_ZN8PfxEntryC2EP8AffixMgr: 90| 8| : pmyMgr(pmgr) 91| 8| , next(nullptr) 92| 8| , nexteq(nullptr) 93| 8| , nextne(nullptr) 94| 8| , flgnxt(nullptr) {} _ZN8PfxEntry3addEPKcm: 97| 1|std::string PfxEntry::add(const char* word, size_t len) { 98| 1| std::string result; 99| 1| if ((len > strip.size() || (len == 0 && pmyMgr->get_fullstrip())) && ------------------ | Branch (99:7): [True: 1, False: 0] | Branch (99:8): [True: 1, False: 0] | Branch (99:31): [True: 0, False: 0] | Branch (99:43): [True: 0, False: 0] ------------------ 100| 1| (len >= numconds) && test_condition(word) && ------------------ | Branch (100:7): [True: 1, False: 0] | Branch (100:28): [True: 1, False: 0] ------------------ 101| 1| (strip.empty() || ------------------ | Branch (101:8): [True: 1, False: 0] ------------------ 102| 1| (len >= strip.size() && strncmp(word, strip.c_str(), strip.size()) == 0))) { ------------------ | Branch (102:8): [True: 0, False: 0] | Branch (102:31): [True: 0, False: 0] ------------------ 103| | /* we have a match so add prefix */ 104| 1| result.assign(appnd); 105| 1| result.append(word + strip.size()); 106| 1| } 107| 1| return result; 108| 1|} _ZN8PfxEntry9checkwordERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEiictR12AffixScratch: 213| 1.31M| AffixScratch& scratch) { 214| 1.31M| struct hentry* he; // hash entry of root word or NULL 215| | 216| | // on entry prefix is 0 length or already matches the beginning of the word. 217| | // So if the remaining root word has positive length 218| | // and if there are enough chars in root word and added back strip chars 219| | // to meet the number of characters conditions, then test it 220| | 221| 1.31M| int tmpl = len - appnd.size(); // length of tmpword 222| | 223| 1.31M| if (tmpl > 0 || (tmpl == 0 && pmyMgr->get_fullstrip())) { ------------------ | Branch (223:7): [True: 1.31M, False: 3.20k] | Branch (223:20): [True: 3.20k, False: 0] | Branch (223:33): [True: 0, False: 3.20k] ------------------ 224| | // generate new root word by removing prefix and adding 225| | // back any characters that would have been stripped 226| | 227| 1.31M| std::string& tmpword = scratch.pfx_check_word; 228| 1.31M| tmpword.assign(strip); 229| 1.31M| tmpword.append(word, start + appnd.size(), tmpl); 230| | 231| | // now make sure all of the conditions on characters 232| | // are met. Please see the appendix at the end of 233| | // this file for more info on exactly what is being 234| | // tested 235| | 236| | // if all conditions are met then check if resulting 237| | // root word in the dictionary 238| | 239| 1.31M| if (test_condition(tmpword)) { ------------------ | Branch (239:9): [True: 16.6k, False: 1.29M] ------------------ 240| 16.6k| tmpl += strip.size(); 241| 16.6k| if ((he = pmyMgr->lookup(tmpword.c_str(), tmpword.size())) != nullptr) { ------------------ | Branch (241:11): [True: 2, False: 16.6k] ------------------ 242| 2| do { 243| 2| if (TESTAFF(he->astr, aflag, he->alen) && ------------------ | | 99| 4|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 2] | | ------------------ ------------------ | Branch (243:15): [True: 0, False: 2] ------------------ 244| | // forbid single prefixes with needaffix flag 245| 0| !TESTAFF(contclass, pmyMgr->get_needaffix(), contclasslen) && ------------------ | | 99| 2|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (245:15): [True: 0, False: 0] ------------------ 246| | // needflag 247| 0| ((!needflag) || TESTAFF(he->astr, needflag, he->alen) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (247:16): [True: 0, False: 0] ------------------ 248| 0| (contclass && TESTAFF(contclass, needflag, contclasslen)))) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (248:17): [True: 0, False: 0] ------------------ 249| 0| return he; 250| 2| he = he->next_homonym; // check homonyms 251| 2| } while (he); ------------------ | Branch (251:18): [True: 0, False: 2] ------------------ 252| 2| } 253| | 254| | // prefix matched but no root word was found 255| | // if aeXPRODUCT is allowed, try again but now 256| | // ross checked combined with a suffix 257| | 258| | // if ((opts & aeXPRODUCT) && in_compound) { 259| 16.6k| if ((opts & aeXPRODUCT)) { ------------------ | | 65| 16.6k|#define aeXPRODUCT (1 << 0) ------------------ | Branch (259:11): [True: 16.6k, False: 0] ------------------ 260| 16.6k| he = pmyMgr->suffix_check(tmpword, 0, tmpl, aeXPRODUCT, this, ------------------ | | 65| 16.6k|#define aeXPRODUCT (1 << 0) ------------------ 261| 16.6k| scratch, FLAG_NULL, needflag, in_compound); ------------------ | | 96| 16.6k|#define FLAG_NULL 0x00 ------------------ 262| 16.6k| if (he) ------------------ | Branch (262:13): [True: 0, False: 16.6k] ------------------ 263| 0| return he; 264| 16.6k| } 265| 16.6k| } 266| 1.31M| } 267| 1.31M| return nullptr; 268| 1.31M|} _ZN8PfxEntry12check_twosfxERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEiictR12AffixScratch: 276| 1.31M| AffixScratch& scratch) { 277| | // on entry prefix is 0 length or already matches the beginning of the word. 278| | // So if the remaining root word has positive length 279| | // and if there are enough chars in root word and added back strip chars 280| | // to meet the number of characters conditions, then test it 281| | 282| 1.31M| int tmpl = len - appnd.size(); // length of tmpword 283| | 284| 1.31M| if ((tmpl > 0 || (tmpl == 0 && pmyMgr->get_fullstrip())) && ------------------ | Branch (284:8): [True: 1.30M, False: 3.16k] | Branch (284:21): [True: 3.16k, False: 0] | Branch (284:34): [True: 0, False: 3.16k] ------------------ 285| 1.30M| (tmpl + strip.size() >= numconds)) { ------------------ | Branch (285:7): [True: 1.26M, False: 45.6k] ------------------ 286| | // generate new root word by removing prefix and adding 287| | // back any characters that would have been stripped 288| | 289| 1.26M| std::string& tmpword = scratch.pfx_check_twosfx; 290| 1.26M| tmpword.assign(strip); 291| 1.26M| tmpword.append(word, start + appnd.size(), tmpl); 292| | 293| | // now make sure all of the conditions on characters 294| | // are met. Please see the appendix at the end of 295| | // this file for more info on exactly what is being 296| | // tested 297| | 298| | // if all conditions are met then check if resulting 299| | // root word in the dictionary 300| | 301| 1.26M| if (test_condition(tmpword)) { ------------------ | Branch (301:9): [True: 12.3k, False: 1.24M] ------------------ 302| 12.3k| tmpl += strip.size(); 303| | 304| | // prefix matched but no root word was found 305| | // if aeXPRODUCT is allowed, try again but now 306| | // cross checked combined with a suffix 307| | 308| 12.3k| if ((opts & aeXPRODUCT) && (in_compound != IN_CPD_BEGIN)) { ------------------ | | 65| 12.3k|#define aeXPRODUCT (1 << 0) ------------------ if ((opts & aeXPRODUCT) && (in_compound != IN_CPD_BEGIN)) { ------------------ | | 73| 12.3k|#define IN_CPD_BEGIN 1 ------------------ | Branch (308:11): [True: 12.3k, False: 0] | Branch (308:34): [True: 12.3k, False: 0] ------------------ 309| | // hash entry of root word or NULL 310| 12.3k| struct hentry* he = pmyMgr->suffix_check_twosfx(tmpword, 0, tmpl, aeXPRODUCT, this, ------------------ | | 65| 12.3k|#define aeXPRODUCT (1 << 0) ------------------ 311| 12.3k| scratch, needflag); 312| 12.3k| if (he) ------------------ | Branch (312:13): [True: 0, False: 12.3k] ------------------ 313| 0| return he; 314| 12.3k| } 315| 12.3k| } 316| 1.26M| } 317| 1.31M| return nullptr; 318| 1.31M|} _ZN8PfxEntry18check_twosfx_morphERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEiictR12AffixScratch: 326| 674| AffixScratch& scratch) { 327| 674| std::string result; 328| | // on entry prefix is 0 length or already matches the beginning of the word. 329| | // So if the remaining root word has positive length 330| | // and if there are enough chars in root word and added back strip chars 331| | // to meet the number of characters conditions, then test it 332| 674| int tmpl = len - appnd.size(); // length of tmpword 333| | 334| 674| if ((tmpl > 0 || (tmpl == 0 && pmyMgr->get_fullstrip())) && ------------------ | Branch (334:8): [True: 674, False: 0] | Branch (334:21): [True: 0, False: 0] | Branch (334:34): [True: 0, False: 0] ------------------ 335| 674| (tmpl + strip.size() >= numconds)) { ------------------ | Branch (335:7): [True: 543, False: 131] ------------------ 336| | // generate new root word by removing prefix and adding 337| | // back any characters that would have been stripped 338| | 339| 543| std::string& tmpword = scratch.pfx_check_twosfx; 340| 543| tmpword.assign(strip); 341| 543| tmpword.append(word, start + appnd.size(), tmpl); 342| | 343| | // now make sure all of the conditions on characters 344| | // are met. Please see the appendix at the end of 345| | // this file for more info on exactly what is being 346| | // tested 347| | 348| | // if all conditions are met then check if resulting 349| | // root word in the dictionary 350| | 351| 543| if (test_condition(tmpword)) { ------------------ | Branch (351:9): [True: 0, False: 543] ------------------ 352| 0| tmpl += strip.size(); 353| | 354| | // prefix matched but no root word was found 355| | // if aeXPRODUCT is allowed, try again but now 356| | // ross checked combined with a suffix 357| | 358| 0| if ((opts & aeXPRODUCT) && (in_compound != IN_CPD_BEGIN)) { ------------------ | | 65| 0|#define aeXPRODUCT (1 << 0) ------------------ if ((opts & aeXPRODUCT) && (in_compound != IN_CPD_BEGIN)) { ------------------ | | 73| 0|#define IN_CPD_BEGIN 1 ------------------ | Branch (358:11): [True: 0, False: 0] | Branch (358:34): [True: 0, False: 0] ------------------ 359| 0| result = pmyMgr->suffix_check_twosfx_morph(tmpword, 0, tmpl, 360| 0| aeXPRODUCT, ------------------ | | 65| 0|#define aeXPRODUCT (1 << 0) ------------------ 361| 0| this, scratch, needflag); 362| 0| } 363| 0| } 364| 543| } 365| 674| return result; 366| 674|} _ZN8PfxEntry11check_morphERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEiictR12AffixScratch: 374| 674| AffixScratch& scratch) { 375| 674| std::string result; 376| | 377| | // on entry prefix is 0 length or already matches the beginning of the word. 378| | // So if the remaining root word has positive length 379| | // and if there are enough chars in root word and added back strip chars 380| | // to meet the number of characters conditions, then test it 381| | 382| 674| int tmpl = len - appnd.size(); // length of tmpword 383| | 384| 674| if ((tmpl > 0 || (tmpl == 0 && pmyMgr->get_fullstrip())) && ------------------ | Branch (384:8): [True: 674, False: 0] | Branch (384:21): [True: 0, False: 0] | Branch (384:34): [True: 0, False: 0] ------------------ 385| 674| (tmpl + strip.size() >= numconds)) { ------------------ | Branch (385:7): [True: 543, False: 131] ------------------ 386| | // generate new root word by removing prefix and adding 387| | // back any characters that would have been stripped 388| | 389| 543| std::string& tmpword = scratch.pfx_check_word; 390| 543| tmpword.assign(strip); 391| 543| tmpword.append(word, start + appnd.size(), tmpl); 392| | 393| | // now make sure all of the conditions on characters 394| | // are met. Please see the appendix at the end of 395| | // this file for more info on exactly what is being 396| | // tested 397| | 398| | // if all conditions are met then check if resulting 399| | // root word in the dictionary 400| | 401| 543| if (test_condition(tmpword)) { ------------------ | Branch (401:9): [True: 0, False: 543] ------------------ 402| 0| tmpl += strip.size(); 403| 0| struct hentry* he; // hash entry of root word or NULL 404| 0| if ((he = pmyMgr->lookup(tmpword.c_str(), tmpword.size())) != nullptr) { ------------------ | Branch (404:11): [True: 0, False: 0] ------------------ 405| 0| do { 406| 0| if (TESTAFF(he->astr, aflag, he->alen) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (406:15): [True: 0, False: 0] ------------------ 407| | // forbid single prefixes with needaffix flag 408| 0| !TESTAFF(contclass, pmyMgr->get_needaffix(), contclasslen) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (408:15): [True: 0, False: 0] ------------------ 409| | // needflag 410| 0| ((!needflag) || TESTAFF(he->astr, needflag, he->alen) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (410:16): [True: 0, False: 0] ------------------ 411| 0| (contclass && TESTAFF(contclass, needflag, contclasslen)))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (411:17): [True: 0, False: 0] ------------------ 412| 0| if (morphcode) { ------------------ | Branch (412:17): [True: 0, False: 0] ------------------ 413| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 414| 0| result.append(morphcode); 415| 0| } else 416| 0| result.append(getKey()); 417| 0| if (!HENTRY_FIND(he, MORPH_STEM)) { ------------------ | | 103| 0|#define MORPH_STEM "st:" ------------------ | Branch (417:17): [True: 0, False: 0] ------------------ 418| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 419| 0| result.append(MORPH_STEM); ------------------ | | 103| 0|#define MORPH_STEM "st:" ------------------ 420| 0| result.append(HENTRY_WORD(he)); ------------------ | | 54| 0|#define HENTRY_WORD(h) &(h->word[0]) ------------------ 421| 0| } 422| | // store the pointer of the hash entry 423| 0| if (HENTRY_DATA(he)) { ------------------ | Branch (423:17): [True: 0, False: 0] ------------------ 424| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 425| 0| result.append(HENTRY_DATA2(he)); 426| 0| } else { 427| | // return with debug information 428| 0| std::string flag = pmyMgr->encode_flag(getFlag()); 429| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 430| 0| result.append(MORPH_FLAG); ------------------ | | 117| 0|#define MORPH_FLAG "fl:" ------------------ 431| 0| result.append(flag); 432| 0| } 433| 0| result.push_back(MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 434| 0| } 435| 0| he = he->next_homonym; 436| 0| } while (he); ------------------ | Branch (436:18): [True: 0, False: 0] ------------------ 437| 0| } 438| | 439| | // prefix matched but no root word was found 440| | // if aeXPRODUCT is allowed, try again but now 441| | // ross checked combined with a suffix 442| | 443| 0| if ((opts & aeXPRODUCT) && (in_compound != IN_CPD_BEGIN)) { ------------------ | | 65| 0|#define aeXPRODUCT (1 << 0) ------------------ if ((opts & aeXPRODUCT) && (in_compound != IN_CPD_BEGIN)) { ------------------ | | 73| 0|#define IN_CPD_BEGIN 1 ------------------ | Branch (443:11): [True: 0, False: 0] | Branch (443:34): [True: 0, False: 0] ------------------ 444| 0| std::string st = pmyMgr->suffix_check_morph(tmpword, 0, tmpl, aeXPRODUCT, this, ------------------ | | 65| 0|#define aeXPRODUCT (1 << 0) ------------------ 445| 0| scratch, FLAG_NULL, needflag); ------------------ | | 96| 0|#define FLAG_NULL 0x00 ------------------ 446| 0| if (!st.empty()) { ------------------ | Branch (446:13): [True: 0, False: 0] ------------------ 447| 0| result.append(st); 448| 0| } 449| 0| } 450| 0| } 451| 543| } 452| | 453| 674| return result; 454| 674|} _ZN8SfxEntryC2EP8AffixMgr: 457| 34| : pmyMgr(pmgr) // register affix manager 458| 34| , next(nullptr) 459| 34| , nexteq(nullptr) 460| 34| , nextne(nullptr) 461| 34| , flgnxt(nullptr) 462| 34| , l_morph(nullptr) 463| 34| , r_morph(nullptr) 464| 34| , eq_morph(nullptr) {} _ZN8SfxEntry3addEPKcm: 467| 53|std::string SfxEntry::add(const char* word, size_t len) { 468| 53| std::string result; 469| | /* make sure all conditions match */ 470| 53| if ((len > strip.size() || (len == 0 && pmyMgr->get_fullstrip())) && ------------------ | Branch (470:8): [True: 53, False: 0] | Branch (470:31): [True: 0, False: 0] | Branch (470:43): [True: 0, False: 0] ------------------ 471| 53| (len >= numconds) && test_condition(word + len, word) && ------------------ | Branch (471:7): [True: 53, False: 0] | Branch (471:28): [True: 53, False: 0] ------------------ 472| 53| (strip.empty() || ------------------ | Branch (472:8): [True: 53, False: 0] ------------------ 473| 53| (len >= strip.size() && strcmp(word + len - strip.size(), strip.c_str()) == 0))) { ------------------ | Branch (473:9): [True: 0, False: 0] | Branch (473:32): [True: 0, False: 0] ------------------ 474| 53| result.assign(word, len); 475| | /* we have a match so add suffix */ 476| 53| result.replace(len - strip.size(), std::string::npos, appnd); 477| 53| } 478| 53| return result; 479| 53|} _ZN8SfxEntry9checkwordERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEiiiP8PfxEntrytttR12AffixScratch: 627| 65.1k| AffixScratch& scratch) { 628| 65.1k| struct hentry* he; // hash entry pointer 629| 65.1k| PfxEntry* ep = ppfx; 630| | 631| | // if this suffix is being cross checked with a prefix 632| | // but it does not support cross products skip it 633| | 634| 65.1k| if (((optflags & aeXPRODUCT) != 0) && ((opts & aeXPRODUCT) == 0)) ------------------ | | 65| 65.1k|#define aeXPRODUCT (1 << 0) ------------------ if (((optflags & aeXPRODUCT) != 0) && ((opts & aeXPRODUCT) == 0)) ------------------ | | 65| 0|#define aeXPRODUCT (1 << 0) ------------------ | Branch (634:7): [True: 0, False: 65.1k] | Branch (634:41): [True: 0, False: 0] ------------------ 635| 0| return nullptr; 636| | 637| | // upon entry suffix is 0 length or already matches the end of the word. 638| | // So if the remaining root word has positive length 639| | // and if there are enough chars in root word and added back strip chars 640| | // to meet the number of characters conditions, then test it 641| | 642| 65.1k| int tmpl = len - appnd.size(); // length of tmpword 643| | // the second condition is not enough for UTF-8 strings 644| | // it checked in test_condition() 645| | 646| 65.1k| if ((tmpl > 0 || (tmpl == 0 && pmyMgr->get_fullstrip())) && ------------------ | Branch (646:8): [True: 63.7k, False: 1.40k] | Branch (646:21): [True: 1.40k, False: 0] | Branch (646:34): [True: 0, False: 1.40k] ------------------ 647| 63.7k| (tmpl + strip.size() >= numconds)) { ------------------ | Branch (647:7): [True: 63.7k, False: 0] ------------------ 648| | // generate new root word by removing suffix and adding 649| | // back any characters that would have been stripped or 650| | // or null terminating the shorter string 651| | 652| 63.7k| std::string& tmpword = scratch.sfx_check_word; 653| 63.7k| tmpword.assign(word, start, tmpl); 654| 63.7k| if (!strip.empty()) { ------------------ | Branch (654:9): [True: 0, False: 63.7k] ------------------ 655| 0| tmpword.append(strip); 656| 0| } 657| | 658| 63.7k| const char* beg = tmpword.c_str(); 659| 63.7k| const char* end = beg + tmpword.size(); 660| | 661| | // now make sure all of the conditions on characters 662| | // are met. Please see the appendix at the end of 663| | // this file for more info on exactly what is being 664| | // tested 665| | 666| | // if all conditions are met then check if resulting 667| | // root word in the dictionary 668| | 669| 63.7k| if (test_condition(end, beg)) { ------------------ | Branch (669:9): [True: 63.7k, False: 4] ------------------ 670| |#ifdef SZOSZABLYA_POSSIBLE_ROOTS 671| | fprintf(stdout, "%s %s %c\n", word.c_str() + start, beg, aflag); 672| |#endif 673| 63.7k| if ((he = pmyMgr->lookup(tmpword.c_str(), tmpword.size())) != nullptr) { ------------------ | Branch (673:11): [True: 26, False: 63.7k] ------------------ 674| 26| do { 675| | // check conditional suffix (enabled by prefix) 676| 26| if ((TESTAFF(he->astr, aflag, he->alen) || ------------------ | | 99| 52|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 2, False: 24] | | ------------------ ------------------ | Branch (676:15): [True: 2, False: 24] ------------------ 677| 24| (ep && ep->getCont() && ------------------ | Branch (677:17): [True: 0, False: 24] | Branch (677:23): [True: 0, False: 0] ------------------ 678| 0| TESTAFF(ep->getCont(), aflag, ep->getContLen()))) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 679| 2| (((optflags & aeXPRODUCT) == 0) || ------------------ | | 65| 2|#define aeXPRODUCT (1 << 0) ------------------ | Branch (679:16): [True: 2, False: 0] ------------------ 680| 0| (ep && TESTAFF(he->astr, ep->getFlag(), he->alen)) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (680:17): [True: 0, False: 0] ------------------ 681| | // enabled by prefix 682| 0| ((contclass) && ------------------ | Branch (682:17): [True: 0, False: 0] ------------------ 683| 0| (ep && TESTAFF(contclass, ep->getFlag(), contclasslen)))) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (683:18): [True: 0, False: 0] ------------------ 684| | // handle cont. class 685| 2| ((!cclass) || ------------------ | Branch (685:16): [True: 2, False: 0] ------------------ 686| 0| ((contclass) && TESTAFF(contclass, cclass, contclasslen))) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (686:17): [True: 0, False: 0] ------------------ 687| | // check only in compound homonyms (bad flags) 688| 2| (!badflag || !TESTAFF(he->astr, badflag, he->alen)) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (688:16): [True: 2, False: 0] | Branch (688:28): [True: 0, False: 0] ------------------ 689| | // handle required flag 690| 2| ((!needflag) || ------------------ | Branch (690:16): [True: 2, False: 0] ------------------ 691| 0| (TESTAFF(he->astr, needflag, he->alen) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 692| 0| ((contclass) && TESTAFF(contclass, needflag, contclasslen))))) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (692:18): [True: 0, False: 0] ------------------ 693| 2| return he; 694| 24| he = he->next_homonym; // check homonyms 695| 24| } while (he); ------------------ | Branch (695:18): [True: 0, False: 24] ------------------ 696| 26| } 697| 63.7k| } 698| 63.7k| } 699| 65.1k| return nullptr; 700| 65.1k|} _ZN8SfxEntry12check_twosfxERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEiiiP8PfxEntrytR12AffixScratch: 709| 25.7k| AffixScratch& scratch) { 710| 25.7k| PfxEntry* ep = ppfx; 711| | 712| | // if this suffix is being cross checked with a prefix 713| | // but it does not support cross products skip it 714| | 715| 25.7k| if ((optflags & aeXPRODUCT) != 0 && (opts & aeXPRODUCT) == 0) ------------------ | | 65| 25.7k|#define aeXPRODUCT (1 << 0) ------------------ if ((optflags & aeXPRODUCT) != 0 && (opts & aeXPRODUCT) == 0) ------------------ | | 65| 0|#define aeXPRODUCT (1 << 0) ------------------ | Branch (715:7): [True: 0, False: 25.7k] | Branch (715:39): [True: 0, False: 0] ------------------ 716| 0| return nullptr; 717| | 718| | // upon entry suffix is 0 length or already matches the end of the word. 719| | // So if the remaining root word has positive length 720| | // and if there are enough chars in root word and added back strip chars 721| | // to meet the number of characters conditions, then test it 722| | 723| 25.7k| int tmpl = len - appnd.size(); // length of tmpword 724| | 725| 25.7k| if ((tmpl > 0 || (tmpl == 0 && pmyMgr->get_fullstrip())) && ------------------ | Branch (725:8): [True: 25.7k, False: 41] | Branch (725:21): [True: 41, False: 0] | Branch (725:34): [True: 0, False: 41] ------------------ 726| 25.7k| (tmpl + strip.size() >= numconds)) { ------------------ | Branch (726:7): [True: 25.7k, False: 0] ------------------ 727| | // generate new root word by removing suffix and adding 728| | // back any characters that would have been stripped or 729| | // or null terminating the shorter string 730| | 731| 25.7k| std::string& tmpword = scratch.sfx_check_twosfx; 732| 25.7k| tmpword.assign(word, start); 733| 25.7k| tmpword.resize(tmpl); 734| 25.7k| tmpword.append(strip); 735| 25.7k| tmpl += strip.size(); 736| | 737| 25.7k| const char* beg = tmpword.c_str(); 738| 25.7k| const char* end = beg + tmpl; 739| | 740| | // now make sure all of the conditions on characters 741| | // are met. Please see the appendix at the end of 742| | // this file for more info on exactly what is being 743| | // tested 744| | 745| | // if all conditions are met then recall suffix_check 746| | 747| 25.7k| if (test_condition(end, beg)) { ------------------ | Branch (747:9): [True: 25.7k, False: 0] ------------------ 748| 25.7k| struct hentry* he; // hash entry pointer 749| 25.7k| if (ppfx) { ------------------ | Branch (749:11): [True: 0, False: 25.7k] ------------------ 750| | // handle conditional suffix 751| 0| if ((contclass) && TESTAFF(contclass, ep->getFlag(), contclasslen)) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (751:13): [True: 0, False: 0] | Branch (751:13): [True: 0, False: 0] ------------------ 752| 0| he = pmyMgr->suffix_check(tmpword, 0, tmpl, 0, nullptr, scratch, (FLAG)aflag, needflag, IN_CPD_NOT); ------------------ | | 72| 0|#define IN_CPD_NOT 0 ------------------ 753| 0| else 754| 0| he = pmyMgr->suffix_check(tmpword, 0, tmpl, optflags, ppfx, 755| 0| scratch, (FLAG)aflag, needflag, IN_CPD_NOT); ------------------ | | 72| 0|#define IN_CPD_NOT 0 ------------------ 756| 25.7k| } else { 757| 25.7k| he = pmyMgr->suffix_check(tmpword, 0, tmpl, 0, nullptr, scratch, (FLAG)aflag, needflag, IN_CPD_NOT); ------------------ | | 72| 25.7k|#define IN_CPD_NOT 0 ------------------ 758| 25.7k| } 759| 25.7k| if (he) ------------------ | Branch (759:11): [True: 0, False: 25.7k] ------------------ 760| 0| return he; 761| 25.7k| } 762| 25.7k| } 763| 25.7k| return nullptr; 764| 25.7k|} _ZN8SfxEntry15initReverseWordEv: 875| 34|void SfxEntry::initReverseWord() { 876| 34| rappnd = appnd; 877| 34| reverseword(rappnd); 878| 34|} _ZN8PfxEntry14test_conditionERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEE: 125| 2.57M|inline int PfxEntry::test_condition(const std::string& s) { 126| 2.57M| size_t st = 0; 127| 2.57M| size_t pos = std::string::npos; // group with pos input position 128| 2.57M| bool neg = false; // complementer 129| 2.57M| bool ingroup = false; // character in the group 130| 2.57M| if (numconds == 0) ------------------ | Branch (130:7): [True: 4.35k, False: 2.57M] ------------------ 131| 4.35k| return 1; 132| 2.57M| char* p = c.conds; 133| 2.69M| while (true) { ------------------ | Branch (133:10): [True: 2.69M, Folded] ------------------ 134| 2.69M| switch (*p) { 135| 0| case '\0': ------------------ | Branch (135:7): [True: 0, False: 2.69M] ------------------ 136| 0| return 1; 137| 24.6k| case '[': { ------------------ | Branch (137:7): [True: 24.6k, False: 2.66M] ------------------ 138| 24.6k| neg = false; 139| 24.6k| ingroup = false; 140| 24.6k| p = nextchar(p); 141| 24.6k| pos = st; 142| 24.6k| break; 143| 0| } 144| 24.6k| case '^': { ------------------ | Branch (144:7): [True: 24.6k, False: 2.66M] ------------------ 145| 24.6k| p = nextchar(p); 146| 24.6k| neg = true; 147| 24.6k| break; 148| 0| } 149| 24.6k| case ']': { ------------------ | Branch (149:7): [True: 24.6k, False: 2.66M] ------------------ 150| 24.6k| if (bool(neg) == bool(ingroup)) ------------------ | Branch (150:13): [True: 0, False: 24.6k] ------------------ 151| 0| return 0; 152| 24.6k| pos = std::string::npos; 153| 24.6k| p = nextchar(p); 154| | // skip the next character 155| 24.6k| if (!ingroup && st < s.size()) ------------------ | Branch (155:13): [True: 24.6k, False: 0] | Branch (155:25): [True: 23.3k, False: 1.28k] ------------------ 156| 23.3k| st = (opts & aeUTF8) ? utf8_next(s, st) : st + 1; ------------------ | | 66| 23.3k|#define aeUTF8 (1 << 1) ------------------ | Branch (156:16): [True: 23.3k, False: 0] ------------------ 157| 24.6k| if (st == s.size() && p) ------------------ | Branch (157:13): [True: 3.51k, False: 21.0k] | Branch (157:31): [True: 0, False: 3.51k] ------------------ 158| 0| return 0; // word <= condition 159| 24.6k| break; 160| 24.6k| } 161| 24.6k| case '.': ------------------ | Branch (161:7): [True: 0, False: 2.69M] ------------------ 162| 0| if (pos == std::string::npos) { // dots are not metacharacters in groups: [.] ------------------ | Branch (162:13): [True: 0, False: 0] ------------------ 163| 0| if (st == s.size()) ------------------ | Branch (163:15): [True: 0, False: 0] ------------------ 164| 0| return 0; // dot has no character to match 165| 0| p = nextchar(p); 166| | // skip the next character 167| 0| st = (opts & aeUTF8) ? utf8_next(s, st) : st + 1; ------------------ | | 66| 0|#define aeUTF8 (1 << 1) ------------------ | Branch (167:16): [True: 0, False: 0] ------------------ 168| 0| break; 169| 0| } 170| | /* FALLTHROUGH */ 171| 2.61M| default: { ------------------ | Branch (171:7): [True: 2.61M, False: 73.8k] ------------------ 172| 2.61M| if (st < s.size() && s[st] == *p) { ------------------ | Branch (172:13): [True: 2.61M, False: 2.56k] | Branch (172:30): [True: 24.7k, False: 2.59M] ------------------ 173| 24.7k| ++st; 174| 24.7k| p = nextchar(p); 175| 24.7k| if ((opts & aeUTF8) && (s[st - 1] & 0x80)) { // multibyte ------------------ | | 66| 24.7k|#define aeUTF8 (1 << 1) ------------------ | Branch (175:15): [True: 24.7k, False: 0] | Branch (175:34): [True: 24.7k, False: 0] ------------------ 176| 49.3k| while (p && is_utf8_cont(*p)) { // character ------------------ | Branch (176:20): [True: 49.3k, False: 0] | Branch (176:25): [True: 24.7k, False: 24.6k] ------------------ 177| 24.7k| if (st >= s.size() || *p != s[st]) { ------------------ | Branch (177:19): [True: 0, False: 24.7k] | Branch (177:37): [True: 128, False: 24.6k] ------------------ 178| 128| if (pos == std::string::npos) ------------------ | Branch (178:21): [True: 112, False: 16] ------------------ 179| 112| return 0; 180| 16| st = pos; 181| 16| break; 182| 128| } 183| 24.6k| p = nextchar(p); 184| 24.6k| ++st; 185| 24.6k| } 186| 24.6k| if (pos != std::string::npos && st != pos) { ------------------ | Branch (186:17): [True: 16, False: 24.6k] | Branch (186:45): [True: 0, False: 16] ------------------ 187| 0| ingroup = true; 188| 0| while (p && *p != ']' && ((p = nextchar(p)) != nullptr)) { ------------------ | Branch (188:22): [True: 0, False: 0] | Branch (188:27): [True: 0, False: 0] | Branch (188:40): [True: 0, False: 0] ------------------ 189| 0| } 190| 0| } 191| 24.6k| } else if (pos != std::string::npos) { ------------------ | Branch (191:22): [True: 0, False: 0] ------------------ 192| 0| ingroup = true; 193| 0| while (p && *p != ']' && ((p = nextchar(p)) != nullptr)) { ------------------ | Branch (193:20): [True: 0, False: 0] | Branch (193:25): [True: 0, False: 0] | Branch (193:38): [True: 0, False: 0] ------------------ 194| 0| } 195| 0| } 196| 2.59M| } else if (pos != std::string::npos) { // group ------------------ | Branch (196:20): [True: 49.1k, False: 2.54M] ------------------ 197| 49.1k| p = nextchar(p); 198| 49.1k| } else 199| 2.54M| return 0; 200| 2.61M| } 201| 2.69M| } 202| 147k| if (!p) ------------------ | Branch (202:9): [True: 24.6k, False: 123k] ------------------ 203| 24.6k| return 1; 204| 147k| } 205| 2.57M|} _ZN8PfxEntry8nextcharEPc: 110| 172k|inline char* PfxEntry::nextchar(char* p) { 111| 172k| if (p) { ------------------ | Branch (111:7): [True: 172k, False: 0] ------------------ 112| 172k| p++; 113| 172k| if (opts & aeLONGCOND) { ------------------ | | 69| 172k|#define aeLONGCOND (1 << 4) ------------------ | Branch (113:9): [True: 0, False: 172k] ------------------ 114| | // jump to the 2nd part of the condition 115| 0| if (p == c.conds + MAXCONDLEN_1) ------------------ | | 91| 0|#define MAXCONDLEN_1 (MAXCONDLEN - sizeof(char*)) | | ------------------ | | | | 90| 0|#define MAXCONDLEN 20 | | ------------------ ------------------ | Branch (115:11): [True: 0, False: 0] ------------------ 116| 0| return c.l.conds2; 117| | // end of the MAXCONDLEN length condition 118| 172k| } else if (p == c.conds + MAXCONDLEN) ------------------ | | 90| 172k|#define MAXCONDLEN 20 ------------------ | Branch (118:16): [True: 0, False: 172k] ------------------ 119| 0| return nullptr; 120| 172k| return *p ? p : nullptr; ------------------ | Branch (120:12): [True: 147k, False: 24.6k] ------------------ 121| 172k| } 122| 0| return nullptr; 123| 172k|} _ZN8SfxEntry14test_conditionEPKcS1_: 496| 89.5k|inline int SfxEntry::test_condition(const char* st, const char* beg) { 497| 89.5k| const char* pos = nullptr; // group with pos input position 498| 89.5k| bool neg = false; // complementer 499| 89.5k| bool ingroup = false; // character in the group 500| 89.5k| if (numconds == 0) ------------------ | Branch (500:7): [True: 89.5k, False: 4] ------------------ 501| 89.5k| return 1; 502| 4| char* p = c.conds; 503| 4| st--; 504| 4| int i = 1; 505| 32| while (true) { ------------------ | Branch (505:10): [True: 32, Folded] ------------------ 506| 32| switch (*p) { 507| 0| case '\0': ------------------ | Branch (507:7): [True: 0, False: 32] ------------------ 508| 0| return 1; 509| 4| case '[': ------------------ | Branch (509:7): [True: 4, False: 28] ------------------ 510| 4| p = nextchar(p); 511| 4| pos = st; 512| 4| break; 513| 0| case '^': ------------------ | Branch (513:7): [True: 0, False: 32] ------------------ 514| 0| p = nextchar(p); 515| 0| neg = true; 516| 0| break; 517| 4| case ']': ------------------ | Branch (517:7): [True: 4, False: 28] ------------------ 518| 4| if (!neg && !ingroup) ------------------ | Branch (518:13): [True: 4, False: 0] | Branch (518:21): [True: 4, False: 0] ------------------ 519| 4| return 0; 520| 0| i++; 521| | // skip the next character 522| 0| if (!ingroup) { ------------------ | Branch (522:13): [True: 0, False: 0] ------------------ 523| 0| for (; (opts & aeUTF8) && (st >= beg) && is_utf8_cont(*st); st--) ------------------ | | 66| 0|#define aeUTF8 (1 << 1) ------------------ | Branch (523:18): [True: 0, False: 0] | Branch (523:37): [True: 0, False: 0] | Branch (523:52): [True: 0, False: 0] ------------------ 524| 0| ; 525| 0| st--; 526| 0| } 527| 0| pos = nullptr; 528| 0| neg = false; 529| 0| ingroup = false; 530| 0| p = nextchar(p); 531| 0| if (st < beg && p) ------------------ | Branch (531:13): [True: 0, False: 0] | Branch (531:25): [True: 0, False: 0] ------------------ 532| 0| return 0; // word <= condition 533| 0| break; 534| 0| case '.': ------------------ | Branch (534:7): [True: 0, False: 32] ------------------ 535| 0| if (!pos) { ------------------ | Branch (535:13): [True: 0, False: 0] ------------------ 536| | // dots are not metacharacters in groups: [.] 537| 0| p = nextchar(p); 538| | // skip the next character 539| 0| for (st--; (opts & aeUTF8) && (st >= beg) && is_utf8_cont(*st); ------------------ | | 66| 0|#define aeUTF8 (1 << 1) ------------------ | Branch (539:22): [True: 0, False: 0] | Branch (539:41): [True: 0, False: 0] | Branch (539:56): [True: 0, False: 0] ------------------ 540| 0| st--) 541| 0| ; 542| 0| if (st < beg) { // word <= condition ------------------ | Branch (542:15): [True: 0, False: 0] ------------------ 543| 0| if (p) ------------------ | Branch (543:17): [True: 0, False: 0] ------------------ 544| 0| return 0; 545| 0| else 546| 0| return 1; 547| 0| } 548| 0| if ((opts & aeUTF8) && (*st & 0x80)) { // head of the UTF-8 character ------------------ | | 66| 0|#define aeUTF8 (1 << 1) ------------------ | Branch (548:15): [True: 0, False: 0] | Branch (548:34): [True: 0, False: 0] ------------------ 549| 0| st--; 550| 0| if (st < beg) { // word <= condition ------------------ | Branch (550:17): [True: 0, False: 0] ------------------ 551| 0| if (p) ------------------ | Branch (551:19): [True: 0, False: 0] ------------------ 552| 0| return 0; 553| 0| else 554| 0| return 1; 555| 0| } 556| 0| } 557| 0| break; 558| 0| } 559| | /* FALLTHROUGH */ 560| 24| default: { ------------------ | Branch (560:7): [True: 24, False: 8] ------------------ 561| 24| if (*st == *p) { ------------------ | Branch (561:13): [True: 0, False: 24] ------------------ 562| 0| p = nextchar(p); 563| 0| if ((opts & aeUTF8) && (*st & 0x80)) { ------------------ | | 66| 0|#define aeUTF8 (1 << 1) ------------------ | Branch (563:15): [True: 0, False: 0] | Branch (563:34): [True: 0, False: 0] ------------------ 564| 0| st--; 565| 0| while (p && (st >= beg)) { ------------------ | Branch (565:20): [True: 0, False: 0] | Branch (565:25): [True: 0, False: 0] ------------------ 566| 0| if (*p != *st) { ------------------ | Branch (566:19): [True: 0, False: 0] ------------------ 567| 0| if (!pos) ------------------ | Branch (567:21): [True: 0, False: 0] ------------------ 568| 0| return 0; 569| 0| st = pos; 570| 0| break; 571| 0| } 572| | // first byte of the UTF-8 multibyte character 573| 0| if (!is_utf8_cont(*p)) ------------------ | Branch (573:19): [True: 0, False: 0] ------------------ 574| 0| break; 575| 0| p = nextchar(p); 576| 0| st--; 577| 0| } 578| 0| if (pos && st != pos) { ------------------ | Branch (578:17): [True: 0, False: 0] | Branch (578:24): [True: 0, False: 0] ------------------ 579| 0| if (neg) ------------------ | Branch (579:19): [True: 0, False: 0] ------------------ 580| 0| return 0; 581| 0| else if (i == numconds) ------------------ | Branch (581:24): [True: 0, False: 0] ------------------ 582| 0| return 1; 583| 0| ingroup = true; 584| 0| while (p && *p != ']' && ((p = nextchar(p)) != nullptr)) { ------------------ | Branch (584:22): [True: 0, False: 0] | Branch (584:27): [True: 0, False: 0] | Branch (584:40): [True: 0, False: 0] ------------------ 585| 0| } 586| 0| st--; 587| 0| } 588| 0| if (p && *p != ']') ------------------ | Branch (588:17): [True: 0, False: 0] | Branch (588:22): [True: 0, False: 0] ------------------ 589| 0| p = nextchar(p); 590| 0| } else if (pos) { ------------------ | Branch (590:22): [True: 0, False: 0] ------------------ 591| 0| if (neg) ------------------ | Branch (591:17): [True: 0, False: 0] ------------------ 592| 0| return 0; 593| 0| else if (i == numconds) ------------------ | Branch (593:22): [True: 0, False: 0] ------------------ 594| 0| return 1; 595| 0| ingroup = true; 596| 0| while (p && *p != ']' && ((p = nextchar(p)) != nullptr)) { ------------------ | Branch (596:20): [True: 0, False: 0] | Branch (596:25): [True: 0, False: 0] | Branch (596:38): [True: 0, False: 0] ------------------ 597| 0| } 598| | // if (p && *p != ']') p = nextchar(p); 599| 0| st--; 600| 0| } 601| 0| if (!pos) { ------------------ | Branch (601:15): [True: 0, False: 0] ------------------ 602| 0| i++; 603| 0| st--; 604| 0| } 605| 0| if (st < beg && p && *p != ']') ------------------ | Branch (605:15): [True: 0, False: 0] | Branch (605:27): [True: 0, False: 0] | Branch (605:32): [True: 0, False: 0] ------------------ 606| 0| return 0; // word <= condition 607| 24| } else if (pos) { // group ------------------ | Branch (607:20): [True: 24, False: 0] ------------------ 608| 24| p = nextchar(p); 609| 24| } else 610| 0| return 0; 611| 24| } 612| 32| } 613| 28| if (!p) ------------------ | Branch (613:9): [True: 0, False: 28] ------------------ 614| 0| return 1; 615| 28| } 616| 4|} _ZN8SfxEntry8nextcharEPc: 481| 28|inline char* SfxEntry::nextchar(char* p) { 482| 28| if (p) { ------------------ | Branch (482:7): [True: 28, False: 0] ------------------ 483| 28| p++; 484| 28| if (opts & aeLONGCOND) { ------------------ | | 69| 28|#define aeLONGCOND (1 << 4) ------------------ | Branch (484:9): [True: 0, False: 28] ------------------ 485| | // jump to the 2nd part of the condition 486| 0| if (p == c.l.conds1 + MAXCONDLEN_1) ------------------ | | 91| 0|#define MAXCONDLEN_1 (MAXCONDLEN - sizeof(char*)) | | ------------------ | | | | 90| 0|#define MAXCONDLEN 20 | | ------------------ ------------------ | Branch (486:11): [True: 0, False: 0] ------------------ 487| 0| return c.l.conds2; 488| | // end of the MAXCONDLEN length condition 489| 28| } else if (p == c.conds + MAXCONDLEN) ------------------ | | 90| 28|#define MAXCONDLEN 20 ------------------ | Branch (489:16): [True: 0, False: 28] ------------------ 490| 0| return nullptr; 491| 28| return *p ? p : nullptr; ------------------ | Branch (491:12): [True: 28, False: 0] ------------------ 492| 28| } 493| 0| return nullptr; 494| 28|} _ZN8PfxEntry7getNextEv: 134| 2.55M| inline PfxEntry* getNext() { return next; } _ZN8SfxEntry7getNextEv: 216| 228| inline SfxEntry* getNext() { return next; } _ZN8PfxEntry6getKeyEv: 124| 116k| const char* getKey() { return appnd.c_str(); } _ZN8PfxEntry7getFlagEv: 123| 848| FLAG getFlag() { return aflag; } _ZN8PfxEntry9setFlgNxtEPS_: 142| 8| inline void setFlgNxt(PfxEntry* ptr) { flgnxt = ptr; } _ZN8PfxEntry7setNextEPS_: 139| 8| inline void setNext(PfxEntry* ptr) { next = ptr; } _ZN8PfxEntry9setNextEQEPS_: 141| 14| inline void setNextEQ(PfxEntry* ptr) { nexteq = ptr; } _ZN8PfxEntry9setNextNEEPS_: 140| 16| inline void setNextNE(PfxEntry* ptr) { nextne = ptr; } _ZN8PfxEntry9getNextEQEv: 136| 76.1k| inline PfxEntry* getNextEQ() { return nexteq; } _ZN8PfxEntry9getNextNEEv: 135| 39.5k| inline PfxEntry* getNextNE() { return nextne; } _ZN8SfxEntry6getKeyEv: 205| 461k| const char* getKey() { return rappnd.c_str(); } _ZN8SfxEntry7getFlagEv: 204| 123k| FLAG getFlag() { return aflag; } _ZN8SfxEntry9setFlgNxtEPS_: 228| 34| inline void setFlgNxt(SfxEntry* ptr) { flgnxt = ptr; } _ZN8SfxEntry7setNextEPS_: 225| 34| inline void setNext(SfxEntry* ptr) { next = ptr; } _ZN8SfxEntry9setNextEQEPS_: 227| 88| inline void setNextEQ(SfxEntry* ptr) { nexteq = ptr; } _ZN8SfxEntry9setNextNEEPS_: 226| 80| inline void setNextNE(SfxEntry* ptr) { nextne = ptr; } _ZN8SfxEntry9getNextEQEv: 218| 407k| inline SfxEntry* getNextEQ() { return nexteq; } _ZN8SfxEntry9getNextNEEv: 217| 53.5k| inline SfxEntry* getNextNE() { return nextne; } _ZN8PfxEntry7getContEv: 131| 3.93M| inline const unsigned short* getCont() { return contclass; } _ZN8PfxEntry10getContLenEv: 132| 1.31M| inline unsigned short getContLen() { return contclasslen; } _ZN8SfxEntry7getContEv: 210| 544k| inline const unsigned short* getCont() { return contclass; } _ZN8SfxEntry10getContLenEv: 211| 64.9k| inline unsigned short getContLen() { return contclasslen; } _ZN8SfxEntry9getKeyLenEv: 214| 168k| inline int getKeyLen() { return appnd.size(); } _ZN8SfxEntry9getFlgNxtEv: 223| 65.5k| inline SfxEntry* getFlgNxt() { return flgnxt; } _ZN8SfxEntry8getAffixEv: 212| 37.1k| inline const char* getAffix() { return appnd.c_str(); } _ZNK8SfxEntry10allowCrossEv: 171| 53| bool allowCross() const { return ((opts & aeXPRODUCT) != 0); } ------------------ | | 65| 53|#define aeXPRODUCT (1 << 0) ------------------ _ZNK8PfxEntry10allowCrossEv: 94| 1| bool allowCross() const { return ((opts & aeXPRODUCT) != 0); } ------------------ | | 65| 1|#define aeXPRODUCT (1 << 0) ------------------ _ZN8PfxEntry9getKeyLenEv: 127| 2.52k| inline int getKeyLen() { return appnd.size(); } _ZN8PfxEntry9getFlgNxtEv: 137| 840| inline PfxEntry* getFlgNxt() { return flgnxt; } _ZN8AffixMgrC2EPKcRKNSt3__16vectorINS2_10unique_ptrI7HashMgrNS2_14default_deleteIS5_EEEENS2_9allocatorIS8_EEEES1_: 93| 38| : alldic(ptr) 94| 38| , pHMgr(ptr[0].get()) { 95| | 96| | // register hash manager and load affix data from aff file 97| 38| csconv = nullptr; 98| 38| utf8 = 0; 99| 38| complexprefixes = 0; 100| 38| parsedmaptable = false; 101| 38| parsedbreaktable = false; 102| 38| iconvtable = nullptr; 103| 38| oconvtable = nullptr; 104| | // allow simplified compound forms (see 3rd field of CHECKCOMPOUNDPATTERN) 105| 38| simplifiedcpd = 0; 106| 38| parsedcheckcpd = false; 107| 38| parseddefcpd = false; 108| 38| phone = nullptr; 109| 38| compoundflag = FLAG_NULL; // permits word in compound forms ------------------ | | 96| 38|#define FLAG_NULL 0x00 ------------------ 110| 38| compoundbegin = FLAG_NULL; // may be first word in compound forms ------------------ | | 96| 38|#define FLAG_NULL 0x00 ------------------ 111| 38| compoundmiddle = FLAG_NULL; // may be middle word in compound forms ------------------ | | 96| 38|#define FLAG_NULL 0x00 ------------------ 112| 38| compoundend = FLAG_NULL; // may be last word in compound forms ------------------ | | 96| 38|#define FLAG_NULL 0x00 ------------------ 113| 38| compoundroot = FLAG_NULL; // compound word signing flag ------------------ | | 96| 38|#define FLAG_NULL 0x00 ------------------ 114| 38| compoundpermitflag = FLAG_NULL; // compound permitting flag for suffixed word ------------------ | | 96| 38|#define FLAG_NULL 0x00 ------------------ 115| 38| compoundforbidflag = FLAG_NULL; // compound fordidden flag for suffixed word ------------------ | | 96| 38|#define FLAG_NULL 0x00 ------------------ 116| 38| compoundmoresuffixes = 0; // allow more suffixes within compound words 117| 38| checkcompounddup = 0; // forbid double words in compounds 118| 38| checkcompoundrep = 0; // forbid bad compounds (may be non-compound word with 119| | // a REP substitution) 120| 38| checkcompoundcase = 121| 38| 0; // forbid upper and lowercase combinations at word bounds 122| 38| checkcompoundtriple = 0; // forbid compounds with triple letters 123| 38| simplifiedtriple = 0; // allow simplified triple letters in compounds 124| | // (Schiff+fahrt -> Schiffahrt) 125| 38| forbiddenword = FORBIDDENWORD; // forbidden word signing flag ------------------ | | 127| 38|#define FORBIDDENWORD 65510 ------------------ 126| 38| nosuggest = FLAG_NULL; // don't suggest words signed with NOSUGGEST flag ------------------ | | 96| 38|#define FLAG_NULL 0x00 ------------------ 127| 38| nongramsuggest = FLAG_NULL; ------------------ | | 96| 38|#define FLAG_NULL 0x00 ------------------ 128| 38| langnum = 0; // language code (see http://l10n.openoffice.org/languages.html) 129| 38| needaffix = FLAG_NULL; // forbidden root, allowed only with suffixes ------------------ | | 96| 38|#define FLAG_NULL 0x00 ------------------ 130| 38| cpdwordmax = -1; // default: unlimited wordcount in compound words 131| 38| cpdmin = -1; // undefined 132| 38| cpdmaxsyllable = 0; // default: unlimited syllablecount in compound words 133| 38| pfxappnd = nullptr; // previous prefix for counting syllables of the prefix BUG 134| 38| sfxappnd = nullptr; // previous suffix for counting syllables of the suffix BUG 135| 38| sfxextra = 0; // modifier for syllable count of sfxappnd BUG 136| 38| checknum = 0; // checking numbers, and word with numbers 137| 38| havecontclass = 0; // flags of possible continuing classes (double affix) 138| | // LEMMA_PRESENT: not put root into the morphological output. Lemma presents 139| | // in morhological description in dictionary file. It's often combined with 140| | // PSEUDOROOT. 141| 38| lemma_present = FLAG_NULL; ------------------ | | 96| 38|#define FLAG_NULL 0x00 ------------------ 142| 38| circumfix = FLAG_NULL; ------------------ | | 96| 38|#define FLAG_NULL 0x00 ------------------ 143| 38| onlyincompound = FLAG_NULL; ------------------ | | 96| 38|#define FLAG_NULL 0x00 ------------------ 144| 38| maxngramsugs = -1; // undefined 145| 38| maxdiff = -1; // undefined 146| 38| onlymaxdiff = 0; 147| 38| maxcpdsugs = -1; // undefined 148| 38| nosplitsugs = 0; 149| 38| sugswithdots = 0; 150| 38| keepcase = 0; 151| 38| forceucase = 0; 152| 38| warn = 0; 153| 38| forbidwarn = 0; 154| 38| checksharps = 0; 155| 38| substandard = FLAG_NULL; ------------------ | | 96| 38|#define FLAG_NULL 0x00 ------------------ 156| 38| fullstrip = 0; 157| | 158| 38| sfx = nullptr; 159| 38| pfx = nullptr; 160| | 161| 9.76k| for (int i = 0; i < SETSIZE; i++) { ------------------ | | 61| 9.76k|#define SETSIZE 256 ------------------ | Branch (161:19): [True: 9.72k, False: 38] ------------------ 162| 9.72k| pStart[i] = nullptr; 163| 9.72k| sStart[i] = nullptr; 164| 9.72k| pFlag[i] = nullptr; 165| 9.72k| sFlag[i] = nullptr; 166| 9.72k| } 167| | 168| 38| memset(contclasses, 0, CONTSIZE * sizeof(char)); ------------------ | | 62| 38|#define CONTSIZE 65536 ------------------ 169| | 170| 38| if (parse_file(affpath, key)) { ------------------ | Branch (170:7): [True: 0, False: 38] ------------------ 171| 0| fprintf(stderr, "Failure loading aff file %s\n", affpath); 172| 0| } 173| | 174| | /* get encoding for CHECKCOMPOUNDCASE */ 175| 38| if (!utf8) { ------------------ | Branch (175:7): [True: 20, False: 18] ------------------ 176| 20| csconv = get_current_cs(get_encoding()); 177| 5.14k| for (int i = 0; i <= 255; i++) { ------------------ | Branch (177:21): [True: 5.12k, False: 20] ------------------ 178| 5.12k| if ((csconv[i].cupper != csconv[i].clower) && ------------------ | Branch (178:11): [True: 2.24k, False: 2.88k] ------------------ 179| 2.24k| (wordchars.find((char)i) == std::string::npos)) { ------------------ | Branch (179:11): [True: 2.24k, False: 0] ------------------ 180| 2.24k| wordchars.push_back((char)i); 181| 2.24k| } 182| 5.12k| } 183| 20| } 184| | 185| | // default BREAK definition 186| 38| if (!parsedbreaktable) { ------------------ | Branch (186:7): [True: 34, False: 4] ------------------ 187| 34| breaktable.emplace_back("-"); 188| 34| breaktable.emplace_back("^-"); 189| 34| breaktable.emplace_back("-$"); 190| 34| parsedbreaktable = true; 191| 34| } 192| | 193| 38|#if defined(FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION) 194| | // not entirely sure this is invalid, so only for fuzzing for now 195| 38| if (iconvtable && !iconvtable->check_against_breaktable(breaktable)) { ------------------ | Branch (195:7): [True: 2, False: 36] | Branch (195:21): [True: 0, False: 2] ------------------ 196| 0| delete iconvtable; 197| 0| iconvtable = nullptr; 198| 0| } 199| 38|#endif 200| | 201| 38| if (cpdmin == -1) ------------------ | Branch (201:7): [True: 34, False: 4] ------------------ 202| 34| cpdmin = MINCPDLEN; ------------------ | | 88| 34|#define MINCPDLEN 3 ------------------ 203| 38|} _ZN8AffixMgr13finishFileMgrEP7FileMgr: 260| 38|void AffixMgr::finishFileMgr(FileMgr* afflst) { 261| 38| delete afflst; 262| | 263| | // convert affix trees to sorted list 264| 38| process_pfx_tree_to_list(); 265| 38| process_sfx_tree_to_list(); 266| 38|} _ZN8AffixMgr10parse_fileEPKcS1_: 269| 38|int AffixMgr::parse_file(const char* affpath, const char* key) { 270| | 271| | // checking flag duplication 272| 38| char dupflags[CONTSIZE]; 273| 38| char dupflags_ini = 1; 274| | 275| | // first line indicator for removing byte order mark 276| 38| int firstline = 1; 277| | 278| | // open the affix file 279| 38| FileMgr* afflst = new FileMgr(affpath, key); 280| | 281| | // step one is to parse the affix file building up the internal 282| | // affix data structures 283| | 284| | // read in each line ignoring any that do not 285| | // start with a known line type indicator 286| 38| std::string line; 287| 646| while (afflst->getline(line)) { ------------------ | Branch (287:10): [True: 608, False: 38] ------------------ 288| 608| mychomp(line); 289| | 290| | /* remove byte order mark */ 291| 608| if (firstline) { ------------------ | Branch (291:9): [True: 38, False: 570] ------------------ 292| 38| firstline = 0; 293| | // Affix file begins with byte order mark: possible incompatibility with 294| | // old Hunspell versions 295| 38| if (line.compare(0, 3, "\xEF\xBB\xBF", 3) == 0) { ------------------ | Branch (295:11): [True: 0, False: 38] ------------------ 296| 0| line.erase(0, 3); 297| 0| } 298| 38| } 299| | 300| | /* parse in the keyboard string */ 301| 608| if (line.compare(0, 3, "KEY", 3) == 0) { ------------------ | Branch (301:9): [True: 0, False: 608] ------------------ 302| 0| if (!parse_string(line, keystring, afflst->getlinenum())) { ------------------ | Branch (302:11): [True: 0, False: 0] ------------------ 303| 0| finishFileMgr(afflst); 304| 0| return 1; 305| 0| } 306| 0| } 307| | 308| | /* parse in the try string */ 309| 608| if (line.compare(0, 3, "TRY", 3) == 0) { ------------------ | Branch (309:9): [True: 2, False: 606] ------------------ 310| 2| if (!parse_string(line, trystring, afflst->getlinenum())) { ------------------ | Branch (310:11): [True: 0, False: 2] ------------------ 311| 0| finishFileMgr(afflst); 312| 0| return 1; 313| 0| } 314| 2| } 315| | 316| | /* parse in the name of the character set used by the .dict and .aff */ 317| 608| if (line.compare(0, 3, "SET", 3) == 0) { ------------------ | Branch (317:9): [True: 18, False: 590] ------------------ 318| 18| if (!parse_string(line, encoding, afflst->getlinenum())) { ------------------ | Branch (318:11): [True: 0, False: 18] ------------------ 319| 0| finishFileMgr(afflst); 320| 0| return 1; 321| 0| } 322| 18| if (encoding == "UTF-8") { ------------------ | Branch (322:11): [True: 18, False: 0] ------------------ 323| 18| utf8 = 1; 324| 18| } 325| 18| } 326| | 327| | /* parse COMPLEXPREFIXES for agglutinative languages with right-to-left 328| | * writing system */ 329| 608| if (line.compare(0, 15, "COMPLEXPREFIXES", 15) == 0) ------------------ | Branch (329:9): [True: 2, False: 606] ------------------ 330| 2| complexprefixes = 1; 331| | 332| | /* parse in the flag used by the controlled compound words */ 333| 608| if (line.compare(0, 12, "COMPOUNDFLAG", 12) == 0) { ------------------ | Branch (333:9): [True: 8, False: 600] ------------------ 334| 8| if (!parse_flag(line, &compoundflag, afflst)) { ------------------ | Branch (334:11): [True: 0, False: 8] ------------------ 335| 0| finishFileMgr(afflst); 336| 0| return 1; 337| 0| } 338| 8| } 339| | 340| | /* parse in the flag used by compound words */ 341| 608| if (line.compare(0, 13, "COMPOUNDBEGIN", 13) == 0) { ------------------ | Branch (341:9): [True: 0, False: 608] ------------------ 342| 0| if (complexprefixes) { ------------------ | Branch (342:11): [True: 0, False: 0] ------------------ 343| 0| if (!parse_flag(line, &compoundend, afflst)) { ------------------ | Branch (343:13): [True: 0, False: 0] ------------------ 344| 0| finishFileMgr(afflst); 345| 0| return 1; 346| 0| } 347| 0| } else { 348| 0| if (!parse_flag(line, &compoundbegin, afflst)) { ------------------ | Branch (348:13): [True: 0, False: 0] ------------------ 349| 0| finishFileMgr(afflst); 350| 0| return 1; 351| 0| } 352| 0| } 353| 0| } 354| | 355| | /* parse in the flag used by compound words */ 356| 608| if (line.compare(0, 14, "COMPOUNDMIDDLE", 14) == 0) { ------------------ | Branch (356:9): [True: 0, False: 608] ------------------ 357| 0| if (!parse_flag(line, &compoundmiddle, afflst)) { ------------------ | Branch (357:11): [True: 0, False: 0] ------------------ 358| 0| finishFileMgr(afflst); 359| 0| return 1; 360| 0| } 361| 0| } 362| | 363| | /* parse in the flag used by compound words */ 364| 608| if (line.compare(0, 11, "COMPOUNDEND", 11) == 0) { ------------------ | Branch (364:9): [True: 0, False: 608] ------------------ 365| 0| if (complexprefixes) { ------------------ | Branch (365:11): [True: 0, False: 0] ------------------ 366| 0| if (!parse_flag(line, &compoundbegin, afflst)) { ------------------ | Branch (366:13): [True: 0, False: 0] ------------------ 367| 0| finishFileMgr(afflst); 368| 0| return 1; 369| 0| } 370| 0| } else { 371| 0| if (!parse_flag(line, &compoundend, afflst)) { ------------------ | Branch (371:13): [True: 0, False: 0] ------------------ 372| 0| finishFileMgr(afflst); 373| 0| return 1; 374| 0| } 375| 0| } 376| 0| } 377| | 378| | /* parse in the data used by compound_check() method */ 379| 608| if (line.compare(0, 15, "COMPOUNDWORDMAX", 15) == 0) { ------------------ | Branch (379:9): [True: 2, False: 606] ------------------ 380| 2| if (!parse_num(line, &cpdwordmax, afflst)) { ------------------ | Branch (380:11): [True: 0, False: 2] ------------------ 381| 0| finishFileMgr(afflst); 382| 0| return 1; 383| 0| } 384| 2| } 385| | 386| | /* parse in the flag sign compounds in dictionary */ 387| 608| if (line.compare(0, 12, "COMPOUNDROOT", 12) == 0) { ------------------ | Branch (387:9): [True: 0, False: 608] ------------------ 388| 0| if (!parse_flag(line, &compoundroot, afflst)) { ------------------ | Branch (388:11): [True: 0, False: 0] ------------------ 389| 0| finishFileMgr(afflst); 390| 0| return 1; 391| 0| } 392| 0| } 393| | 394| | /* parse in the flag used by compound_check() method */ 395| 608| if (line.compare(0, 18, "COMPOUNDPERMITFLAG", 18) == 0) { ------------------ | Branch (395:9): [True: 2, False: 606] ------------------ 396| 2| if (!parse_flag(line, &compoundpermitflag, afflst)) { ------------------ | Branch (396:11): [True: 0, False: 2] ------------------ 397| 0| finishFileMgr(afflst); 398| 0| return 1; 399| 0| } 400| 2| } 401| | 402| | /* parse in the flag used by compound_check() method */ 403| 608| if (line.compare(0, 18, "COMPOUNDFORBIDFLAG", 18) == 0) { ------------------ | Branch (403:9): [True: 2, False: 606] ------------------ 404| 2| if (!parse_flag(line, &compoundforbidflag, afflst)) { ------------------ | Branch (404:11): [True: 0, False: 2] ------------------ 405| 0| finishFileMgr(afflst); 406| 0| return 1; 407| 0| } 408| 2| } 409| | 410| 608| if (line.compare(0, 20, "COMPOUNDMORESUFFIXES", 20) == 0) { ------------------ | Branch (410:9): [True: 0, False: 608] ------------------ 411| 0| compoundmoresuffixes = 1; 412| 0| } 413| | 414| 608| if (line.compare(0, 16, "CHECKCOMPOUNDDUP", 16) == 0) { ------------------ | Branch (414:9): [True: 0, False: 608] ------------------ 415| 0| checkcompounddup = 1; 416| 0| } 417| | 418| 608| if (line.compare(0, 16, "CHECKCOMPOUNDREP", 16) == 0) { ------------------ | Branch (418:9): [True: 2, False: 606] ------------------ 419| 2| checkcompoundrep = 1; 420| 2| } 421| | 422| 608| if (line.compare(0, 19, "CHECKCOMPOUNDTRIPLE", 19) == 0) { ------------------ | Branch (422:9): [True: 0, False: 608] ------------------ 423| 0| checkcompoundtriple = 1; 424| 0| } 425| | 426| 608| if (line.compare(0, 16, "SIMPLIFIEDTRIPLE", 16) == 0) { ------------------ | Branch (426:9): [True: 0, False: 608] ------------------ 427| 0| simplifiedtriple = 1; 428| 0| } 429| | 430| 608| if (line.compare(0, 17, "CHECKCOMPOUNDCASE", 17) == 0) { ------------------ | Branch (430:9): [True: 0, False: 608] ------------------ 431| 0| checkcompoundcase = 1; 432| 0| } 433| | 434| 608| if (line.compare(0, 9, "NOSUGGEST", 9) == 0) { ------------------ | Branch (434:9): [True: 0, False: 608] ------------------ 435| 0| if (!parse_flag(line, &nosuggest, afflst)) { ------------------ | Branch (435:11): [True: 0, False: 0] ------------------ 436| 0| finishFileMgr(afflst); 437| 0| return 1; 438| 0| } 439| 0| } 440| | 441| 608| if (line.compare(0, 14, "NONGRAMSUGGEST", 14) == 0) { ------------------ | Branch (441:9): [True: 0, False: 608] ------------------ 442| 0| if (!parse_flag(line, &nongramsuggest, afflst)) { ------------------ | Branch (442:11): [True: 0, False: 0] ------------------ 443| 0| finishFileMgr(afflst); 444| 0| return 1; 445| 0| } 446| 0| } 447| | 448| | /* parse in the flag used by forbidden words */ 449| 608| if (line.compare(0, 13, "FORBIDDENWORD", 13) == 0) { ------------------ | Branch (449:9): [True: 2, False: 606] ------------------ 450| 2| if (!parse_flag(line, &forbiddenword, afflst)) { ------------------ | Branch (450:11): [True: 0, False: 2] ------------------ 451| 0| finishFileMgr(afflst); 452| 0| return 1; 453| 0| } 454| 2| } 455| | 456| | /* parse in the flag used by forbidden words (is deprecated) */ 457| 608| if (line.compare(0, 13, "LEMMA_PRESENT", 13) == 0) { ------------------ | Branch (457:9): [True: 0, False: 608] ------------------ 458| 0| if (!parse_flag(line, &lemma_present, afflst)) { ------------------ | Branch (458:11): [True: 0, False: 0] ------------------ 459| 0| finishFileMgr(afflst); 460| 0| return 1; 461| 0| } 462| 0| } 463| | 464| | /* parse in the flag used by circumfixes */ 465| 608| if (line.compare(0, 9, "CIRCUMFIX", 9) == 0) { ------------------ | Branch (465:9): [True: 2, False: 606] ------------------ 466| 2| if (!parse_flag(line, &circumfix, afflst)) { ------------------ | Branch (466:11): [True: 0, False: 2] ------------------ 467| 0| finishFileMgr(afflst); 468| 0| return 1; 469| 0| } 470| 2| } 471| | 472| | /* parse in the flag used by fogemorphemes */ 473| 608| if (line.compare(0, 14, "ONLYINCOMPOUND", 14) == 0) { ------------------ | Branch (473:9): [True: 0, False: 608] ------------------ 474| 0| if (!parse_flag(line, &onlyincompound, afflst)) { ------------------ | Branch (474:11): [True: 0, False: 0] ------------------ 475| 0| finishFileMgr(afflst); 476| 0| return 1; 477| 0| } 478| 0| } 479| | 480| | /* parse in the flag used by `needaffixs' (is deprecated) */ 481| 608| if (line.compare(0, 10, "PSEUDOROOT", 10) == 0) { ------------------ | Branch (481:9): [True: 0, False: 608] ------------------ 482| 0| if (!parse_flag(line, &needaffix, afflst)) { ------------------ | Branch (482:11): [True: 0, False: 0] ------------------ 483| 0| finishFileMgr(afflst); 484| 0| return 1; 485| 0| } 486| 0| } 487| | 488| | /* parse in the flag used by `needaffixs' */ 489| 608| if (line.compare(0, 9, "NEEDAFFIX", 9) == 0) { ------------------ | Branch (489:9): [True: 2, False: 606] ------------------ 490| 2| if (!parse_flag(line, &needaffix, afflst)) { ------------------ | Branch (490:11): [True: 0, False: 2] ------------------ 491| 0| finishFileMgr(afflst); 492| 0| return 1; 493| 0| } 494| 2| } 495| | 496| | /* parse in the minimal length for words in compounds */ 497| 608| if (line.compare(0, 11, "COMPOUNDMIN", 11) == 0) { ------------------ | Branch (497:9): [True: 4, False: 604] ------------------ 498| 4| if (!parse_num(line, &cpdmin, afflst)) { ------------------ | Branch (498:11): [True: 0, False: 4] ------------------ 499| 0| finishFileMgr(afflst); 500| 0| return 1; 501| 0| } 502| 4| if (cpdmin < 1) ------------------ | Branch (502:11): [True: 0, False: 4] ------------------ 503| 0| cpdmin = 1; 504| 4| } 505| | 506| | /* parse in the max. words and syllables in compounds */ 507| 608| if (line.compare(0, 16, "COMPOUNDSYLLABLE", 16) == 0) { ------------------ | Branch (507:9): [True: 2, False: 606] ------------------ 508| 2| if (!parse_cpdsyllable(line, afflst)) { ------------------ | Branch (508:11): [True: 0, False: 2] ------------------ 509| 0| finishFileMgr(afflst); 510| 0| return 1; 511| 0| } 512| 2| } 513| | 514| | /* parse in the flag used by compound_check() method */ 515| 608| if (line.compare(0, 11, "SYLLABLENUM", 11) == 0) { ------------------ | Branch (515:9): [True: 0, False: 608] ------------------ 516| 0| if (!parse_string(line, cpdsyllablenum, afflst->getlinenum())) { ------------------ | Branch (516:11): [True: 0, False: 0] ------------------ 517| 0| finishFileMgr(afflst); 518| 0| return 1; 519| 0| } 520| 0| } 521| | 522| | /* parse in the flag used by the controlled compound words */ 523| 608| if (line.compare(0, 8, "CHECKNUM", 8) == 0) { ------------------ | Branch (523:9): [True: 0, False: 608] ------------------ 524| 0| checknum = 1; 525| 0| } 526| | 527| | /* parse in the extra word characters */ 528| 608| if (line.compare(0, 9, "WORDCHARS", 9) == 0) { ------------------ | Branch (528:9): [True: 6, False: 602] ------------------ 529| 6| if (!parse_array(line, wordchars, wordchars_utf16, ------------------ | Branch (529:11): [True: 0, False: 6] ------------------ 530| 6| utf8, afflst->getlinenum())) { 531| 0| finishFileMgr(afflst); 532| 0| return 1; 533| 0| } 534| 6| } 535| | 536| | /* parse in the ignored characters (for example, Arabic optional diacretics 537| | * charachters */ 538| 608| if (line.compare(0, 6, "IGNORE", 6) == 0) { ------------------ | Branch (538:9): [True: 4, False: 604] ------------------ 539| 4| if (!parse_array(line, ignorechars, ignorechars_utf16, ------------------ | Branch (539:11): [True: 0, False: 4] ------------------ 540| 4| utf8, afflst->getlinenum())) { 541| 0| finishFileMgr(afflst); 542| 0| return 1; 543| 0| } 544| 4| } 545| | 546| | /* parse in the input conversion table */ 547| 608| if (line.compare(0, 5, "ICONV", 5) == 0) { ------------------ | Branch (547:9): [True: 2, False: 606] ------------------ 548| 2| if (!parse_convtable(line, afflst, &iconvtable, "ICONV")) { ------------------ | Branch (548:11): [True: 0, False: 2] ------------------ 549| 0| finishFileMgr(afflst); 550| 0| return 1; 551| 0| } 552| 2| } 553| | 554| | /* parse in the output conversion table */ 555| 608| if (line.compare(0, 5, "OCONV", 5) == 0) { ------------------ | Branch (555:9): [True: 2, False: 606] ------------------ 556| 2| if (!parse_convtable(line, afflst, &oconvtable, "OCONV")) { ------------------ | Branch (556:11): [True: 0, False: 2] ------------------ 557| 0| finishFileMgr(afflst); 558| 0| return 1; 559| 0| } 560| 2| } 561| | 562| | /* parse in the phonetic translation table */ 563| 608| if (line.compare(0, 5, "PHONE", 5) == 0) { ------------------ | Branch (563:9): [True: 2, False: 606] ------------------ 564| 2| if (!parse_phonetable(line, afflst)) { ------------------ | Branch (564:11): [True: 0, False: 2] ------------------ 565| 0| finishFileMgr(afflst); 566| 0| return 1; 567| 0| } 568| 2| } 569| | 570| | /* parse in the checkcompoundpattern table */ 571| 608| if (line.compare(0, 20, "CHECKCOMPOUNDPATTERN", 20) == 0) { ------------------ | Branch (571:9): [True: 2, False: 606] ------------------ 572| 2| if (!parse_checkcpdtable(line, afflst)) { ------------------ | Branch (572:11): [True: 0, False: 2] ------------------ 573| 0| finishFileMgr(afflst); 574| 0| return 1; 575| 0| } 576| 2| } 577| | 578| | /* parse in the defcompound table */ 579| 608| if (line.compare(0, 12, "COMPOUNDRULE", 12) == 0) { ------------------ | Branch (579:9): [True: 0, False: 608] ------------------ 580| 0| if (!parse_defcpdtable(line, afflst)) { ------------------ | Branch (580:11): [True: 0, False: 0] ------------------ 581| 0| finishFileMgr(afflst); 582| 0| return 1; 583| 0| } 584| 0| } 585| | 586| | /* parse in the related character map table */ 587| 608| if (line.compare(0, 3, "MAP", 3) == 0) { ------------------ | Branch (587:9): [True: 2, False: 606] ------------------ 588| 2| if (!parse_maptable(line, afflst)) { ------------------ | Branch (588:11): [True: 0, False: 2] ------------------ 589| 0| finishFileMgr(afflst); 590| 0| return 1; 591| 0| } 592| 2| } 593| | 594| | /* parse in the word breakpoints table */ 595| 608| if (line.compare(0, 5, "BREAK", 5) == 0) { ------------------ | Branch (595:9): [True: 4, False: 604] ------------------ 596| 4| if (!parse_breaktable(line, afflst)) { ------------------ | Branch (596:11): [True: 0, False: 4] ------------------ 597| 0| finishFileMgr(afflst); 598| 0| return 1; 599| 0| } 600| 4| } 601| | 602| | /* parse in the language for language specific codes */ 603| 608| if (line.compare(0, 4, "LANG", 4) == 0) { ------------------ | Branch (603:9): [True: 2, False: 606] ------------------ 604| 2| if (!parse_string(line, lang, afflst->getlinenum())) { ------------------ | Branch (604:11): [True: 0, False: 2] ------------------ 605| 0| finishFileMgr(afflst); 606| 0| return 1; 607| 0| } 608| 2| langnum = get_lang_num(lang); 609| 2| } 610| | 611| 608| if (line.compare(0, 7, "VERSION", 7) == 0) { ------------------ | Branch (611:9): [True: 0, False: 608] ------------------ 612| 0| size_t startpos = line.find_first_not_of(" \t", 7); 613| 0| if (startpos != std::string::npos) { ------------------ | Branch (613:11): [True: 0, False: 0] ------------------ 614| 0| version = line.substr(startpos); 615| 0| } 616| 0| } 617| | 618| 608| if (line.compare(0, 12, "MAXNGRAMSUGS", 12) == 0) { ------------------ | Branch (618:9): [True: 4, False: 604] ------------------ 619| 4| if (!parse_num(line, &maxngramsugs, afflst)) { ------------------ | Branch (619:11): [True: 0, False: 4] ------------------ 620| 0| finishFileMgr(afflst); 621| 0| return 1; 622| 0| } 623| 4| } 624| | 625| 608| if (line.compare(0, 11, "ONLYMAXDIFF", 11) == 0) ------------------ | Branch (625:9): [True: 0, False: 608] ------------------ 626| 0| onlymaxdiff = 1; 627| | 628| 608| if (line.compare(0, 7, "MAXDIFF", 7) == 0) { ------------------ | Branch (628:9): [True: 0, False: 608] ------------------ 629| 0| if (!parse_num(line, &maxdiff, afflst)) { ------------------ | Branch (629:11): [True: 0, False: 0] ------------------ 630| 0| finishFileMgr(afflst); 631| 0| return 1; 632| 0| } 633| 0| } 634| | 635| 608| if (line.compare(0, 10, "MAXCPDSUGS", 10) == 0) { ------------------ | Branch (635:9): [True: 0, False: 608] ------------------ 636| 0| if (!parse_num(line, &maxcpdsugs, afflst)) { ------------------ | Branch (636:11): [True: 0, False: 0] ------------------ 637| 0| finishFileMgr(afflst); 638| 0| return 1; 639| 0| } 640| 0| } 641| | 642| 608| if (line.compare(0, 11, "NOSPLITSUGS", 11) == 0) { ------------------ | Branch (642:9): [True: 0, False: 608] ------------------ 643| 0| nosplitsugs = 1; 644| 0| } 645| | 646| 608| if (line.compare(0, 9, "FULLSTRIP", 9) == 0) { ------------------ | Branch (646:9): [True: 0, False: 608] ------------------ 647| 0| fullstrip = 1; 648| 0| } 649| | 650| 608| if (line.compare(0, 12, "SUGSWITHDOTS", 12) == 0) { ------------------ | Branch (650:9): [True: 0, False: 608] ------------------ 651| 0| sugswithdots = 1; 652| 0| } 653| | 654| | /* parse in the flag used by forbidden words */ 655| 608| if (line.compare(0, 8, "KEEPCASE", 8) == 0) { ------------------ | Branch (655:9): [True: 2, False: 606] ------------------ 656| 2| if (!parse_flag(line, &keepcase, afflst)) { ------------------ | Branch (656:11): [True: 0, False: 2] ------------------ 657| 0| finishFileMgr(afflst); 658| 0| return 1; 659| 0| } 660| 2| } 661| | 662| | /* parse in the flag used by `forceucase' */ 663| 608| if (line.compare(0, 10, "FORCEUCASE", 10) == 0) { ------------------ | Branch (663:9): [True: 0, False: 608] ------------------ 664| 0| if (!parse_flag(line, &forceucase, afflst)) { ------------------ | Branch (664:11): [True: 0, False: 0] ------------------ 665| 0| finishFileMgr(afflst); 666| 0| return 1; 667| 0| } 668| 0| } 669| | 670| | /* parse in the flag used by `warn' */ 671| 608| if (line.compare(0, 4, "WARN", 4) == 0) { ------------------ | Branch (671:9): [True: 2, False: 606] ------------------ 672| 2| if (!parse_flag(line, &warn, afflst)) { ------------------ | Branch (672:11): [True: 0, False: 2] ------------------ 673| 0| finishFileMgr(afflst); 674| 0| return 1; 675| 0| } 676| 2| } 677| | 678| 608| if (line.compare(0, 10, "FORBIDWARN", 10) == 0) { ------------------ | Branch (678:9): [True: 0, False: 608] ------------------ 679| 0| forbidwarn = 1; 680| 0| } 681| | 682| | /* parse in the flag used by the affix generator */ 683| 608| if (line.compare(0, 11, "SUBSTANDARD", 11) == 0) { ------------------ | Branch (683:9): [True: 0, False: 608] ------------------ 684| 0| if (!parse_flag(line, &substandard, afflst)) { ------------------ | Branch (684:11): [True: 0, False: 0] ------------------ 685| 0| finishFileMgr(afflst); 686| 0| return 1; 687| 0| } 688| 0| } 689| | 690| 608| if (line.compare(0, 11, "CHECKSHARPS", 11) == 0) { ------------------ | Branch (690:9): [True: 0, False: 608] ------------------ 691| 0| checksharps = 1; 692| 0| } 693| | 694| | /* parse this affix: P - prefix, S - suffix */ 695| | // affix type 696| 608| char ft = ' '; 697| 608| if (line.compare(0, 3, "PFX", 3) == 0) ------------------ | Branch (697:9): [True: 12, False: 596] ------------------ 698| 12| ft = complexprefixes ? 'S' : 'P'; ------------------ | Branch (698:12): [True: 4, False: 8] ------------------ 699| 608| if (line.compare(0, 3, "SFX", 3) == 0) ------------------ | Branch (699:9): [True: 14, False: 594] ------------------ 700| 14| ft = complexprefixes ? 'P' : 'S'; ------------------ | Branch (700:12): [True: 0, False: 14] ------------------ 701| 608| if (ft != ' ') { ------------------ | Branch (701:9): [True: 26, False: 582] ------------------ 702| 26| if (dupflags_ini) { ------------------ | Branch (702:11): [True: 18, False: 8] ------------------ 703| 18| memset(dupflags, 0, sizeof(dupflags)); 704| 18| dupflags_ini = 0; 705| 18| } 706| 26| if (!parse_affix(line, ft, afflst, dupflags)) { ------------------ | Branch (706:11): [True: 0, False: 26] ------------------ 707| 0| finishFileMgr(afflst); 708| 0| return 1; 709| 0| } 710| 26| } 711| 608| } 712| | 713| 38| finishFileMgr(afflst); 714| | // affix trees are sorted now 715| | 716| | // now we can speed up performance greatly taking advantage of the 717| | // relationship between the affixes and the idea of "subsets". 718| | 719| | // View each prefix as a potential leading subset of another and view 720| | // each suffix (reversed) as a potential trailing subset of another. 721| | 722| | // To illustrate this relationship if we know the prefix "ab" is found in the 723| | // word to examine, only prefixes that "ab" is a leading subset of need be 724| | // examined. 725| | // Furthermore is "ab" is not present then none of the prefixes that "ab" is 726| | // is a subset need be examined. 727| | // The same argument goes for suffix string that are reversed. 728| | 729| | // Then to top this off why not examine the first char of the word to quickly 730| | // limit the set of prefixes to examine (i.e. the prefixes to examine must 731| | // be leading supersets of the first character of the word (if they exist) 732| | 733| | // To take advantage of this "subset" relationship, we need to add two links 734| | // from entry. One to take next if the current prefix is found (call it 735| | // nexteq) 736| | // and one to take next if the current prefix is not found (call it nextne). 737| | 738| | // Since we have built ordered lists, all that remains is to properly 739| | // initialize 740| | // the nextne and nexteq pointers that relate them 741| | 742| 38| process_pfx_order(); 743| 38| process_sfx_order(); 744| | 745| 38| return 0; 746| 38|} _ZN8AffixMgr13build_pfxtreeEP8PfxEntry: 752| 8|int AffixMgr::build_pfxtree(PfxEntry* pfxptr) { 753| 8| PfxEntry* ptr; 754| 8| PfxEntry* pptr; 755| 8| PfxEntry* ep = pfxptr; 756| | 757| | // get the right starting points 758| 8| const char* key = ep->getKey(); 759| 8| const auto flg = (unsigned char)(ep->getFlag() & 0x00FF); 760| | 761| | // first index by flag which must exist 762| 8| ptr = pFlag[flg]; 763| 8| ep->setFlgNxt(ptr); 764| 8| pFlag[flg] = ep; 765| | 766| | // handle the special case of null affix string 767| 8| if (*key == '\0') { ------------------ | Branch (767:7): [True: 2, False: 6] ------------------ 768| | // always inset them at head of list at element 0 769| 2| ptr = pStart[0]; 770| 2| ep->setNext(ptr); 771| 2| pStart[0] = ep; 772| 2| return 0; 773| 2| } 774| | 775| | // now handle the normal case 776| 6| ep->setNextEQ(nullptr); 777| 6| ep->setNextNE(nullptr); 778| | 779| 6| unsigned char sp = *((const unsigned char*)key); 780| 6| ptr = pStart[sp]; 781| | 782| | // handle the first insert 783| 6| if (!ptr) { ------------------ | Branch (783:7): [True: 4, False: 2] ------------------ 784| 4| pStart[sp] = ep; 785| 4| return 0; 786| 4| } 787| | 788| | // otherwise use binary tree insertion so that a sorted 789| | // list can easily be generated later 790| 2| pptr = nullptr; 791| 2| for (;;) { 792| 2| pptr = ptr; 793| 2| if (strcmp(ep->getKey(), ptr->getKey()) <= 0) { ------------------ | Branch (793:9): [True: 0, False: 2] ------------------ 794| 0| ptr = ptr->getNextEQ(); 795| 0| if (!ptr) { ------------------ | Branch (795:11): [True: 0, False: 0] ------------------ 796| 0| pptr->setNextEQ(ep); 797| 0| break; 798| 0| } 799| 2| } else { 800| 2| ptr = ptr->getNextNE(); 801| 2| if (!ptr) { ------------------ | Branch (801:11): [True: 2, False: 0] ------------------ 802| 2| pptr->setNextNE(ep); 803| 2| break; 804| 2| } 805| 2| } 806| 2| } 807| 2| return 0; 808| 6|} _ZN8AffixMgr13build_sfxtreeEP8SfxEntry: 813| 34|int AffixMgr::build_sfxtree(SfxEntry* sfxptr) { 814| | 815| 34| sfxptr->initReverseWord(); 816| | 817| 34| SfxEntry* ptr; 818| 34| SfxEntry* pptr; 819| 34| SfxEntry* ep = sfxptr; 820| | 821| | /* get the right starting point */ 822| 34| const char* key = ep->getKey(); 823| 34| const auto flg = (unsigned char)(ep->getFlag() & 0x00FF); 824| | 825| | // first index by flag which must exist 826| 34| ptr = sFlag[flg]; 827| 34| ep->setFlgNxt(ptr); 828| 34| sFlag[flg] = ep; 829| | 830| | // next index by affix string 831| | 832| | // handle the special case of null affix string 833| 34| if (*key == '\0') { ------------------ | Branch (833:7): [True: 0, False: 34] ------------------ 834| | // always inset them at head of list at element 0 835| 0| ptr = sStart[0]; 836| 0| ep->setNext(ptr); 837| 0| sStart[0] = ep; 838| 0| return 0; 839| 0| } 840| | 841| | // now handle the normal case 842| 34| ep->setNextEQ(nullptr); 843| 34| ep->setNextNE(nullptr); 844| | 845| 34| unsigned char sp = *((const unsigned char*)key); 846| 34| ptr = sStart[sp]; 847| | 848| | // handle the first insert 849| 34| if (!ptr) { ------------------ | Branch (849:7): [True: 22, False: 12] ------------------ 850| 22| sStart[sp] = ep; 851| 22| return 0; 852| 22| } 853| | 854| | // otherwise use binary tree insertion so that a sorted 855| | // list can easily be generated later 856| 12| pptr = nullptr; 857| 20| for (;;) { 858| 20| pptr = ptr; 859| 20| if (strcmp(ep->getKey(), ptr->getKey()) <= 0) { ------------------ | Branch (859:9): [True: 18, False: 2] ------------------ 860| 18| ptr = ptr->getNextEQ(); 861| 18| if (!ptr) { ------------------ | Branch (861:11): [True: 10, False: 8] ------------------ 862| 10| pptr->setNextEQ(ep); 863| 10| break; 864| 10| } 865| 18| } else { 866| 2| ptr = ptr->getNextNE(); 867| 2| if (!ptr) { ------------------ | Branch (867:11): [True: 2, False: 0] ------------------ 868| 2| pptr->setNextNE(ep); 869| 2| break; 870| 2| } 871| 2| } 872| 20| } 873| 12| return 0; 874| 34|} _ZN8AffixMgr24process_pfx_tree_to_listEv: 877| 38|int AffixMgr::process_pfx_tree_to_list() { 878| 9.72k| for (int i = 1; i < SETSIZE; i++) { ------------------ | | 61| 9.72k|#define SETSIZE 256 ------------------ | Branch (878:19): [True: 9.69k, False: 38] ------------------ 879| 9.69k| pStart[i] = process_pfx_in_order(pStart[i], nullptr); 880| 9.69k| } 881| 38| return 0; 882| 38|} _ZN8AffixMgr20process_pfx_in_orderEP8PfxEntryS1_: 884| 9.70k|PfxEntry* AffixMgr::process_pfx_in_order(PfxEntry* ptr, PfxEntry* nptr) { 885| 9.70k| if (ptr) { ------------------ | Branch (885:7): [True: 6, False: 9.69k] ------------------ 886| 6| nptr = process_pfx_in_order(ptr->getNextNE(), nptr); 887| 6| ptr->setNext(nptr); 888| 6| nptr = process_pfx_in_order(ptr->getNextEQ(), ptr); 889| 6| } 890| 9.70k| return nptr; 891| 9.70k|} _ZN8AffixMgr24process_sfx_tree_to_listEv: 894| 38|int AffixMgr::process_sfx_tree_to_list() { 895| 9.72k| for (int i = 1; i < SETSIZE; i++) { ------------------ | | 61| 9.72k|#define SETSIZE 256 ------------------ | Branch (895:19): [True: 9.69k, False: 38] ------------------ 896| 9.69k| sStart[i] = process_sfx_in_order(sStart[i], nullptr); 897| 9.69k| } 898| 38| return 0; 899| 38|} _ZN8AffixMgr20process_sfx_in_orderEP8SfxEntryS1_: 901| 9.75k|SfxEntry* AffixMgr::process_sfx_in_order(SfxEntry* ptr, SfxEntry* nptr) { 902| 9.75k| if (ptr) { ------------------ | Branch (902:7): [True: 34, False: 9.72k] ------------------ 903| 34| nptr = process_sfx_in_order(ptr->getNextNE(), nptr); 904| 34| ptr->setNext(nptr); 905| 34| nptr = process_sfx_in_order(ptr->getNextEQ(), ptr); 906| 34| } 907| 9.75k| return nptr; 908| 9.75k|} _ZN8AffixMgr17process_pfx_orderEv: 912| 38|int AffixMgr::process_pfx_order() { 913| 38| PfxEntry* ptr; 914| | 915| | // loop through each prefix list starting point 916| 9.72k| for (int i = 1; i < SETSIZE; i++) { ------------------ | | 61| 9.72k|#define SETSIZE 256 ------------------ | Branch (916:19): [True: 9.69k, False: 38] ------------------ 917| 9.69k| ptr = pStart[i]; 918| | 919| | // look through the remainder of the list 920| | // and find next entry with affix that 921| | // the current one is not a subset of 922| | // mark that as destination for NextNE 923| | // use next in list that you are a subset 924| | // of as NextEQ 925| | 926| 9.69k| for (; ptr != nullptr; ptr = ptr->getNext()) { ------------------ | Branch (926:12): [True: 6, False: 9.69k] ------------------ 927| 6| PfxEntry* nptr = ptr->getNext(); 928| 8| for (; nptr != nullptr; nptr = nptr->getNext()) { ------------------ | Branch (928:14): [True: 2, False: 6] ------------------ 929| 2| if (!isSubset(ptr->getKey(), nptr->getKey())) ------------------ | Branch (929:13): [True: 0, False: 2] ------------------ 930| 0| break; 931| 2| } 932| 6| ptr->setNextNE(nptr); 933| 6| ptr->setNextEQ(nullptr); 934| 6| if ((ptr->getNext()) && ------------------ | Branch (934:11): [True: 2, False: 4] ------------------ 935| 2| isSubset(ptr->getKey(), (ptr->getNext())->getKey())) ------------------ | Branch (935:11): [True: 2, False: 0] ------------------ 936| 2| ptr->setNextEQ(ptr->getNext()); 937| 6| } 938| | 939| | // now clean up by adding smart search termination strings: 940| | // if you are already a superset of the previous prefix 941| | // but not a subset of the next, search can end here 942| | // so set NextNE properly 943| | 944| 9.69k| ptr = pStart[i]; 945| 9.69k| for (; ptr != nullptr; ptr = ptr->getNext()) { ------------------ | Branch (945:12): [True: 6, False: 9.69k] ------------------ 946| 6| PfxEntry* nptr = ptr->getNext(); 947| 6| PfxEntry* mptr = nullptr; 948| 8| for (; nptr != nullptr; nptr = nptr->getNext()) { ------------------ | Branch (948:14): [True: 2, False: 6] ------------------ 949| 2| if (!isSubset(ptr->getKey(), nptr->getKey())) ------------------ | Branch (949:13): [True: 0, False: 2] ------------------ 950| 0| break; 951| 2| mptr = nptr; 952| 2| } 953| 6| if (mptr) ------------------ | Branch (953:11): [True: 2, False: 4] ------------------ 954| 2| mptr->setNextNE(nullptr); 955| 6| } 956| 9.69k| } 957| 38| return 0; 958| 38|} _ZN8AffixMgr17process_sfx_orderEv: 962| 38|int AffixMgr::process_sfx_order() { 963| 38| SfxEntry* ptr; 964| | 965| | // loop through each prefix list starting point 966| 9.72k| for (int i = 1; i < SETSIZE; i++) { ------------------ | | 61| 9.72k|#define SETSIZE 256 ------------------ | Branch (966:19): [True: 9.69k, False: 38] ------------------ 967| 9.69k| ptr = sStart[i]; 968| | 969| | // look through the remainder of the list 970| | // and find next entry with affix that 971| | // the current one is not a subset of 972| | // mark that as destination for NextNE 973| | // use next in list that you are a subset 974| | // of as NextEQ 975| | 976| 9.72k| for (; ptr != nullptr; ptr = ptr->getNext()) { ------------------ | Branch (976:12): [True: 34, False: 9.69k] ------------------ 977| 34| SfxEntry* nptr = ptr->getNext(); 978| 52| for (; nptr != nullptr; nptr = nptr->getNext()) { ------------------ | Branch (978:14): [True: 20, False: 32] ------------------ 979| 20| if (!isSubset(ptr->getKey(), nptr->getKey())) ------------------ | Branch (979:13): [True: 2, False: 18] ------------------ 980| 2| break; 981| 20| } 982| 34| ptr->setNextNE(nptr); 983| 34| ptr->setNextEQ(nullptr); 984| 34| if ((ptr->getNext()) && ------------------ | Branch (984:11): [True: 12, False: 22] ------------------ 985| 12| isSubset(ptr->getKey(), (ptr->getNext())->getKey())) ------------------ | Branch (985:11): [True: 10, False: 2] ------------------ 986| 10| ptr->setNextEQ(ptr->getNext()); 987| 34| } 988| | 989| | // now clean up by adding smart search termination strings: 990| | // if you are already a superset of the previous suffix 991| | // but not a subset of the next, search can end here 992| | // so set NextNE properly 993| | 994| 9.69k| ptr = sStart[i]; 995| 9.72k| for (; ptr != nullptr; ptr = ptr->getNext()) { ------------------ | Branch (995:12): [True: 34, False: 9.69k] ------------------ 996| 34| SfxEntry* nptr = ptr->getNext(); 997| 34| SfxEntry* mptr = nullptr; 998| 52| for (; nptr != nullptr; nptr = nptr->getNext()) { ------------------ | Branch (998:14): [True: 20, False: 32] ------------------ 999| 20| if (!isSubset(ptr->getKey(), nptr->getKey())) ------------------ | Branch (999:13): [True: 2, False: 18] ------------------ 1000| 2| break; 1001| 18| mptr = nptr; 1002| 18| } 1003| 34| if (mptr) ------------------ | Branch (1003:11): [True: 10, False: 24] ------------------ 1004| 10| mptr->setNextNE(nullptr); 1005| 34| } 1006| 9.69k| } 1007| 38| return 0; 1008| 38|} _ZN8AffixMgr7condlenERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEE: 1021| 8|int AffixMgr::condlen(const std::string& s) { 1022| 8| int l = 0; 1023| 8| bool group = false; 1024| 8| size_t i = 0; 1025| 42| while (i < s.size()) { ------------------ | Branch (1025:10): [True: 34, False: 8] ------------------ 1026| 34| if (s[i] == '[') { ------------------ | Branch (1026:9): [True: 4, False: 30] ------------------ 1027| 4| group = true; 1028| 4| ++l; 1029| 4| ++i; 1030| 30| } else if (s[i] == ']') { ------------------ | Branch (1030:16): [True: 4, False: 26] ------------------ 1031| 4| group = false; 1032| 4| ++i; 1033| 26| } else if (group) { ------------------ | Branch (1033:16): [True: 18, False: 8] ------------------ 1034| 18| ++i; 1035| 18| } else { 1036| 8| ++l; 1037| 8| i = utf8 ? utf8_next(s, i) : i + 1; ------------------ | Branch (1037:11): [True: 8, False: 0] ------------------ 1038| 8| } 1039| 34| } 1040| 8| return l; 1041| 8|} _ZN8AffixMgr8encodeitER8AffEntryRKNSt3__112basic_stringIcNS2_11char_traitsIcEENS2_9allocatorIcEEEE: 1043| 42|int AffixMgr::encodeit(AffEntry& entry, const std::string& cs) { 1044| 42| if (cs.compare(".") != 0) { ------------------ | Branch (1044:7): [True: 8, False: 34] ------------------ 1045| 8| entry.numconds = (char)condlen(cs); 1046| 8| const size_t cslen = cs.size(); 1047| 8| const size_t short_part = std::min(MAXCONDLEN, cslen); ------------------ | | 90| 8|#define MAXCONDLEN 20 ------------------ 1048| 8| memcpy(entry.c.conds, cs.data(), short_part); 1049| 8| if (short_part < MAXCONDLEN) { ------------------ | | 90| 8|#define MAXCONDLEN 20 ------------------ | Branch (1049:9): [True: 8, False: 0] ------------------ 1050| | //blank out the remaining space 1051| 8| memset(entry.c.conds + short_part, 0, MAXCONDLEN - short_part); ------------------ | | 90| 8|#define MAXCONDLEN 20 ------------------ 1052| 8| } else if (cs[MAXCONDLEN]) { ------------------ | | 90| 0|#define MAXCONDLEN 20 ------------------ | Branch (1052:16): [True: 0, False: 0] ------------------ 1053| | //there is more conditions than fit in fixed space, so its 1054| | //a long condition 1055| 0| entry.opts |= aeLONGCOND; ------------------ | | 69| 0|#define aeLONGCOND (1 << 4) ------------------ 1056| 0| size_t remaining = cs.size() - MAXCONDLEN_1; ------------------ | | 91| 0|#define MAXCONDLEN_1 (MAXCONDLEN - sizeof(char*)) | | ------------------ | | | | 90| 0|#define MAXCONDLEN 20 | | ------------------ ------------------ 1057| 0| entry.c.l.conds2 = new char[1 + remaining]; 1058| 0| memcpy(entry.c.l.conds2, cs.data() + MAXCONDLEN_1, remaining); ------------------ | | 91| 0|#define MAXCONDLEN_1 (MAXCONDLEN - sizeof(char*)) | | ------------------ | | | | 90| 0|#define MAXCONDLEN 20 | | ------------------ ------------------ 1059| 0| entry.c.l.conds2[remaining] = 0; 1060| 0| } 1061| 34| } else { 1062| 34| entry.numconds = 0; 1063| 34| entry.c.conds[0] = '\0'; 1064| 34| } 1065| 42| return 0; 1066| 42|} _ZN8AffixMgr12prefix_checkERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEiicR12AffixScratcht: 1083| 525M| const FLAG needflag) { 1084| 525M| struct hentry* rv = nullptr; 1085| | 1086| 525M| pfx = nullptr; 1087| 525M| pfxappnd = nullptr; 1088| 525M| sfxappnd = nullptr; 1089| 525M| sfxextra = 0; 1090| | 1091| | // first handle the special case of 0 length prefixes 1092| 525M| PfxEntry* pe = pStart[0]; 1093| 526M| while (pe) { ------------------ | Branch (1093:10): [True: 1.27M, False: 525M] ------------------ 1094| 1.27M| if ( 1095| | // fogemorpheme 1096| 1.27M| ((in_compound != IN_CPD_NOT) || ------------------ | | 72| 1.27M|#define IN_CPD_NOT 0 ------------------ | Branch (1096:10): [True: 0, False: 1.27M] ------------------ 1097| 1.27M| !(pe->getCont() && ------------------ | Branch (1097:12): [True: 1.27M, False: 0] ------------------ 1098| 1.27M| (TESTAFF(pe->getCont(), onlyincompound, pe->getContLen())))) && ------------------ | | 99| 1.27M|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (1098:12): [True: 0, False: 1.27M] ------------------ 1099| | // permit prefixes in compounds 1100| 1.27M| ((in_compound != IN_CPD_END) || ------------------ | | 74| 1.27M|#define IN_CPD_END 2 ------------------ | Branch (1100:10): [True: 1.27M, False: 0] ------------------ 1101| 0| (pe->getCont() && ------------------ | Branch (1101:11): [True: 0, False: 0] ------------------ 1102| 1.27M| (TESTAFF(pe->getCont(), compoundpermitflag, pe->getContLen()))))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (1102:11): [True: 0, False: 0] ------------------ 1103| | // check prefix 1104| 1.27M| rv = pe->checkword(word, start, len, in_compound, needflag, scratch); 1105| 1.27M| if (rv) { ------------------ | Branch (1105:11): [True: 0, False: 1.27M] ------------------ 1106| 0| pfx = pe; // BUG: pfx not stateless 1107| 0| return rv; 1108| 0| } 1109| 1.27M| } 1110| 1.27M| pe = pe->getNext(); 1111| 1.27M| } 1112| | 1113| | // now handle the general case 1114| 525M| unsigned char sp = word[start]; 1115| 525M| PfxEntry* pptr = pStart[sp]; 1116| | 1117| 525M| while (pptr) { ------------------ | Branch (1117:10): [True: 60.0k, False: 525M] ------------------ 1118| 60.0k| if (isSubset(pptr->getKey(), word.c_str() + start)) { ------------------ | Branch (1118:9): [True: 40.2k, False: 19.7k] ------------------ 1119| 40.2k| if ( 1120| | // fogemorpheme 1121| 40.2k| ((in_compound != IN_CPD_NOT) || ------------------ | | 72| 40.2k|#define IN_CPD_NOT 0 ------------------ | Branch (1121:12): [True: 0, False: 40.2k] ------------------ 1122| 40.2k| !(pptr->getCont() && ------------------ | Branch (1122:14): [True: 35.8k, False: 4.38k] ------------------ 1123| 35.8k| (TESTAFF(pptr->getCont(), onlyincompound, pptr->getContLen())))) && ------------------ | | 99| 35.8k|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (1123:14): [True: 0, False: 35.8k] ------------------ 1124| | // permit prefixes in compounds 1125| 40.2k| ((in_compound != IN_CPD_END) || ------------------ | | 74| 40.2k|#define IN_CPD_END 2 ------------------ | Branch (1125:12): [True: 40.2k, False: 0] ------------------ 1126| 0| (pptr->getCont() && (TESTAFF(pptr->getCont(), compoundpermitflag, ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (1126:13): [True: 0, False: 0] | Branch (1126:32): [True: 0, False: 0] ------------------ 1127| 40.2k| pptr->getContLen()))))) { 1128| | // check prefix 1129| 40.2k| rv = pptr->checkword(word, start, len, in_compound, needflag, scratch); 1130| 40.2k| if (rv) { ------------------ | Branch (1130:13): [True: 0, False: 40.2k] ------------------ 1131| 0| pfx = pptr; // BUG: pfx not stateless 1132| 0| return rv; 1133| 0| } 1134| 40.2k| } 1135| 40.2k| pptr = pptr->getNextEQ(); 1136| 40.2k| } else { 1137| 19.7k| pptr = pptr->getNextNE(); 1138| 19.7k| } 1139| 60.0k| } 1140| | 1141| 525M| return nullptr; 1142| 525M|} _ZN8AffixMgr19prefix_check_twosfxERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEiicR12AffixScratcht: 1150| 14.7M| const FLAG needflag) { 1151| 14.7M| struct hentry* rv = nullptr; 1152| | 1153| 14.7M| pfx = nullptr; 1154| 14.7M| sfxappnd = nullptr; 1155| 14.7M| sfxextra = 0; 1156| | 1157| | // first handle the special case of 0 length prefixes 1158| 14.7M| PfxEntry* pe = pStart[0]; 1159| | 1160| 16.0M| while (pe) { ------------------ | Branch (1160:10): [True: 1.27M, False: 14.7M] ------------------ 1161| 1.27M| rv = pe->check_twosfx(word, start, len, in_compound, needflag, scratch); 1162| 1.27M| if (rv) ------------------ | Branch (1162:9): [True: 0, False: 1.27M] ------------------ 1163| 0| return rv; 1164| 1.27M| pe = pe->getNext(); 1165| 1.27M| } 1166| | 1167| | // now handle the general case 1168| 14.7M| unsigned char sp = word[start]; 1169| 14.7M| PfxEntry* pptr = pStart[sp]; 1170| | 1171| 14.7M| while (pptr) { ------------------ | Branch (1171:10): [True: 55.6k, False: 14.7M] ------------------ 1172| 55.6k| if (isSubset(pptr->getKey(), word.c_str() + start)) { ------------------ | Branch (1172:9): [True: 35.8k, False: 19.7k] ------------------ 1173| 35.8k| rv = pptr->check_twosfx(word, start, len, in_compound, needflag, scratch); 1174| 35.8k| if (rv) { ------------------ | Branch (1174:11): [True: 0, False: 35.8k] ------------------ 1175| 0| pfx = pptr; 1176| 0| return rv; 1177| 0| } 1178| 35.8k| pptr = pptr->getNextEQ(); 1179| 35.8k| } else { 1180| 19.7k| pptr = pptr->getNextNE(); 1181| 19.7k| } 1182| 55.6k| } 1183| | 1184| 14.7M| return nullptr; 1185| 14.7M|} _ZN8AffixMgr18prefix_check_morphERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEiicR12AffixScratcht: 1193| 12.7k| const FLAG needflag) { 1194| | 1195| 12.7k| std::string result; 1196| | 1197| 12.7k| pfx = nullptr; 1198| 12.7k| sfxappnd = nullptr; 1199| 12.7k| sfxextra = 0; 1200| | 1201| | // first handle the special case of 0 length prefixes 1202| 12.7k| PfxEntry* pe = pStart[0]; 1203| 13.4k| while (pe) { ------------------ | Branch (1203:10): [True: 674, False: 12.7k] ------------------ 1204| 674| std::string st = pe->check_morph(word, start, len, in_compound, needflag, scratch); 1205| 674| if (!st.empty()) { ------------------ | Branch (1205:9): [True: 0, False: 674] ------------------ 1206| 0| result.append(st); 1207| 0| } 1208| 674| pe = pe->getNext(); 1209| 674| } 1210| | 1211| | // now handle the general case 1212| 12.7k| unsigned char sp = word[start]; 1213| 12.7k| PfxEntry* pptr = pStart[sp]; 1214| | 1215| 12.7k| while (pptr) { ------------------ | Branch (1215:10): [True: 0, False: 12.7k] ------------------ 1216| 0| if (isSubset(pptr->getKey(), word.c_str() + start)) { ------------------ | Branch (1216:9): [True: 0, False: 0] ------------------ 1217| 0| std::string st = pptr->check_morph(word, start, len, in_compound, needflag, scratch); 1218| 0| if (!st.empty()) { ------------------ | Branch (1218:11): [True: 0, False: 0] ------------------ 1219| | // fogemorpheme 1220| 0| if ((in_compound != IN_CPD_NOT) || ------------------ | | 72| 0|#define IN_CPD_NOT 0 ------------------ | Branch (1220:13): [True: 0, False: 0] ------------------ 1221| 0| !((pptr->getCont() && (TESTAFF(pptr->getCont(), onlyincompound, ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (1221:16): [True: 0, False: 0] | Branch (1221:35): [True: 0, False: 0] ------------------ 1222| 0| pptr->getContLen()))))) { 1223| 0| result.append(st); 1224| 0| pfx = pptr; 1225| 0| } 1226| 0| } 1227| 0| pptr = pptr->getNextEQ(); 1228| 0| } else { 1229| 0| pptr = pptr->getNextNE(); 1230| 0| } 1231| 0| } 1232| | 1233| 12.7k| return result; 1234| 12.7k|} _ZN8AffixMgr25prefix_check_twosfx_morphERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEiicR12AffixScratcht: 1242| 3.96k| const FLAG needflag) { 1243| 3.96k| std::string result; 1244| | 1245| 3.96k| pfx = nullptr; 1246| 3.96k| sfxappnd = nullptr; 1247| 3.96k| sfxextra = 0; 1248| | 1249| | // first handle the special case of 0 length prefixes 1250| 3.96k| PfxEntry* pe = pStart[0]; 1251| 4.64k| while (pe) { ------------------ | Branch (1251:10): [True: 674, False: 3.96k] ------------------ 1252| 674| std::string st = pe->check_twosfx_morph(word, start, len, in_compound, needflag, scratch); 1253| 674| if (!st.empty()) { ------------------ | Branch (1253:9): [True: 0, False: 674] ------------------ 1254| 0| result.append(st); 1255| 0| } 1256| 674| pe = pe->getNext(); 1257| 674| } 1258| | 1259| | // now handle the general case 1260| 3.96k| unsigned char sp = word[start]; 1261| 3.96k| PfxEntry* pptr = pStart[sp]; 1262| | 1263| 3.96k| while (pptr) { ------------------ | Branch (1263:10): [True: 0, False: 3.96k] ------------------ 1264| 0| if (isSubset(pptr->getKey(), word.c_str() + start)) { ------------------ | Branch (1264:9): [True: 0, False: 0] ------------------ 1265| 0| std::string st = pptr->check_twosfx_morph(word, start, len, in_compound, needflag, scratch); 1266| 0| if (!st.empty()) { ------------------ | Branch (1266:11): [True: 0, False: 0] ------------------ 1267| 0| result.append(st); 1268| 0| pfx = pptr; 1269| 0| } 1270| 0| pptr = pptr->getNextEQ(); 1271| 0| } else { 1272| 0| pptr = pptr->getNextNE(); 1273| 0| } 1274| 0| } 1275| | 1276| 3.96k| return result; 1277| 3.96k|} _ZN8AffixMgr12get_syllableERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEE: 1553| 376|short AffixMgr::get_syllable(const std::string& word) { 1554| 376| if (cpdmaxsyllable == 0) ------------------ | Branch (1554:7): [True: 0, False: 376] ------------------ 1555| 0| return 0; 1556| | 1557| 376| short num = 0; 1558| | 1559| 376| if (!utf8) { ------------------ | Branch (1559:7): [True: 0, False: 376] ------------------ 1560| 0| num = (short)std::count_if(word.begin(), word.end(), 1561| 0| [&](char c) { 1562| 0| return std::binary_search(cpdvowels.begin(), cpdvowels.end(), c); 1563| 0| }); 1564| 376| } else if (!cpdvowels_utf16.empty()) { ------------------ | Branch (1564:14): [True: 376, False: 0] ------------------ 1565| 376| std::vector w; 1566| 376| u8_u16(w, word); 1567| 376| num = (short)std::count_if(w.begin(), w.end(), 1568| 376| [&](w_char wc) { 1569| 376| return std::binary_search(cpdvowels_utf16.begin(), cpdvowels_utf16.end(), wc); 1570| 376| }); 1571| 376| } 1572| | 1573| 376| return num; 1574| 376|} _ZN8AffixMgr10setcminmaxEPmS0_PKcm: 1576| 19.7M|void AffixMgr::setcminmax(size_t* cmin, size_t* cmax, const char* word, size_t len) { 1577| 19.7M| if (utf8) { ------------------ | Branch (1577:7): [True: 1.58M, False: 18.1M] ------------------ 1578| 1.58M| int i; 1579| 4.72M| for (*cmin = 0, i = 0; (i < cpdmin) && *cmin < len; i++) { ------------------ | Branch (1579:28): [True: 3.16M, False: 1.56M] | Branch (1579:44): [True: 3.14M, False: 24.4k] ------------------ 1580| 3.52M| for ((*cmin)++; *cmin < len && is_utf8_cont(word[*cmin]); (*cmin)++) ------------------ | Branch (1580:23): [True: 3.47M, False: 41.8k] | Branch (1580:38): [True: 378k, False: 3.10M] ------------------ 1581| 378k| ; 1582| 3.14M| } 1583| 3.16M| for (*cmax = len, i = 0; (i < (cpdmin - 1)) && *cmax > 0; i++) { ------------------ | Branch (1583:30): [True: 1.58M, False: 1.58M] | Branch (1583:52): [True: 1.58M, False: 3.26k] ------------------ 1584| 2.54M| for ((*cmax)--; *cmax > 0 && is_utf8_cont(word[*cmax]); (*cmax)--) ------------------ | Branch (1584:23): [True: 2.52M, False: 21.2k] | Branch (1584:36): [True: 967k, False: 1.56M] ------------------ 1585| 967k| ; 1586| 1.58M| } 1587| 18.1M| } else { 1588| 18.1M| *cmin = cpdmin; 1589| 18.1M| *cmax = len - cpdmin + 1; 1590| 18.1M| } 1591| 19.7M|} _ZN8AffixMgr14compound_checkERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEssssPP6hentrySB_ccPiR12AffixScratch: 1605| 19.7M| AffixScratch& scratch) { 1606| 19.7M| short oldnumsyllable, oldnumsyllable2, oldwordnum, oldwordnum2; 1607| 19.7M| hentry *rv = nullptr, *rv_first; 1608| 19.7M| std::string st; 1609| 19.7M| char ch = '\0', affixed; 1610| 19.7M| size_t cmin, cmax; 1611| 19.7M| int striple = 0, soldi = 0, oldcmin = 0, oldcmax = 0, oldlen = 0, checkedstriple = 0; 1612| 19.7M| hentry** oldwords = words; 1613| 19.7M| size_t scpd = 0, len = word.size(); 1614| | 1615| | // protect subsequent words[wnum + 1] reads and any recursion 1616| 19.7M| if (wnum + 1 >= maxwordnum) ------------------ | Branch (1616:7): [True: 0, False: 19.7M] ------------------ 1617| 0| return nullptr; 1618| | 1619| 19.7M| int checked_prefix; 1620| | 1621| | // add a time limit to handle possible 1622| | // combinatorical explosion of the overlapping words 1623| | 1624| 19.7M| HUNSPELL_THREAD_LOCAL std::chrono::steady_clock::time_point clock_time_start; ------------------ | | 60| 19.7M|# define HUNSPELL_THREAD_LOCAL thread_local ------------------ 1625| 19.7M| HUNSPELL_THREAD_LOCAL bool timelimit_exceeded; ------------------ | | 60| 19.7M|# define HUNSPELL_THREAD_LOCAL thread_local ------------------ 1626| | 1627| | // get the current time 1628| 19.7M| std::chrono::steady_clock::time_point clock_now = std::chrono::steady_clock::now(); 1629| | 1630| 19.7M| if (wnum == 0) { ------------------ | Branch (1630:7): [True: 19.7M, False: 119] ------------------ 1631| | // set the start time 1632| 19.7M| clock_time_start = clock_now; 1633| 19.7M| timelimit_exceeded = false; 1634| 19.7M| } 1635| 119| else if (clock_now - clock_time_start > TIMELIMIT_MS) ------------------ | | 106| 119|#define TIMELIMIT_MS std::chrono::milliseconds(50) ------------------ | Branch (1635:12): [True: 0, False: 119] ------------------ 1636| 0| timelimit_exceeded = true; 1637| | 1638| 19.7M| setcminmax(&cmin, &cmax, word.c_str(), len); 1639| | 1640| 19.7M| st.assign(word); 1641| | 1642| 504M| for (size_t i = cmin; i < cmax; ++i) { ------------------ | Branch (1642:25): [True: 485M, False: 18.8M] ------------------ 1643| | // go to end of the UTF-8 character 1644| 485M| if (utf8) { ------------------ | Branch (1644:9): [True: 36.7M, False: 448M] ------------------ 1645| 46.1M| for (; is_utf8_cont(st[i]); i++) ------------------ | Branch (1645:14): [True: 9.36M, False: 36.7M] ------------------ 1646| 9.36M| ; 1647| 36.7M| if (i >= cmax) ------------------ | Branch (1647:11): [True: 203k, False: 36.5M] ------------------ 1648| 203k| return nullptr; 1649| 36.7M| } 1650| | 1651| 485M| words = oldwords; 1652| 485M| int onlycpdrule = (words) ? 1 : 0; ------------------ | Branch (1652:23): [True: 0, False: 485M] ------------------ 1653| | 1654| 485M| do { // onlycpdrule loop 1655| | 1656| 485M| oldnumsyllable = numsyllable; 1657| 485M| oldwordnum = wordnum; 1658| 485M| checked_prefix = 0; 1659| | 1660| 485M| do { // simplified checkcompoundpattern loop 1661| | 1662| 485M| if (timelimit_exceeded || ------------------ | Branch (1662:13): [True: 0, False: 485M] | Branch (1662:13): [True: 0, False: 485M] ------------------ 1663| 485M| std::chrono::steady_clock::now() - clock_time_start > TIMELIMIT_MS) { ------------------ | | 106| 485M|#define TIMELIMIT_MS std::chrono::milliseconds(50) ------------------ | Branch (1663:13): [True: 0, False: 485M] ------------------ 1664| 0| timelimit_exceeded = true; 1665| 0| return nullptr; 1666| 0| } 1667| | 1668| 485M| if (scpd > 0) { ------------------ | Branch (1668:13): [True: 0, False: 485M] ------------------ 1669| 0| for (; scpd <= checkcpdtable.size() && ------------------ | Branch (1669:18): [True: 0, False: 0] ------------------ 1670| 0| (checkcpdtable[scpd - 1].pattern3.empty() || ------------------ | Branch (1670:19): [True: 0, False: 0] ------------------ 1671| 0| i > word.size() || ------------------ | Branch (1671:19): [True: 0, False: 0] ------------------ 1672| 0| word.compare(i, checkcpdtable[scpd - 1].pattern3.size(), checkcpdtable[scpd - 1].pattern3) != 0); ------------------ | Branch (1672:19): [True: 0, False: 0] ------------------ 1673| 0| scpd++) 1674| 0| ; 1675| | 1676| 0| if (scpd > checkcpdtable.size()) ------------------ | Branch (1676:15): [True: 0, False: 0] ------------------ 1677| 0| break; // break simplified checkcompoundpattern loop 1678| 0| st.replace(i, std::string::npos, checkcpdtable[scpd - 1].pattern); 1679| 0| soldi = i; 1680| 0| i += checkcpdtable[scpd - 1].pattern.size(); 1681| 0| st.replace(i, std::string::npos, checkcpdtable[scpd - 1].pattern2); 1682| 0| st.replace(i + checkcpdtable[scpd - 1].pattern2.size(), std::string::npos, 1683| 0| word.substr(soldi + checkcpdtable[scpd - 1].pattern3.size())); 1684| | 1685| 0| oldlen = len; 1686| 0| len += checkcpdtable[scpd - 1].pattern.size() + 1687| 0| checkcpdtable[scpd - 1].pattern2.size() - 1688| 0| checkcpdtable[scpd - 1].pattern3.size(); 1689| 0| oldcmin = cmin; 1690| 0| oldcmax = cmax; 1691| 0| setcminmax(&cmin, &cmax, st.c_str(), len); 1692| | 1693| 0| cmax = len - cpdmin + 1; 1694| 0| } 1695| | 1696| 485M| if (i >= st.size()) ------------------ | Branch (1696:13): [True: 717k, False: 484M] ------------------ 1697| 717k| return nullptr; 1698| | 1699| 484M| ch = st[i]; 1700| 484M| st[i] = '\0'; 1701| | 1702| 484M| sfx = nullptr; 1703| 484M| pfx = nullptr; 1704| | 1705| | // FIRST WORD 1706| | 1707| 484M| affixed = 1; 1708| 484M| rv = lookup(st.c_str(), i); // perhaps without prefix 1709| | 1710| | // forbid dictionary stems with COMPOUNDFORBIDFLAG in 1711| | // compound words, overriding the effect of COMPOUNDPERMITFLAG 1712| 484M| if ((rv) && compoundforbidflag && ------------------ | Branch (1712:13): [True: 964k, False: 483M] | Branch (1712:21): [True: 28.1k, False: 936k] ------------------ 1713| 28.1k| TESTAFF(rv->astr, compoundforbidflag, rv->alen) && !hu_mov_rule) { ------------------ | | 99| 484M|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 28.1k] | | ------------------ ------------------ | Branch (1713:68): [True: 0, False: 0] ------------------ 1714| 0| bool would_continue = !onlycpdrule && simplifiedcpd; ------------------ | Branch (1714:35): [True: 0, False: 0] | Branch (1714:51): [True: 0, False: 0] ------------------ 1715| 0| if (!scpd && would_continue) { ------------------ | Branch (1715:17): [True: 0, False: 0] | Branch (1715:26): [True: 0, False: 0] ------------------ 1716| | // given the while conditions that continue jumps to, this situation 1717| | // never ends 1718| 0| HUNSPELL_WARNING(stderr, "break infinite loop\n"); 1719| 0| break; 1720| 0| } 1721| | 1722| 0| if (scpd > 0 && would_continue) { ------------------ | Branch (1722:17): [True: 0, False: 0] | Branch (1722:29): [True: 0, False: 0] ------------------ 1723| | // under these conditions we loop again, but the assumption above 1724| | // appears to be that cmin and cmax are the original values they 1725| | // had in the outside loop 1726| 0| cmin = oldcmin; 1727| 0| cmax = oldcmax; 1728| 0| } 1729| 0| continue; 1730| 0| } 1731| | 1732| | // search homonym with compound flag 1733| 485M| while ((rv) && !hu_mov_rule && ------------------ | Branch (1733:16): [True: 964k, False: 484M] | Branch (1733:24): [True: 964k, False: 90] ------------------ 1734| 964k| ((needaffix && TESTAFF(rv->astr, needaffix, rv->alen)) || ------------------ | | 99| 312k|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 312k] | | ------------------ ------------------ | Branch (1734:18): [True: 312k, False: 652k] ------------------ 1735| 964k| !((compoundflag && !words && !onlycpdrule && ------------------ | Branch (1735:20): [True: 964k, False: 0] | Branch (1735:36): [True: 964k, False: 0] | Branch (1735:46): [True: 964k, False: 0] ------------------ 1736| 964k| TESTAFF(rv->astr, compoundflag, rv->alen)) || ------------------ | | 99| 964k|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 194, False: 964k] | | ------------------ ------------------ 1737| 964k| (compoundbegin && !wordnum && !onlycpdrule && ------------------ | Branch (1737:20): [True: 0, False: 964k] | Branch (1737:37): [True: 0, False: 0] | Branch (1737:49): [True: 0, False: 0] ------------------ 1738| 0| TESTAFF(rv->astr, compoundbegin, rv->alen)) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1739| 964k| (compoundmiddle && wordnum && !words && !onlycpdrule && ------------------ | Branch (1739:20): [True: 0, False: 964k] | Branch (1739:38): [True: 0, False: 0] | Branch (1739:49): [True: 0, False: 0] | Branch (1739:59): [True: 0, False: 0] ------------------ 1740| 0| TESTAFF(rv->astr, compoundmiddle, rv->alen)) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1741| 964k| (!defcpdtable.empty() && onlycpdrule && ------------------ | Branch (1741:20): [True: 0, False: 964k] | Branch (1741:44): [True: 0, False: 0] ------------------ 1742| 0| ((!words && !wordnum && ------------------ | Branch (1742:22): [True: 0, False: 0] | Branch (1742:32): [True: 0, False: 0] ------------------ 1743| 0| defcpd_check(&words, wnum, maxwordnum, rv, rwords, 0)) || ------------------ | Branch (1743:22): [True: 0, False: 0] ------------------ 1744| 0| (words && ------------------ | Branch (1744:22): [True: 0, False: 0] ------------------ 1745| 0| defcpd_check(&words, wnum, maxwordnum, rv, rwords, 0))))) || ------------------ | Branch (1745:22): [True: 0, False: 0] ------------------ 1746| 194| (scpd != 0 && checkcpdtable[scpd - 1].cond != FLAG_NULL && ------------------ | | 96| 194|#define FLAG_NULL 0x00 ------------------ | Branch (1746:18): [True: 0, False: 194] | Branch (1746:31): [True: 0, False: 0] ------------------ 1747| 964k| !TESTAFF(rv->astr, checkcpdtable[scpd - 1].cond, rv->alen)))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (1747:18): [True: 0, False: 0] ------------------ 1748| 964k| rv = rv->next_homonym; 1749| 964k| } 1750| | 1751| 484M| if (rv) ------------------ | Branch (1751:13): [True: 284, False: 484M] ------------------ 1752| 284| affixed = 0; 1753| | 1754| 484M| if (!rv) { ------------------ | Branch (1754:13): [True: 484M, False: 284] ------------------ 1755| 484M| if (onlycpdrule) ------------------ | Branch (1755:15): [True: 0, False: 484M] ------------------ 1756| 0| break; 1757| 484M| if (compoundflag && ------------------ | Branch (1757:15): [True: 484M, False: 0] ------------------ 1758| 484M| !(rv = prefix_check(st, 0, i, ------------------ | Branch (1758:15): [True: 484M, False: 0] ------------------ 1759| 484M| hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN, ------------------ | | 75| 114k|#define IN_CPD_OTHER 3 ------------------ hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN, ------------------ | | 73| 968M|#define IN_CPD_BEGIN 1 ------------------ | Branch (1759:35): [True: 114k, False: 484M] ------------------ 1760| 484M| scratch, compoundflag))) { 1761| 484M| if (((rv = suffix_check(st, 0, i, 0, nullptr, scratch, FLAG_NULL, compoundflag, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || ------------------ | | 96| 484M|#define FLAG_NULL 0x00 ------------------ if (((rv = suffix_check(st, 0, i, 0, nullptr, scratch, FLAG_NULL, compoundflag, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || ------------------ | | 75| 114k|#define IN_CPD_OTHER 3 ------------------ if (((rv = suffix_check(st, 0, i, 0, nullptr, scratch, FLAG_NULL, compoundflag, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || ------------------ | | 73| 968M|#define IN_CPD_BEGIN 1 ------------------ | Branch (1761:18): [True: 0, False: 484M] | Branch (1761:93): [True: 114k, False: 484M] ------------------ 1762| 484M| (compoundmoresuffixes && (rv = suffix_check_twosfx(st, 0, i, 0, nullptr, scratch, compoundflag)))) && ------------------ | Branch (1762:19): [True: 0, False: 484M] | Branch (1762:43): [True: 0, False: 0] ------------------ 1763| 0| !hu_mov_rule && sfx->getCont() && ------------------ | Branch (1763:17): [True: 0, False: 0] | Branch (1763:33): [True: 0, False: 0] ------------------ 1764| 0| ((compoundforbidflag && TESTAFF(sfx->getCont(), compoundforbidflag, sfx->getContLen())) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (1764:19): [True: 0, False: 0] ------------------ 1765| 0| (compoundend && TESTAFF(sfx->getCont(), compoundend, sfx->getContLen())))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (1765:19): [True: 0, False: 0] ------------------ 1766| 0| rv = nullptr; 1767| 0| } 1768| 484M| } 1769| | 1770| 484M| if (rv || ------------------ | Branch (1770:15): [True: 0, False: 484M] ------------------ 1771| 484M| (((wordnum == 0) && compoundbegin && ------------------ | Branch (1771:17): [True: 484M, False: 115k] | Branch (1771:35): [True: 0, False: 484M] ------------------ 1772| 0| ((rv = suffix_check(st, 0, i, 0, nullptr, scratch, FLAG_NULL, compoundbegin, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || ------------------ | | 96| 0|#define FLAG_NULL 0x00 ------------------ ((rv = suffix_check(st, 0, i, 0, nullptr, scratch, FLAG_NULL, compoundbegin, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || ------------------ | | 75| 0|#define IN_CPD_OTHER 3 ------------------ ((rv = suffix_check(st, 0, i, 0, nullptr, scratch, FLAG_NULL, compoundbegin, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || ------------------ | | 73| 0|#define IN_CPD_BEGIN 1 ------------------ | Branch (1772:18): [True: 0, False: 0] | Branch (1772:94): [True: 0, False: 0] ------------------ 1773| 0| (compoundmoresuffixes && (rv = suffix_check_twosfx(st, 0, i, 0, nullptr, scratch, ------------------ | Branch (1773:19): [True: 0, False: 0] | Branch (1773:43): [True: 0, False: 0] ------------------ 1774| 0| compoundbegin))) || // twofold suffixes + compound 1775| 0| (rv = prefix_check(st, 0, i, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN, scratch, compoundbegin)))) || ------------------ | | 75| 0|#define IN_CPD_OTHER 3 ------------------ (rv = prefix_check(st, 0, i, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN, scratch, compoundbegin)))) || ------------------ | | 73| 0|#define IN_CPD_BEGIN 1 ------------------ | Branch (1775:18): [True: 0, False: 0] | Branch (1775:47): [True: 0, False: 0] ------------------ 1776| 484M| ((wordnum > 0) && compoundmiddle && ------------------ | Branch (1776:17): [True: 503, False: 484M] | Branch (1776:34): [True: 0, False: 503] ------------------ 1777| 0| ((rv = suffix_check(st, 0, i, 0, nullptr, scratch, FLAG_NULL, compoundmiddle, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || ------------------ | | 96| 0|#define FLAG_NULL 0x00 ------------------ ((rv = suffix_check(st, 0, i, 0, nullptr, scratch, FLAG_NULL, compoundmiddle, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || ------------------ | | 75| 0|#define IN_CPD_OTHER 3 ------------------ ((rv = suffix_check(st, 0, i, 0, nullptr, scratch, FLAG_NULL, compoundmiddle, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || ------------------ | | 73| 0|#define IN_CPD_BEGIN 1 ------------------ | Branch (1777:18): [True: 0, False: 0] | Branch (1777:95): [True: 0, False: 0] ------------------ 1778| 0| (compoundmoresuffixes && (rv = suffix_check_twosfx(st, 0, i, 0, nullptr, scratch, ------------------ | Branch (1778:19): [True: 0, False: 0] | Branch (1778:43): [True: 0, False: 0] ------------------ 1779| 0| compoundmiddle))) || // twofold suffixes + compound 1780| 0| (rv = prefix_check(st, 0, i, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN, scratch, compoundmiddle)))))) ------------------ | | 75| 0|#define IN_CPD_OTHER 3 ------------------ (rv = prefix_check(st, 0, i, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN, scratch, compoundmiddle)))))) ------------------ | | 73| 0|#define IN_CPD_BEGIN 1 ------------------ | Branch (1780:18): [True: 0, False: 0] | Branch (1780:47): [True: 0, False: 0] ------------------ 1781| 0| checked_prefix = 1; 1782| | // else check forbiddenwords and needaffix 1783| 484M| } else if (rv->astr && (TESTAFF(rv->astr, forbiddenword, rv->alen) || ------------------ | | 99| 440|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 220] | | ------------------ ------------------ | Branch (1783:20): [True: 220, False: 64] | Branch (1783:20): [True: 1, False: 283] ------------------ 1784| 220| TESTAFF(rv->astr, needaffix, rv->alen) || ------------------ | | 99| 440|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 220] | | ------------------ ------------------ 1785| 220| TESTAFF(rv->astr, ONLYUPCASEFLAG, rv->alen) || ------------------ | | 99| 440|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 1, False: 219] | | ------------------ ------------------ 1786| 219| (is_sug && nosuggest && ------------------ | Branch (1786:34): [True: 0, False: 219] | Branch (1786:44): [True: 0, False: 0] ------------------ 1787| 1| TESTAFF(rv->astr, nosuggest, rv->alen)))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1788| 1| st[i] = ch; 1789| | // continue; 1790| 1| break; 1791| 1| } 1792| | 1793| | // check non_compound flag in suffix and prefix 1794| 484M| if ((rv) && !hu_mov_rule && ------------------ | Branch (1794:13): [True: 283, False: 484M] | Branch (1794:21): [True: 194, False: 89] ------------------ 1795| 194| ((pfx && pfx->getCont() && ------------------ | Branch (1795:15): [True: 0, False: 194] | Branch (1795:22): [True: 0, False: 0] ------------------ 1796| 0| TESTAFF(pfx->getCont(), compoundforbidflag, pfx->getContLen())) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1797| 194| (sfx && sfx->getCont() && ------------------ | Branch (1797:15): [True: 0, False: 194] | Branch (1797:22): [True: 0, False: 0] ------------------ 1798| 0| TESTAFF(sfx->getCont(), compoundforbidflag, ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1799| 194| sfx->getContLen())))) { 1800| 0| rv = nullptr; 1801| 0| } 1802| | 1803| | // check compoundend flag in suffix and prefix 1804| 484M| if ((rv) && !checked_prefix && compoundend && !hu_mov_rule && ------------------ | Branch (1804:13): [True: 283, False: 484M] | Branch (1804:21): [True: 283, False: 0] | Branch (1804:40): [True: 0, False: 283] | Branch (1804:55): [True: 0, False: 0] ------------------ 1805| 0| ((pfx && pfx->getCont() && ------------------ | Branch (1805:15): [True: 0, False: 0] | Branch (1805:22): [True: 0, False: 0] ------------------ 1806| 0| TESTAFF(pfx->getCont(), compoundend, pfx->getContLen())) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1807| 0| (sfx && sfx->getCont() && ------------------ | Branch (1807:15): [True: 0, False: 0] | Branch (1807:22): [True: 0, False: 0] ------------------ 1808| 0| TESTAFF(sfx->getCont(), compoundend, sfx->getContLen())))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1809| 0| rv = nullptr; 1810| 0| } 1811| | 1812| | // check compoundmiddle flag in suffix and prefix 1813| 484M| if ((rv) && !checked_prefix && (wordnum == 0) && compoundmiddle && ------------------ | Branch (1813:13): [True: 283, False: 484M] | Branch (1813:21): [True: 283, False: 0] | Branch (1813:40): [True: 194, False: 89] | Branch (1813:58): [True: 0, False: 194] ------------------ 1814| 0| !hu_mov_rule && ------------------ | Branch (1814:13): [True: 0, False: 0] ------------------ 1815| 0| ((pfx && pfx->getCont() && ------------------ | Branch (1815:15): [True: 0, False: 0] | Branch (1815:22): [True: 0, False: 0] ------------------ 1816| 0| TESTAFF(pfx->getCont(), compoundmiddle, pfx->getContLen())) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1817| 0| (sfx && sfx->getCont() && ------------------ | Branch (1817:15): [True: 0, False: 0] | Branch (1817:22): [True: 0, False: 0] ------------------ 1818| 0| TESTAFF(sfx->getCont(), compoundmiddle, sfx->getContLen())))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1819| 0| rv = nullptr; 1820| 0| } 1821| | 1822| | // check forbiddenwords 1823| 484M| if ((rv) && (rv->astr) && ------------------ | Branch (1823:13): [True: 283, False: 484M] | Branch (1823:13): [True: 0, False: 484M] | Branch (1823:21): [True: 219, False: 64] ------------------ 1824| 219| (TESTAFF(rv->astr, forbiddenword, rv->alen) || ------------------ | | 99| 438|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 219] | | ------------------ ------------------ 1825| 219| TESTAFF(rv->astr, ONLYUPCASEFLAG, rv->alen) || ------------------ | | 99| 438|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 219] | | ------------------ ------------------ 1826| 219| (is_sug && nosuggest && TESTAFF(rv->astr, nosuggest, rv->alen)))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (1826:15): [True: 0, False: 219] | Branch (1826:25): [True: 0, False: 0] ------------------ 1827| 0| return nullptr; 1828| 0| } 1829| | 1830| | // increment word number, if the second root has a compoundroot flag 1831| 484M| if ((rv) && compoundroot && ------------------ | Branch (1831:13): [True: 283, False: 484M] | Branch (1831:21): [True: 0, False: 283] ------------------ 1832| 0| (TESTAFF(rv->astr, compoundroot, rv->alen))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (1832:13): [True: 0, False: 0] ------------------ 1833| 0| wordnum++; 1834| 0| } 1835| | 1836| | // first word is acceptable in compound words? 1837| 484M| if (((rv) && ------------------ | Branch (1837:13): [True: 216, False: 484M] | Branch (1837:14): [True: 283, False: 484M] ------------------ 1838| 283| (checked_prefix || (words && words[wnum]) || ------------------ | Branch (1838:15): [True: 0, False: 283] | Branch (1838:34): [True: 0, False: 283] | Branch (1838:43): [True: 0, False: 0] ------------------ 1839| 283| (compoundflag && TESTAFF(rv->astr, compoundflag, rv->alen)) || ------------------ | | 99| 283|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 216, False: 67] | | ------------------ ------------------ | Branch (1839:16): [True: 283, False: 0] ------------------ 1840| 67| ((oldwordnum == 0) && compoundbegin && ------------------ | Branch (1840:16): [True: 0, False: 67] | Branch (1840:37): [True: 0, False: 0] ------------------ 1841| 0| TESTAFF(rv->astr, compoundbegin, rv->alen)) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1842| 67| ((oldwordnum > 0) && compoundmiddle && ------------------ | Branch (1842:16): [True: 0, False: 67] | Branch (1842:36): [True: 0, False: 0] ------------------ 1843| 0| TESTAFF(rv->astr, compoundmiddle, rv->alen)) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1844| | 1845| | // LANG_hu section: spec. Hungarian rule 1846| 67| || ((langnum == LANG_hu) && hu_mov_rule && ------------------ | Branch (1846:19): [True: 67, False: 0] | Branch (1846:43): [True: 67, False: 0] ------------------ 1847| 67| (TESTAFF( ------------------ | | 99| 134|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 67] | | ------------------ ------------------ 1848| 67| rv->astr, 'F', 1849| 67| rv->alen) || // XXX hardwired Hungarian dictionary codes 1850| 67| TESTAFF(rv->astr, 'G', rv->alen) || ------------------ | | 99| 134|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 67] | | ------------------ ------------------ 1851| 67| TESTAFF(rv->astr, 'H', rv->alen))) ------------------ | | 99| 67|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 67] | | ------------------ ------------------ 1852| | // END of LANG_hu section 1853| 283| ) && 1854| 216| ( 1855| | // test CHECKCOMPOUNDPATTERN conditions 1856| 216| scpd == 0 || checkcpdtable[scpd - 1].cond == FLAG_NULL || ------------------ | | 96| 216|#define FLAG_NULL 0x00 ------------------ | Branch (1856:18): [True: 216, False: 0] | Branch (1856:31): [True: 0, False: 0] ------------------ 1857| 0| TESTAFF(rv->astr, checkcpdtable[scpd - 1].cond, rv->alen)) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1858| 216| !((checkcompoundtriple && scpd == 0 && ------------------ | Branch (1858:17): [True: 0, False: 216] | Branch (1858:40): [True: 0, False: 0] ------------------ 1859| 0| !words && i < word.size() && // test triple letters ------------------ | Branch (1859:17): [True: 0, False: 0] | Branch (1859:27): [True: 0, False: 0] ------------------ 1860| 0| (word[i - 1] == word[i]) && ------------------ | Branch (1860:17): [True: 0, False: 0] ------------------ 1861| 0| (((i > 1) && (word[i - 1] == word[i - 2])) || ------------------ | Branch (1861:19): [True: 0, False: 0] | Branch (1861:30): [True: 0, False: 0] ------------------ 1862| 0| ((word[i - 1] == word[i + 1])) // may be word[i+1] == '\0' ------------------ | Branch (1862:18): [True: 0, False: 0] ------------------ 1863| 0| )) || 1864| 216| (checkcompoundcase && scpd == 0 && !words && i < word.size() && ------------------ | Branch (1864:17): [True: 0, False: 216] | Branch (1864:38): [True: 0, False: 0] | Branch (1864:51): [True: 0, False: 0] | Branch (1864:61): [True: 0, False: 0] ------------------ 1865| 0| cpdcase_check(word, i)))) ------------------ | Branch (1865:17): [True: 0, False: 0] ------------------ 1866| | // LANG_hu section: spec. Hungarian rule 1867| 484M| || ((!rv) && (langnum == LANG_hu) && hu_mov_rule && ------------------ | Branch (1867:17): [True: 484M, False: 67] | Branch (1867:26): [True: 36.5M, False: 447M] | Branch (1867:50): [True: 114k, False: 36.4M] ------------------ 1868| 114k| (rv = affix_check(st, 0, i, scratch)) && ------------------ | Branch (1868:17): [True: 0, False: 114k] ------------------ 1869| 0| (sfx && sfx->getCont() && ------------------ | Branch (1869:18): [True: 0, False: 0] | Branch (1869:25): [True: 0, False: 0] ------------------ 1870| 0| ( // XXX hardwired Hungarian dic. codes 1871| 0| TESTAFF(sfx->getCont(), (unsigned short)'x', ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1872| 0| sfx->getContLen()) || 1873| 0| TESTAFF( ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1874| 0| sfx->getCont(), (unsigned short)'%', 1875| 216| sfx->getContLen()))))) { // first word is ok condition 1876| | 1877| | // LANG_hu section: spec. Hungarian rule 1878| 216| if (langnum == LANG_hu) { ------------------ | Branch (1878:15): [True: 188, False: 28] ------------------ 1879| | // calculate syllable number of the word 1880| 188| numsyllable += get_syllable(st.substr(0, i)); 1881| | // + 1 word, if syllable number of the prefix > 1 (hungarian 1882| | // convention) 1883| 188| if (pfx && (get_syllable(pfx->getKey()) > 1)) ------------------ | Branch (1883:17): [True: 0, False: 188] | Branch (1883:17): [True: 0, False: 188] | Branch (1883:24): [True: 0, False: 0] ------------------ 1884| 0| wordnum++; 1885| 188| } 1886| | // END of LANG_hu section 1887| | 1888| | // NEXT WORD(S) 1889| 216| rv_first = rv; 1890| 216| st[i] = ch; 1891| | 1892| 216| do { // striple loop 1893| | 1894| | // check simplifiedtriple 1895| 216| if (simplifiedtriple) { ------------------ | Branch (1895:17): [True: 0, False: 216] ------------------ 1896| 0| if (striple) { ------------------ | Branch (1896:19): [True: 0, False: 0] ------------------ 1897| 0| checkedstriple = 1; 1898| 0| i--; // check "fahrt" instead of "ahrt" in "Schiffahrt" 1899| 0| } else if (i > 2 && i <= word.size() && word[i - 1] == word[i - 2]) ------------------ | Branch (1899:26): [True: 0, False: 0] | Branch (1899:35): [True: 0, False: 0] | Branch (1899:55): [True: 0, False: 0] ------------------ 1900| 0| striple = 1; 1901| 0| } 1902| | 1903| 216| rv = lookup(st.c_str() + i, st.size() - i); // perhaps without prefix 1904| | 1905| | // search homonym with compound flag 1906| 216| while ((rv) && ((needaffix && TESTAFF(rv->astr, needaffix, rv->alen)) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (1906:20): [True: 0, False: 216] | Branch (1906:30): [True: 0, False: 0] ------------------ 1907| 0| !((compoundflag && !words && TESTAFF(rv->astr, compoundflag, rv->alen)) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (1907:32): [True: 0, False: 0] | Branch (1907:48): [True: 0, False: 0] ------------------ 1908| 0| (compoundend && !words && TESTAFF(rv->astr, compoundend, rv->alen)) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (1908:32): [True: 0, False: 0] | Branch (1908:47): [True: 0, False: 0] ------------------ 1909| 0| (!defcpdtable.empty() && words && defcpd_check(&words, wnum + 1, maxwordnum, rv, nullptr, 1))) || ------------------ | Branch (1909:32): [True: 0, False: 0] | Branch (1909:56): [True: 0, False: 0] | Branch (1909:65): [True: 0, False: 0] ------------------ 1910| 0| (scpd != 0 && checkcpdtable[scpd - 1].cond2 != FLAG_NULL && ------------------ | | 96| 0|#define FLAG_NULL 0x00 ------------------ | Branch (1910:30): [True: 0, False: 0] | Branch (1910:43): [True: 0, False: 0] ------------------ 1911| 0| !TESTAFF(rv->astr, checkcpdtable[scpd - 1].cond2, rv->alen)))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (1911:30): [True: 0, False: 0] ------------------ 1912| 0| rv = rv->next_homonym; 1913| 0| } 1914| | 1915| | // check FORCEUCASE 1916| 216| if (rv && forceucase && ------------------ | Branch (1916:17): [True: 0, False: 216] | Branch (1916:23): [True: 0, False: 0] ------------------ 1917| 0| (TESTAFF(rv->astr, forceucase, rv->alen)) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (1917:17): [True: 0, False: 0] ------------------ 1918| 0| !(info && *info & SPELL_ORIGCAP)) ------------------ | | 83| 0|#define SPELL_ORIGCAP (1 << 5) ------------------ | Branch (1918:19): [True: 0, False: 0] | Branch (1918:27): [True: 0, False: 0] ------------------ 1919| 0| rv = nullptr; 1920| | 1921| 216| if (rv && words && words[wnum + 1]) ------------------ | Branch (1921:17): [True: 0, False: 216] | Branch (1921:23): [True: 0, False: 0] | Branch (1921:32): [True: 0, False: 0] ------------------ 1922| 0| return rv_first; 1923| | 1924| 216| oldnumsyllable2 = numsyllable; 1925| 216| oldwordnum2 = wordnum; 1926| | 1927| | // LANG_hu section: spec. Hungarian rule, XXX hardwired dictionary 1928| | // code 1929| 216| if ((rv) && (langnum == LANG_hu) && ------------------ | Branch (1929:17): [True: 0, False: 216] | Branch (1929:17): [True: 0, False: 216] | Branch (1929:25): [True: 0, False: 0] ------------------ 1930| 0| (TESTAFF(rv->astr, 'I', rv->alen)) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (1930:17): [True: 0, False: 0] ------------------ 1931| 0| !(TESTAFF(rv->astr, 'J', rv->alen))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (1931:17): [True: 0, False: 0] ------------------ 1932| 0| numsyllable--; 1933| 0| } 1934| | // END of LANG_hu section 1935| | 1936| | // increment word number, if the second root has a compoundroot flag 1937| 216| if ((rv) && (compoundroot) && ------------------ | Branch (1937:17): [True: 0, False: 216] | Branch (1937:25): [True: 0, False: 0] ------------------ 1938| 0| (TESTAFF(rv->astr, compoundroot, rv->alen))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (1938:17): [True: 0, False: 0] ------------------ 1939| 0| wordnum++; 1940| 0| } 1941| | 1942| | // check forbiddenwords 1943| 216| if ((rv) && (rv->astr) && ------------------ | Branch (1943:17): [True: 0, False: 216] | Branch (1943:17): [True: 0, False: 216] | Branch (1943:25): [True: 0, False: 0] ------------------ 1944| 0| (TESTAFF(rv->astr, forbiddenword, rv->alen) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1945| 0| TESTAFF(rv->astr, ONLYUPCASEFLAG, rv->alen) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1946| 0| (is_sug && nosuggest && ------------------ | Branch (1946:19): [True: 0, False: 0] | Branch (1946:29): [True: 0, False: 0] ------------------ 1947| 0| TESTAFF(rv->astr, nosuggest, rv->alen)))) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1948| 0| return nullptr; 1949| | 1950| | // second word is acceptable, as a root? 1951| | // hungarian conventions: compounding is acceptable, 1952| | // when compound forms consist of 2 words, or if more, 1953| | // then the syllable number of root words must be 6, or lesser. 1954| | 1955| 216| if ((rv) && ------------------ | Branch (1955:17): [True: 0, False: 216] | Branch (1955:17): [True: 0, False: 216] ------------------ 1956| 0| ((compoundflag && TESTAFF(rv->astr, compoundflag, rv->alen)) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (1956:19): [True: 0, False: 0] ------------------ 1957| 0| (compoundend && TESTAFF(rv->astr, compoundend, rv->alen))) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (1957:19): [True: 0, False: 0] ------------------ 1958| 0| (((cpdwordmax == -1) || (wordnum + 1 < cpdwordmax)) || ------------------ | Branch (1958:19): [True: 0, False: 0] | Branch (1958:41): [True: 0, False: 0] ------------------ 1959| 0| ((cpdmaxsyllable != 0) && ------------------ | Branch (1959:19): [True: 0, False: 0] ------------------ 1960| 0| (numsyllable + get_syllable(std::string(HENTRY_WORD(rv), rv->blen)) <= ------------------ | | 54| 0|#define HENTRY_WORD(h) &(h->word[0]) ------------------ | Branch (1960:19): [True: 0, False: 0] ------------------ 1961| 0| cpdmaxsyllable))) && 1962| 0| ( 1963| | // test CHECKCOMPOUNDPATTERN 1964| 0| checkcpdtable.empty() || scpd != 0 || ------------------ | Branch (1964:21): [True: 0, False: 0] | Branch (1964:46): [True: 0, False: 0] ------------------ 1965| 0| (i < word.size() && !cpdpat_check(word, i, rv_first, rv, 0))) && ------------------ | Branch (1965:22): [True: 0, False: 0] | Branch (1965:41): [True: 0, False: 0] ------------------ 1966| 0| ((!checkcompounddup || (rv != rv_first))) ------------------ | Branch (1966:19): [True: 0, False: 0] | Branch (1966:40): [True: 0, False: 0] ------------------ 1967| | // test CHECKCOMPOUNDPATTERN conditions 1968| 0| && 1969| 0| (scpd == 0 || checkcpdtable[scpd - 1].cond2 == FLAG_NULL || ------------------ | | 96| 0|#define FLAG_NULL 0x00 ------------------ | Branch (1969:18): [True: 0, False: 0] | Branch (1969:31): [True: 0, False: 0] ------------------ 1970| 0| TESTAFF(rv->astr, checkcpdtable[scpd - 1].cond2, rv->alen))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1971| | // forbid compound word, if it is a non-compound word with typical 1972| | // fault 1973| 0| if ((checkcompoundrep && cpdrep_check(word, len, scratch, timelimit_exceeded, clock_time_start)) || ------------------ | Branch (1973:20): [True: 0, False: 0] | Branch (1973:40): [True: 0, False: 0] ------------------ 1974| 0| cpdwordpair_check(word, len, scratch, timelimit_exceeded, clock_time_start)) ------------------ | Branch (1974:19): [True: 0, False: 0] ------------------ 1975| 0| return nullptr; 1976| 0| return rv_first; 1977| 0| } 1978| | 1979| 216| numsyllable = oldnumsyllable2; 1980| 216| wordnum = oldwordnum2; 1981| | 1982| | // perhaps second word has prefix or/and suffix 1983| 216| sfx = nullptr; 1984| 216| sfxflag = FLAG_NULL; ------------------ | | 96| 216|#define FLAG_NULL 0x00 ------------------ 1985| 216| rv = (compoundflag && !onlycpdrule && i < word.size()) ? affix_check(word, i, word.size() - i, scratch, compoundflag, IN_CPD_END) ------------------ | | 74| 216|#define IN_CPD_END 2 ------------------ | Branch (1985:19): [True: 216, False: 0] | Branch (1985:35): [True: 216, False: 0] | Branch (1985:51): [True: 216, False: 0] ------------------ 1986| 216| : nullptr; 1987| 216| if (!rv && compoundend && !onlycpdrule) { ------------------ | Branch (1987:17): [True: 216, False: 0] | Branch (1987:24): [True: 0, False: 216] | Branch (1987:39): [True: 0, False: 0] ------------------ 1988| 0| sfx = nullptr; 1989| 0| pfx = nullptr; 1990| 0| if (i < word.size()) ------------------ | Branch (1990:19): [True: 0, False: 0] ------------------ 1991| 0| rv = affix_check(word, i, word.size() - i, scratch, compoundend, IN_CPD_END); ------------------ | | 74| 0|#define IN_CPD_END 2 ------------------ 1992| 0| } 1993| | 1994| 216| if (!rv && !defcpdtable.empty() && words) { ------------------ | Branch (1994:17): [True: 216, False: 0] | Branch (1994:24): [True: 0, False: 216] | Branch (1994:48): [True: 0, False: 0] ------------------ 1995| 0| if (i < word.size()) ------------------ | Branch (1995:19): [True: 0, False: 0] ------------------ 1996| 0| rv = affix_check(word, i, word.size() - i, scratch, 0, IN_CPD_END); ------------------ | | 74| 0|#define IN_CPD_END 2 ------------------ 1997| 0| if (rv && defcpd_check(&words, wnum + 1, maxwordnum, rv, nullptr, 1)) ------------------ | Branch (1997:19): [True: 0, False: 0] | Branch (1997:25): [True: 0, False: 0] ------------------ 1998| 0| return rv_first; 1999| 0| rv = nullptr; 2000| 0| } 2001| | 2002| | // test CHECKCOMPOUNDPATTERN conditions (allowed forms) 2003| 216| if (rv && ------------------ | Branch (2003:17): [True: 0, False: 216] ------------------ 2004| 0| !(scpd == 0 || checkcpdtable[scpd - 1].cond2 == FLAG_NULL || ------------------ | | 96| 0|#define FLAG_NULL 0x00 ------------------ | Branch (2004:19): [True: 0, False: 0] | Branch (2004:32): [True: 0, False: 0] ------------------ 2005| 0| TESTAFF(rv->astr, checkcpdtable[scpd - 1].cond2, rv->alen))) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2006| 0| rv = nullptr; 2007| | 2008| | // test CHECKCOMPOUNDPATTERN conditions (forbidden compounds) 2009| 216| if (rv && !checkcpdtable.empty() && scpd == 0 && ------------------ | Branch (2009:17): [True: 0, False: 216] | Branch (2009:23): [True: 0, False: 0] | Branch (2009:49): [True: 0, False: 0] ------------------ 2010| 0| cpdpat_check(word, i, rv_first, rv, affixed)) ------------------ | Branch (2010:17): [True: 0, False: 0] ------------------ 2011| 0| rv = nullptr; 2012| | 2013| | // check non_compound flag in suffix and prefix 2014| 216| if ((rv) && ((pfx && pfx->getCont() && ------------------ | Branch (2014:17): [True: 0, False: 216] | Branch (2014:27): [True: 0, False: 0] | Branch (2014:34): [True: 0, False: 0] ------------------ 2015| 0| TESTAFF(pfx->getCont(), compoundforbidflag, ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2016| 0| pfx->getContLen())) || 2017| 0| (sfx && sfx->getCont() && ------------------ | Branch (2017:27): [True: 0, False: 0] | Branch (2017:34): [True: 0, False: 0] ------------------ 2018| 0| TESTAFF(sfx->getCont(), compoundforbidflag, ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2019| 0| sfx->getContLen())))) { 2020| 0| rv = nullptr; 2021| 0| } 2022| | 2023| | // check FORCEUCASE 2024| 216| if (rv && forceucase && ------------------ | Branch (2024:17): [True: 0, False: 216] | Branch (2024:23): [True: 0, False: 0] ------------------ 2025| 0| (TESTAFF(rv->astr, forceucase, rv->alen)) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (2025:17): [True: 0, False: 0] ------------------ 2026| 0| !(info && *info & SPELL_ORIGCAP)) ------------------ | | 83| 0|#define SPELL_ORIGCAP (1 << 5) ------------------ | Branch (2026:19): [True: 0, False: 0] | Branch (2026:27): [True: 0, False: 0] ------------------ 2027| 0| rv = nullptr; 2028| | 2029| | // check forbiddenwords 2030| 216| if ((rv) && (rv->astr) && ------------------ | Branch (2030:17): [True: 0, False: 216] | Branch (2030:17): [True: 0, False: 216] | Branch (2030:25): [True: 0, False: 0] ------------------ 2031| 0| (TESTAFF(rv->astr, forbiddenword, rv->alen) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2032| 0| TESTAFF(rv->astr, ONLYUPCASEFLAG, rv->alen) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2033| 0| (is_sug && nosuggest && ------------------ | Branch (2033:19): [True: 0, False: 0] | Branch (2033:29): [True: 0, False: 0] ------------------ 2034| 0| TESTAFF(rv->astr, nosuggest, rv->alen)))) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2035| 0| return nullptr; 2036| | 2037| | // pfxappnd = prefix of word+i, or NULL 2038| | // calculate syllable number of prefix. 2039| | // hungarian convention: when syllable number of prefix is more, 2040| | // than 1, the prefix+word counts as two words. 2041| | 2042| 216| if (langnum == LANG_hu) { ------------------ | Branch (2042:17): [True: 188, False: 28] ------------------ 2043| 188| if (i < word.size()) { ------------------ | Branch (2043:19): [True: 188, False: 0] ------------------ 2044| | // calculate syllable number of the word 2045| 188| numsyllable += get_syllable(word.substr(i)); 2046| 188| } 2047| | 2048| | // - affix syllable num. 2049| | // XXX only second suffix (inflections, not derivations) 2050| 188| if (sfxappnd) { ------------------ | Branch (2050:19): [True: 0, False: 188] ------------------ 2051| 0| std::string tmp(sfxappnd); 2052| 0| reverseword(tmp); 2053| 0| numsyllable -= short(get_syllable(tmp) + sfxextra); 2054| 188| } else { 2055| 188| numsyllable -= short(sfxextra); 2056| 188| } 2057| | 2058| | // + 1 word, if syllable number of the prefix > 1 (hungarian 2059| | // convention) 2060| 188| if (pfx && (get_syllable(pfx->getKey()) > 1)) ------------------ | Branch (2060:19): [True: 0, False: 188] | Branch (2060:19): [True: 0, False: 188] | Branch (2060:26): [True: 0, False: 0] ------------------ 2061| 0| wordnum++; 2062| | 2063| | // increment syllable num, if last word has a SYLLABLENUM flag 2064| | // and the suffix is beginning `s' 2065| | 2066| 188| if (!cpdsyllablenum.empty()) { ------------------ | Branch (2066:19): [True: 0, False: 188] ------------------ 2067| 0| switch (sfxflag) { ------------------ | Branch (2067:25): [True: 0, False: 0] ------------------ 2068| 0| case 'c': { ------------------ | Branch (2068:19): [True: 0, False: 0] ------------------ 2069| 0| numsyllable += 2; 2070| 0| break; 2071| 0| } 2072| 0| case 'J': { ------------------ | Branch (2072:19): [True: 0, False: 0] ------------------ 2073| 0| numsyllable += 1; 2074| 0| break; 2075| 0| } 2076| 0| case 'I': { ------------------ | Branch (2076:19): [True: 0, False: 0] ------------------ 2077| 0| if (rv && TESTAFF(rv->astr, 'J', rv->alen)) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (2077:25): [True: 0, False: 0] | Branch (2077:25): [True: 0, False: 0] ------------------ 2078| 0| numsyllable += 1; 2079| 0| break; 2080| 0| } 2081| 0| } 2082| 0| } 2083| 188| } 2084| | 2085| | // increment word number, if the second word has a compoundroot flag 2086| 216| if ((rv) && (compoundroot) && ------------------ | Branch (2086:17): [True: 0, False: 216] | Branch (2086:25): [True: 0, False: 0] ------------------ 2087| 0| (TESTAFF(rv->astr, compoundroot, rv->alen))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (2087:17): [True: 0, False: 0] ------------------ 2088| 0| wordnum++; 2089| 0| } 2090| | // second word is acceptable, as a word with prefix or/and suffix? 2091| | // hungarian conventions: compounding is acceptable, 2092| | // when compound forms consist 2 word, otherwise 2093| | // the syllable number of root words is 6, or lesser. 2094| 216| if ((rv) && ------------------ | Branch (2094:17): [True: 0, False: 216] ------------------ 2095| 0| (((cpdwordmax == -1) || (wordnum + 1 < cpdwordmax)) || ------------------ | Branch (2095:19): [True: 0, False: 0] | Branch (2095:41): [True: 0, False: 0] ------------------ 2096| 0| ((cpdmaxsyllable != 0) && (numsyllable <= cpdmaxsyllable))) && ------------------ | Branch (2096:19): [True: 0, False: 0] | Branch (2096:44): [True: 0, False: 0] ------------------ 2097| 0| ((!checkcompounddup || (rv != rv_first)))) { ------------------ | Branch (2097:19): [True: 0, False: 0] | Branch (2097:40): [True: 0, False: 0] ------------------ 2098| | // forbid compound word, if it is a non-compound word with typical 2099| | // fault 2100| 0| if ((checkcompoundrep && cpdrep_check(word, len, scratch, timelimit_exceeded, clock_time_start)) || ------------------ | Branch (2100:20): [True: 0, False: 0] | Branch (2100:40): [True: 0, False: 0] ------------------ 2101| 0| cpdwordpair_check(word, len, scratch, timelimit_exceeded, clock_time_start)) ------------------ | Branch (2101:19): [True: 0, False: 0] ------------------ 2102| 0| return nullptr; 2103| 0| return rv_first; 2104| 0| } 2105| | 2106| 216| numsyllable = oldnumsyllable2; 2107| 216| wordnum = oldwordnum2; 2108| | 2109| | // perhaps second word is a compound word (recursive call) 2110| | // (only if SPELL_COMPOUND_2 is not set and maxwordnum is not exceeded) 2111| 216| if ((!info || !(*info & SPELL_COMPOUND_2)) && wordnum + 2 < maxwordnum && wnum + 1 < maxwordnum) { ------------------ | | 85| 216|#define SPELL_COMPOUND_2 (1 << 7) // permit only 2 dictionary words in the compound ------------------ | Branch (2111:18): [True: 0, False: 216] | Branch (2111:27): [True: 119, False: 97] | Branch (2111:59): [True: 119, False: 0] | Branch (2111:87): [True: 119, False: 0] ------------------ 2112| 119| rv = compound_check(st.substr(i), wordnum + 1, 2113| 119| numsyllable, maxwordnum, wnum + 1, words, rwords, 0, 2114| 119| is_sug, info, scratch); 2115| | 2116| 119| if (rv && !checkcpdtable.empty() && i < word.size() && ------------------ | Branch (2116:19): [True: 0, False: 119] | Branch (2116:25): [True: 0, False: 0] | Branch (2116:51): [True: 0, False: 0] ------------------ 2117| 0| ((scpd == 0 && ------------------ | Branch (2117:21): [True: 0, False: 0] ------------------ 2118| 0| cpdpat_check(word, i, rv_first, rv, affixed)) || ------------------ | Branch (2118:21): [True: 0, False: 0] ------------------ 2119| 0| (scpd != 0 && ------------------ | Branch (2119:21): [True: 0, False: 0] ------------------ 2120| 0| !cpdpat_check(word, i, rv_first, rv, affixed)))) ------------------ | Branch (2120:21): [True: 0, False: 0] ------------------ 2121| 0| rv = nullptr; 2122| 119| } else { 2123| 97| rv = nullptr; 2124| 97| } 2125| 216| if (rv) { ------------------ | Branch (2125:17): [True: 0, False: 216] ------------------ 2126| | // forbid compound word, if it is a non-compound word with typical 2127| | // fault, or a dictionary word pair 2128| | 2129| 0| if (cpdwordpair_check(word, len, scratch, timelimit_exceeded, clock_time_start)) ------------------ | Branch (2129:19): [True: 0, False: 0] ------------------ 2130| 0| return nullptr; 2131| | 2132| 0| if (checkcompoundrep || forbiddenword) { ------------------ | Branch (2132:19): [True: 0, False: 0] | Branch (2132:39): [True: 0, False: 0] ------------------ 2133| 0| if (checkcompoundrep && cpdrep_check(word, len, scratch, timelimit_exceeded, clock_time_start)) ------------------ | Branch (2133:21): [True: 0, False: 0] | Branch (2133:41): [True: 0, False: 0] ------------------ 2134| 0| return nullptr; 2135| | 2136| | // check first part 2137| 0| if (i < word.size() && word.compare(i, rv->blen, rv->word, rv->blen) == 0) { ------------------ | Branch (2137:21): [True: 0, False: 0] | Branch (2137:40): [True: 0, False: 0] ------------------ 2138| 0| char r = st[i + rv->blen]; 2139| 0| st[i + rv->blen] = '\0'; 2140| | 2141| 0| if ((checkcompoundrep && cpdrep_check(st, i + rv->blen, scratch, timelimit_exceeded, clock_time_start)) || ------------------ | Branch (2141:24): [True: 0, False: 0] | Branch (2141:44): [True: 0, False: 0] ------------------ 2142| 0| cpdwordpair_check(st, i + rv->blen, scratch, timelimit_exceeded, clock_time_start)) { ------------------ | Branch (2142:23): [True: 0, False: 0] ------------------ 2143| 0| st[ + i + rv->blen] = r; 2144| 0| continue; 2145| 0| } 2146| | 2147| 0| if (forbiddenword) { ------------------ | Branch (2147:23): [True: 0, False: 0] ------------------ 2148| 0| struct hentry* rv2 = lookup(word.c_str(), word.size()); 2149| 0| if (!rv2 && len <= word.size()) ------------------ | Branch (2149:25): [True: 0, False: 0] | Branch (2149:33): [True: 0, False: 0] ------------------ 2150| 0| rv2 = affix_check(word, 0, len, scratch); 2151| 0| if (rv2 && rv2->astr && ------------------ | Branch (2151:25): [True: 0, False: 0] | Branch (2151:32): [True: 0, False: 0] ------------------ 2152| 0| TESTAFF(rv2->astr, forbiddenword, rv2->alen) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2153| 0| (strncmp(rv2->word, st.c_str(), i + rv->blen) == 0)) { ------------------ | Branch (2153:25): [True: 0, False: 0] ------------------ 2154| 0| return nullptr; 2155| 0| } 2156| 0| } 2157| 0| st[i + rv->blen] = r; 2158| 0| } 2159| 0| } 2160| 0| return rv_first; 2161| 0| } 2162| 216| } while (striple && !checkedstriple); // end of striple loop ------------------ | Branch (2162:20): [True: 0, False: 216] | Branch (2162:31): [True: 0, False: 0] ------------------ 2163| | 2164| 216| if (checkedstriple) { ------------------ | Branch (2164:15): [True: 0, False: 216] ------------------ 2165| 0| i++; 2166| 0| checkedstriple = 0; 2167| 0| striple = 0; 2168| 0| } 2169| | 2170| 216| } // first word is ok condition 2171| | 2172| 484M| if (soldi != 0) { ------------------ | Branch (2172:13): [True: 0, False: 484M] ------------------ 2173| 0| i = soldi; 2174| 0| soldi = 0; 2175| 0| len = oldlen; 2176| 0| cmin = oldcmin; 2177| 0| cmax = oldcmax; 2178| 0| } 2179| 484M| scpd++; 2180| | 2181| 484M| } while (!onlycpdrule && simplifiedcpd && ------------------ | Branch (2181:16): [True: 484M, False: 0] | Branch (2181:32): [True: 0, False: 484M] ------------------ 2182| 0| scpd <= checkcpdtable.size()); // end of simplifiedcpd loop ------------------ | Branch (2182:16): [True: 0, False: 0] ------------------ 2183| | 2184| 484M| scpd = 0; 2185| 484M| wordnum = oldwordnum; 2186| 484M| numsyllable = oldnumsyllable; 2187| | 2188| 484M| if (soldi != 0) { ------------------ | Branch (2188:11): [True: 0, False: 484M] ------------------ 2189| 0| i = soldi; 2190| 0| st.assign(word); // XXX add more optim. 2191| 0| soldi = 0; 2192| 0| len = oldlen; 2193| 0| cmin = oldcmin; 2194| 0| cmax = oldcmax; 2195| 0| } else 2196| 484M| st[i] = ch; 2197| | 2198| 484M| } while (!defcpdtable.empty() && oldwordnum == 0 && ------------------ | Branch (2198:14): [True: 0, False: 484M] | Branch (2198:38): [True: 0, False: 0] ------------------ 2199| 0| onlycpdrule++ < 1); // end of onlycpd loop ------------------ | Branch (2199:14): [True: 0, False: 0] ------------------ 2200| 485M| } 2201| | 2202| 18.8M| return nullptr; 2203| 19.7M|} _ZN8AffixMgr20compound_check_morphERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEssssPP6hentrySB_cRS6_PS7_R12AffixScratch: 2217| 2.06k| AffixScratch& scratch) { 2218| 2.06k| short oldnumsyllable, oldnumsyllable2, oldwordnum, oldwordnum2; 2219| 2.06k| hentry *rv = nullptr, *rv_first; 2220| 2.06k| std::string st, presult; 2221| 2.06k| char ch, affixed = 0; 2222| 2.06k| int checked_prefix, ok = 0; 2223| 2.06k| size_t cmin, cmax; 2224| 2.06k| hentry** oldwords = words; 2225| 2.06k| size_t len = word.size(); 2226| | 2227| | // protect subsequent words[wnum + 1] reads and any recursion 2228| 2.06k| if (wnum + 1 >= maxwordnum) ------------------ | Branch (2228:7): [True: 0, False: 2.06k] ------------------ 2229| 0| return 0; 2230| | 2231| | // add a time limit to handle possible 2232| | // combinatorical explosion of the overlapping words 2233| | 2234| 2.06k| HUNSPELL_THREAD_LOCAL std::chrono::steady_clock::time_point clock_time_start; ------------------ | | 60| 2.06k|# define HUNSPELL_THREAD_LOCAL thread_local ------------------ 2235| 2.06k| HUNSPELL_THREAD_LOCAL bool timelimit_exceeded; ------------------ | | 60| 2.06k|# define HUNSPELL_THREAD_LOCAL thread_local ------------------ 2236| | 2237| | // get the current time 2238| 2.06k| std::chrono::steady_clock::time_point clock_now = std::chrono::steady_clock::now(); 2239| | 2240| 2.06k| if (wnum == 0) { ------------------ | Branch (2240:7): [True: 2.06k, False: 0] ------------------ 2241| | // set the start time 2242| 2.06k| clock_time_start = clock_now; 2243| 2.06k| timelimit_exceeded = false; 2244| 2.06k| } 2245| 0| else if (clock_now - clock_time_start > TIMELIMIT_MS) ------------------ | | 106| 0|#define TIMELIMIT_MS std::chrono::milliseconds(50) ------------------ | Branch (2245:12): [True: 0, False: 0] ------------------ 2246| 0| timelimit_exceeded = true; 2247| | 2248| 2.06k| setcminmax(&cmin, &cmax, word.c_str(), len); 2249| | 2250| 2.06k| st.assign(word); 2251| | 2252| 32.4k| for (size_t i = cmin; i < cmax; ++i) { ------------------ | Branch (2252:25): [True: 30.5k, False: 1.83k] ------------------ 2253| | // go to end of the UTF-8 character 2254| 30.5k| if (utf8) { ------------------ | Branch (2254:9): [True: 9.76k, False: 20.8k] ------------------ 2255| 13.4k| for (; is_utf8_cont(st[i]); i++) ------------------ | Branch (2255:14): [True: 3.66k, False: 9.76k] ------------------ 2256| 3.66k| ; 2257| 9.76k| if (i >= cmax) ------------------ | Branch (2257:11): [True: 97, False: 9.66k] ------------------ 2258| 97| return 0; 2259| 9.76k| } 2260| | 2261| 30.4k| words = oldwords; 2262| 30.4k| int onlycpdrule = (words) ? 1 : 0; ------------------ | Branch (2262:23): [True: 0, False: 30.4k] ------------------ 2263| | 2264| 30.4k| do { // onlycpdrule loop 2265| | 2266| 30.4k| if (timelimit_exceeded || ------------------ | Branch (2266:11): [True: 0, False: 30.4k] | Branch (2266:11): [True: 0, False: 30.4k] ------------------ 2267| 30.4k| std::chrono::steady_clock::now() - clock_time_start > TIMELIMIT_MS) { ------------------ | | 106| 30.4k|#define TIMELIMIT_MS std::chrono::milliseconds(50) ------------------ | Branch (2267:11): [True: 0, False: 30.4k] ------------------ 2268| 0| timelimit_exceeded = true; 2269| 0| return 0; 2270| 0| } 2271| | 2272| 30.4k| oldnumsyllable = numsyllable; 2273| 30.4k| oldwordnum = wordnum; 2274| 30.4k| checked_prefix = 0; 2275| | 2276| 30.4k| if (i >= st.size()) ------------------ | Branch (2276:11): [True: 132, False: 30.3k] ------------------ 2277| 132| return 0; 2278| | 2279| 30.3k| ch = st[i]; 2280| 30.3k| st[i] = '\0'; 2281| 30.3k| sfx = nullptr; 2282| | 2283| | // FIRST WORD 2284| | 2285| 30.3k| affixed = 1; 2286| | 2287| 30.3k| presult.clear(); 2288| 30.3k| if (partresult) ------------------ | Branch (2288:11): [True: 0, False: 30.3k] ------------------ 2289| 0| presult.append(*partresult); 2290| | 2291| 30.3k| rv = lookup(st.c_str(), i); // perhaps without prefix 2292| | 2293| | // forbid dictionary stems with COMPOUNDFORBIDFLAG in 2294| | // compound words, overriding the effect of COMPOUNDPERMITFLAG 2295| 30.3k| if ((rv) && compoundforbidflag && ------------------ | Branch (2295:11): [True: 133, False: 30.2k] | Branch (2295:19): [True: 1, False: 132] ------------------ 2296| 1| TESTAFF(rv->astr, compoundforbidflag, rv->alen) && !hu_mov_rule) ------------------ | | 99| 30.3k|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 1] | | ------------------ ------------------ | Branch (2296:66): [True: 0, False: 0] ------------------ 2297| 0| continue; 2298| | 2299| | // search homonym with compound flag 2300| 30.4k| while ((rv) && !hu_mov_rule && ------------------ | Branch (2300:14): [True: 133, False: 30.3k] | Branch (2300:22): [True: 133, False: 0] ------------------ 2301| 133| ((needaffix && TESTAFF(rv->astr, needaffix, rv->alen)) || ------------------ | | 99| 44|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 44] | | ------------------ ------------------ | Branch (2301:16): [True: 44, False: 89] ------------------ 2302| 133| !((compoundflag && !words && !onlycpdrule && ------------------ | Branch (2302:18): [True: 133, False: 0] | Branch (2302:34): [True: 133, False: 0] | Branch (2302:44): [True: 133, False: 0] ------------------ 2303| 133| TESTAFF(rv->astr, compoundflag, rv->alen)) || ------------------ | | 99| 133|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 133] | | ------------------ ------------------ 2304| 133| (compoundbegin && !wordnum && !onlycpdrule && ------------------ | Branch (2304:18): [True: 0, False: 133] | Branch (2304:35): [True: 0, False: 0] | Branch (2304:47): [True: 0, False: 0] ------------------ 2305| 0| TESTAFF(rv->astr, compoundbegin, rv->alen)) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2306| 133| (compoundmiddle && wordnum && !words && !onlycpdrule && ------------------ | Branch (2306:18): [True: 0, False: 133] | Branch (2306:36): [True: 0, False: 0] | Branch (2306:47): [True: 0, False: 0] | Branch (2306:57): [True: 0, False: 0] ------------------ 2307| 0| TESTAFF(rv->astr, compoundmiddle, rv->alen)) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2308| 133| (!defcpdtable.empty() && onlycpdrule && ------------------ | Branch (2308:18): [True: 0, False: 133] | Branch (2308:42): [True: 0, False: 0] ------------------ 2309| 0| ((!words && !wordnum && ------------------ | Branch (2309:20): [True: 0, False: 0] | Branch (2309:30): [True: 0, False: 0] ------------------ 2310| 0| defcpd_check(&words, wnum, maxwordnum, rv, rwords, 0)) || ------------------ | Branch (2310:20): [True: 0, False: 0] ------------------ 2311| 0| (words && ------------------ | Branch (2311:20): [True: 0, False: 0] ------------------ 2312| 133| defcpd_check(&words, wnum, maxwordnum, rv, rwords, 0))))))) { ------------------ | Branch (2312:20): [True: 0, False: 0] ------------------ 2313| 133| rv = rv->next_homonym; 2314| 133| } 2315| | 2316| | 2317| 30.3k| if (rv) ------------------ | Branch (2317:11): [True: 0, False: 30.3k] ------------------ 2318| 0| affixed = 0; 2319| | 2320| 30.3k| if (rv) { ------------------ | Branch (2320:11): [True: 0, False: 30.3k] ------------------ 2321| 0| presult.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 2322| 0| presult.append(MORPH_PART); ------------------ | | 116| 0|#define MORPH_PART "pa:" ------------------ 2323| 0| presult.append(st, 0, i); 2324| 0| if (!HENTRY_FIND(rv, MORPH_STEM)) { ------------------ | | 103| 0|#define MORPH_STEM "st:" ------------------ | Branch (2324:13): [True: 0, False: 0] ------------------ 2325| 0| presult.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 2326| 0| presult.append(MORPH_STEM); ------------------ | | 103| 0|#define MORPH_STEM "st:" ------------------ 2327| 0| presult.append(st, 0, i); 2328| 0| } 2329| 0| if (HENTRY_DATA(rv)) { ------------------ | Branch (2329:13): [True: 0, False: 0] ------------------ 2330| 0| presult.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 2331| 0| presult.append(HENTRY_DATA2(rv)); 2332| 0| } 2333| 0| } 2334| | 2335| 30.3k| if (!rv) { ------------------ | Branch (2335:11): [True: 30.3k, False: 0] ------------------ 2336| 30.3k| if (compoundflag && ------------------ | Branch (2336:13): [True: 30.3k, False: 0] ------------------ 2337| 30.3k| !(rv = ------------------ | Branch (2337:13): [True: 30.3k, False: 0] ------------------ 2338| 30.3k| prefix_check(st, 0, i, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN, ------------------ | | 75| 0|#define IN_CPD_OTHER 3 ------------------ prefix_check(st, 0, i, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN, ------------------ | | 73| 60.7k|#define IN_CPD_BEGIN 1 ------------------ | Branch (2338:42): [True: 0, False: 30.3k] ------------------ 2339| 30.3k| scratch, compoundflag))) { 2340| 30.3k| if (((rv = suffix_check(st, 0, i, 0, nullptr, scratch, FLAG_NULL, compoundflag, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || ------------------ | | 96| 30.3k|#define FLAG_NULL 0x00 ------------------ if (((rv = suffix_check(st, 0, i, 0, nullptr, scratch, FLAG_NULL, compoundflag, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || ------------------ | | 75| 0|#define IN_CPD_OTHER 3 ------------------ if (((rv = suffix_check(st, 0, i, 0, nullptr, scratch, FLAG_NULL, compoundflag, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || ------------------ | | 73| 60.7k|#define IN_CPD_BEGIN 1 ------------------ | Branch (2340:16): [True: 0, False: 30.3k] | Branch (2340:91): [True: 0, False: 30.3k] ------------------ 2341| 30.3k| (compoundmoresuffixes && (rv = suffix_check_twosfx(st, 0, i, 0, nullptr, scratch, compoundflag)))) && ------------------ | Branch (2341:17): [True: 0, False: 30.3k] | Branch (2341:41): [True: 0, False: 0] ------------------ 2342| 0| !hu_mov_rule && sfx->getCont() && ------------------ | Branch (2342:15): [True: 0, False: 0] | Branch (2342:31): [True: 0, False: 0] ------------------ 2343| 0| ((compoundforbidflag && TESTAFF(sfx->getCont(), compoundforbidflag, sfx->getContLen())) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (2343:17): [True: 0, False: 0] ------------------ 2344| 0| (compoundend && TESTAFF(sfx->getCont(), compoundend, sfx->getContLen())))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (2344:17): [True: 0, False: 0] ------------------ 2345| 0| rv = nullptr; 2346| 0| } 2347| 30.3k| } 2348| | 2349| 30.3k| if (rv || ------------------ | Branch (2349:13): [True: 0, False: 30.3k] ------------------ 2350| 30.3k| (((wordnum == 0) && compoundbegin && ------------------ | Branch (2350:15): [True: 30.3k, False: 0] | Branch (2350:33): [True: 0, False: 30.3k] ------------------ 2351| 0| ((rv = suffix_check(st, 0, i, 0, nullptr, scratch, FLAG_NULL, compoundbegin, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || ------------------ | | 96| 0|#define FLAG_NULL 0x00 ------------------ ((rv = suffix_check(st, 0, i, 0, nullptr, scratch, FLAG_NULL, compoundbegin, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || ------------------ | | 75| 0|#define IN_CPD_OTHER 3 ------------------ ((rv = suffix_check(st, 0, i, 0, nullptr, scratch, FLAG_NULL, compoundbegin, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || ------------------ | | 73| 0|#define IN_CPD_BEGIN 1 ------------------ | Branch (2351:16): [True: 0, False: 0] | Branch (2351:92): [True: 0, False: 0] ------------------ 2352| 0| (compoundmoresuffixes && (rv = suffix_check_twosfx(st, 0, i, 0, nullptr, scratch, ------------------ | Branch (2352:17): [True: 0, False: 0] | Branch (2352:41): [True: 0, False: 0] ------------------ 2353| 0| compoundbegin))) || // twofold suffix+compound 2354| 0| (rv = prefix_check(st, 0, i, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN, scratch, compoundbegin)))) || ------------------ | | 75| 0|#define IN_CPD_OTHER 3 ------------------ (rv = prefix_check(st, 0, i, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN, scratch, compoundbegin)))) || ------------------ | | 73| 0|#define IN_CPD_BEGIN 1 ------------------ | Branch (2354:16): [True: 0, False: 0] | Branch (2354:45): [True: 0, False: 0] ------------------ 2355| 30.3k| ((wordnum > 0) && compoundmiddle && ------------------ | Branch (2355:15): [True: 0, False: 30.3k] | Branch (2355:32): [True: 0, False: 0] ------------------ 2356| 0| ((rv = suffix_check(st, 0, i, 0, nullptr, scratch, FLAG_NULL, compoundmiddle, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || ------------------ | | 96| 0|#define FLAG_NULL 0x00 ------------------ ((rv = suffix_check(st, 0, i, 0, nullptr, scratch, FLAG_NULL, compoundmiddle, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || ------------------ | | 75| 0|#define IN_CPD_OTHER 3 ------------------ ((rv = suffix_check(st, 0, i, 0, nullptr, scratch, FLAG_NULL, compoundmiddle, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || ------------------ | | 73| 0|#define IN_CPD_BEGIN 1 ------------------ | Branch (2356:16): [True: 0, False: 0] | Branch (2356:93): [True: 0, False: 0] ------------------ 2357| 0| (compoundmoresuffixes && (rv = suffix_check_twosfx(st, 0, i, 0, nullptr, scratch, ------------------ | Branch (2357:17): [True: 0, False: 0] | Branch (2357:41): [True: 0, False: 0] ------------------ 2358| 0| compoundmiddle))) || // twofold suffix+compound 2359| 0| (rv = prefix_check(st, 0, i, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN, scratch, compoundmiddle)))))) { ------------------ | | 75| 0|#define IN_CPD_OTHER 3 ------------------ (rv = prefix_check(st, 0, i, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN, scratch, compoundmiddle)))))) { ------------------ | | 73| 0|#define IN_CPD_BEGIN 1 ------------------ | Branch (2359:16): [True: 0, False: 0] | Branch (2359:45): [True: 0, False: 0] ------------------ 2360| 0| std::string p; 2361| 0| if (compoundflag) ------------------ | Branch (2361:15): [True: 0, False: 0] ------------------ 2362| 0| p = affix_check_morph(st, 0, i, scratch, compoundflag); 2363| 0| if (p.empty()) { ------------------ | Branch (2363:15): [True: 0, False: 0] ------------------ 2364| 0| if ((wordnum == 0) && compoundbegin) { ------------------ | Branch (2364:17): [True: 0, False: 0] | Branch (2364:35): [True: 0, False: 0] ------------------ 2365| 0| p = affix_check_morph(st, 0, i, scratch, compoundbegin); 2366| 0| } else if ((wordnum > 0) && compoundmiddle) { ------------------ | Branch (2366:24): [True: 0, False: 0] | Branch (2366:41): [True: 0, False: 0] ------------------ 2367| 0| p = affix_check_morph(st, 0, i, scratch, compoundmiddle); 2368| 0| } 2369| 0| } 2370| 0| if (!p.empty()) { ------------------ | Branch (2370:15): [True: 0, False: 0] ------------------ 2371| 0| presult.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 2372| 0| presult.append(MORPH_PART); ------------------ | | 116| 0|#define MORPH_PART "pa:" ------------------ 2373| 0| presult.append(st, 0, i); 2374| 0| line_uniq_app(p, MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 2375| 0| presult.append(p); 2376| 0| } 2377| 0| checked_prefix = 1; 2378| 0| } 2379| | // else check forbiddenwords 2380| 30.3k| } else if (rv->astr && (TESTAFF(rv->astr, forbiddenword, rv->alen) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (2380:18): [True: 0, False: 0] | Branch (2380:18): [True: 0, False: 0] ------------------ 2381| 0| TESTAFF(rv->astr, ONLYUPCASEFLAG, rv->alen) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2382| 0| TESTAFF(rv->astr, needaffix, rv->alen))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2383| 0| st[i] = ch; 2384| 0| continue; 2385| 0| } 2386| | 2387| | // check non_compound flag in suffix and prefix 2388| 30.3k| if ((rv) && !hu_mov_rule && ------------------ | Branch (2388:11): [True: 0, False: 30.3k] | Branch (2388:19): [True: 0, False: 0] ------------------ 2389| 0| ((pfx && pfx->getCont() && ------------------ | Branch (2389:13): [True: 0, False: 0] | Branch (2389:20): [True: 0, False: 0] ------------------ 2390| 0| TESTAFF(pfx->getCont(), compoundforbidflag, pfx->getContLen())) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2391| 0| (sfx && sfx->getCont() && ------------------ | Branch (2391:13): [True: 0, False: 0] | Branch (2391:20): [True: 0, False: 0] ------------------ 2392| 0| TESTAFF(sfx->getCont(), compoundforbidflag, sfx->getContLen())))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2393| 0| continue; 2394| 0| } 2395| | 2396| | // check compoundend flag in suffix and prefix 2397| 30.3k| if ((rv) && !checked_prefix && compoundend && !hu_mov_rule && ------------------ | Branch (2397:11): [True: 0, False: 30.3k] | Branch (2397:19): [True: 0, False: 0] | Branch (2397:38): [True: 0, False: 0] | Branch (2397:53): [True: 0, False: 0] ------------------ 2398| 0| ((pfx && pfx->getCont() && ------------------ | Branch (2398:13): [True: 0, False: 0] | Branch (2398:20): [True: 0, False: 0] ------------------ 2399| 0| TESTAFF(pfx->getCont(), compoundend, pfx->getContLen())) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2400| 0| (sfx && sfx->getCont() && ------------------ | Branch (2400:13): [True: 0, False: 0] | Branch (2400:20): [True: 0, False: 0] ------------------ 2401| 0| TESTAFF(sfx->getCont(), compoundend, sfx->getContLen())))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2402| 0| continue; 2403| 0| } 2404| | 2405| | // check compoundmiddle flag in suffix and prefix 2406| 30.3k| if ((rv) && !checked_prefix && (wordnum == 0) && compoundmiddle && ------------------ | Branch (2406:11): [True: 0, False: 30.3k] | Branch (2406:19): [True: 0, False: 0] | Branch (2406:38): [True: 0, False: 0] | Branch (2406:56): [True: 0, False: 0] ------------------ 2407| 0| !hu_mov_rule && ------------------ | Branch (2407:11): [True: 0, False: 0] ------------------ 2408| 0| ((pfx && pfx->getCont() && ------------------ | Branch (2408:13): [True: 0, False: 0] | Branch (2408:20): [True: 0, False: 0] ------------------ 2409| 0| TESTAFF(pfx->getCont(), compoundmiddle, pfx->getContLen())) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2410| 0| (sfx && sfx->getCont() && ------------------ | Branch (2410:13): [True: 0, False: 0] | Branch (2410:20): [True: 0, False: 0] ------------------ 2411| 0| TESTAFF(sfx->getCont(), compoundmiddle, sfx->getContLen())))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2412| 0| rv = nullptr; 2413| 0| } 2414| | 2415| | // check forbiddenwords 2416| 30.3k| if ((rv) && (rv->astr) && (TESTAFF(rv->astr, forbiddenword, rv->alen) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (2416:11): [True: 0, False: 30.3k] | Branch (2416:11): [True: 0, False: 30.3k] | Branch (2416:19): [True: 0, False: 0] ------------------ 2417| 0| TESTAFF(rv->astr, ONLYUPCASEFLAG, rv->alen))) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2418| 0| continue; 2419| | 2420| | // increment word number, if the second root has a compoundroot flag 2421| 30.3k| if ((rv) && (compoundroot) && ------------------ | Branch (2421:11): [True: 0, False: 30.3k] | Branch (2421:19): [True: 0, False: 0] ------------------ 2422| 0| (TESTAFF(rv->astr, compoundroot, rv->alen))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (2422:11): [True: 0, False: 0] ------------------ 2423| 0| wordnum++; 2424| 0| } 2425| | 2426| | // first word is acceptable in compound words? 2427| 30.3k| if (((rv) && ------------------ | Branch (2427:11): [True: 0, False: 30.3k] | Branch (2427:12): [True: 0, False: 30.3k] ------------------ 2428| 0| (checked_prefix || (words && words[wnum]) || (compoundflag && TESTAFF(rv->astr, compoundflag, rv->alen)) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (2428:13): [True: 0, False: 0] | Branch (2428:32): [True: 0, False: 0] | Branch (2428:41): [True: 0, False: 0] | Branch (2428:58): [True: 0, False: 0] ------------------ 2429| 0| ((oldwordnum == 0) && compoundbegin && TESTAFF(rv->astr, compoundbegin, rv->alen)) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (2429:14): [True: 0, False: 0] | Branch (2429:35): [True: 0, False: 0] ------------------ 2430| 0| ((oldwordnum > 0) && compoundmiddle && TESTAFF(rv->astr, compoundmiddle, rv->alen)) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (2430:14): [True: 0, False: 0] | Branch (2430:34): [True: 0, False: 0] ------------------ 2431| | // LANG_hu section: spec. Hungarian rule 2432| 0| || ((langnum == LANG_hu) && // hu_mov_rule ------------------ | Branch (2432:17): [True: 0, False: 0] ------------------ 2433| 0| hu_mov_rule && ------------------ | Branch (2433:17): [True: 0, False: 0] ------------------ 2434| 0| (TESTAFF(rv->astr, 'F', rv->alen) || TESTAFF(rv->astr, 'G', rv->alen) || TESTAFF(rv->astr, 'H', rv->alen))) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ (TESTAFF(rv->astr, 'F', rv->alen) || TESTAFF(rv->astr, 'G', rv->alen) || TESTAFF(rv->astr, 'H', rv->alen))) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ (TESTAFF(rv->astr, 'F', rv->alen) || TESTAFF(rv->astr, 'G', rv->alen) || TESTAFF(rv->astr, 'H', rv->alen))) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2435| | // END of LANG_hu section 2436| 0| ) && 2437| 0| !((checkcompoundtriple && !words && // test triple letters ------------------ | Branch (2437:15): [True: 0, False: 0] | Branch (2437:38): [True: 0, False: 0] ------------------ 2438| 0| (word[i - 1] == word[i]) && ------------------ | Branch (2438:15): [True: 0, False: 0] ------------------ 2439| 0| (((i > 1) && (word[i - 1] == word[i - 2])) || ((word[i - 1] == word[i + 1])) // may be word[i+1] == '\0' ------------------ | Branch (2439:17): [True: 0, False: 0] | Branch (2439:28): [True: 0, False: 0] | Branch (2439:61): [True: 0, False: 0] ------------------ 2440| 0| )) || 2441| 0| ( 2442| | // test CHECKCOMPOUNDPATTERN 2443| 0| !checkcpdtable.empty() && !words && cpdpat_check(word, i, rv, nullptr, affixed)) || ------------------ | Branch (2443:18): [True: 0, False: 0] | Branch (2443:44): [True: 0, False: 0] | Branch (2443:54): [True: 0, False: 0] ------------------ 2444| 0| (checkcompoundcase && !words && cpdcase_check(word, i)))) ------------------ | Branch (2444:15): [True: 0, False: 0] | Branch (2444:36): [True: 0, False: 0] | Branch (2444:46): [True: 0, False: 0] ------------------ 2445| | // LANG_hu section: spec. Hungarian rule 2446| 30.3k| || ((!rv) && (langnum == LANG_hu) && hu_mov_rule && (rv = affix_check(st, 0, i, scratch)) && ------------------ | Branch (2446:15): [True: 30.3k, False: 0] | Branch (2446:24): [True: 9.66k, False: 20.7k] | Branch (2446:48): [True: 0, False: 9.66k] | Branch (2446:63): [True: 0, False: 0] ------------------ 2447| 0| (sfx && sfx->getCont() && ------------------ | Branch (2447:16): [True: 0, False: 0] | Branch (2447:23): [True: 0, False: 0] ------------------ 2448| 0| (TESTAFF(sfx->getCont(), (unsigned short)'x', sfx->getContLen()) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2449| 0| TESTAFF(sfx->getCont(), (unsigned short)'%', sfx->getContLen())))) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2450| | // END of LANG_hu section 2451| 30.3k| ) { 2452| | // LANG_hu section: spec. Hungarian rule 2453| 0| if (langnum == LANG_hu) { ------------------ | Branch (2453:13): [True: 0, False: 0] ------------------ 2454| | // calculate syllable number of the word 2455| 0| numsyllable += get_syllable(st.substr(0, i)); 2456| | 2457| | // + 1 word, if syllable number of the prefix > 1 (hungarian 2458| | // convention) 2459| 0| if (pfx && (get_syllable(pfx->getKey()) > 1)) ------------------ | Branch (2459:15): [True: 0, False: 0] | Branch (2459:15): [True: 0, False: 0] | Branch (2459:22): [True: 0, False: 0] ------------------ 2460| 0| wordnum++; 2461| 0| } 2462| | // END of LANG_hu section 2463| | 2464| | // NEXT WORD(S) 2465| 0| rv_first = rv; 2466| 0| rv = lookup(word.c_str() + i, word.size() - i); // perhaps without prefix 2467| | 2468| | // search homonym with compound flag 2469| 0| while ((rv) && ((needaffix && TESTAFF(rv->astr, needaffix, rv->alen)) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (2469:16): [True: 0, False: 0] | Branch (2469:26): [True: 0, False: 0] ------------------ 2470| 0| !((compoundflag && !words && TESTAFF(rv->astr, compoundflag, rv->alen)) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (2470:28): [True: 0, False: 0] | Branch (2470:44): [True: 0, False: 0] ------------------ 2471| 0| (compoundend && !words && TESTAFF(rv->astr, compoundend, rv->alen)) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (2471:28): [True: 0, False: 0] | Branch (2471:43): [True: 0, False: 0] ------------------ 2472| 0| (!defcpdtable.empty() && words && defcpd_check(&words, wnum + 1, maxwordnum, rv, nullptr, 1))))) { ------------------ | Branch (2472:28): [True: 0, False: 0] | Branch (2472:52): [True: 0, False: 0] | Branch (2472:61): [True: 0, False: 0] ------------------ 2473| 0| rv = rv->next_homonym; 2474| 0| } 2475| | 2476| 0| if (rv && words && words[wnum + 1]) { ------------------ | Branch (2476:13): [True: 0, False: 0] | Branch (2476:19): [True: 0, False: 0] | Branch (2476:28): [True: 0, False: 0] ------------------ 2477| 0| result.append(presult); 2478| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 2479| 0| result.append(MORPH_PART); ------------------ | | 116| 0|#define MORPH_PART "pa:" ------------------ 2480| 0| result.append(word, i, word.size()); 2481| 0| if (complexprefixes && HENTRY_DATA(rv)) ------------------ | Branch (2481:15): [True: 0, False: 0] | Branch (2481:34): [True: 0, False: 0] ------------------ 2482| 0| result.append(HENTRY_DATA2(rv)); 2483| 0| if (!HENTRY_FIND(rv, MORPH_STEM)) { ------------------ | | 103| 0|#define MORPH_STEM "st:" ------------------ | Branch (2483:15): [True: 0, False: 0] ------------------ 2484| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 2485| 0| result.append(MORPH_STEM); ------------------ | | 103| 0|#define MORPH_STEM "st:" ------------------ 2486| 0| result.append(HENTRY_WORD(rv)); ------------------ | | 54| 0|#define HENTRY_WORD(h) &(h->word[0]) ------------------ 2487| 0| } 2488| | // store the pointer of the hash entry 2489| 0| if (!complexprefixes && HENTRY_DATA(rv)) { ------------------ | Branch (2489:15): [True: 0, False: 0] | Branch (2489:35): [True: 0, False: 0] ------------------ 2490| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 2491| 0| result.append(HENTRY_DATA2(rv)); 2492| 0| } 2493| 0| result.push_back(MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 2494| 0| return 0; 2495| 0| } 2496| | 2497| 0| oldnumsyllable2 = numsyllable; 2498| 0| oldwordnum2 = wordnum; 2499| | 2500| | // LANG_hu section: spec. Hungarian rule 2501| 0| if ((rv) && (langnum == LANG_hu) && ------------------ | Branch (2501:13): [True: 0, False: 0] | Branch (2501:13): [True: 0, False: 0] | Branch (2501:21): [True: 0, False: 0] ------------------ 2502| 0| (TESTAFF(rv->astr, 'I', rv->alen)) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (2502:13): [True: 0, False: 0] ------------------ 2503| 0| !(TESTAFF(rv->astr, 'J', rv->alen))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (2503:13): [True: 0, False: 0] ------------------ 2504| 0| numsyllable--; 2505| 0| } 2506| | // END of LANG_hu section 2507| | // increment word number, if the second root has a compoundroot flag 2508| 0| if ((rv) && (compoundroot) && ------------------ | Branch (2508:13): [True: 0, False: 0] | Branch (2508:21): [True: 0, False: 0] ------------------ 2509| 0| (TESTAFF(rv->astr, compoundroot, rv->alen))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (2509:13): [True: 0, False: 0] ------------------ 2510| 0| wordnum++; 2511| 0| } 2512| | 2513| | // check forbiddenwords 2514| 0| if ((rv) && (rv->astr) && ------------------ | Branch (2514:13): [True: 0, False: 0] | Branch (2514:13): [True: 0, False: 0] | Branch (2514:21): [True: 0, False: 0] ------------------ 2515| 0| (TESTAFF(rv->astr, forbiddenword, rv->alen) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2516| 0| TESTAFF(rv->astr, ONLYUPCASEFLAG, rv->alen))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2517| 0| st[i] = ch; 2518| 0| continue; 2519| 0| } 2520| | 2521| | // second word is acceptable, as a root? 2522| | // hungarian conventions: compounding is acceptable, 2523| | // when compound forms consist of 2 words, or if more, 2524| | // then the syllable number of root words must be 6, or lesser. 2525| 0| if ((rv) && ------------------ | Branch (2525:13): [True: 0, False: 0] | Branch (2525:13): [True: 0, False: 0] ------------------ 2526| 0| ((compoundflag && TESTAFF(rv->astr, compoundflag, rv->alen)) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (2526:15): [True: 0, False: 0] ------------------ 2527| 0| (compoundend && TESTAFF(rv->astr, compoundend, rv->alen))) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (2527:15): [True: 0, False: 0] ------------------ 2528| 0| (((cpdwordmax == -1) || (wordnum + 1 < cpdwordmax)) || ------------------ | Branch (2528:15): [True: 0, False: 0] | Branch (2528:37): [True: 0, False: 0] ------------------ 2529| 0| ((cpdmaxsyllable != 0) && ------------------ | Branch (2529:15): [True: 0, False: 0] ------------------ 2530| 0| (numsyllable + get_syllable(std::string(HENTRY_WORD(rv), rv->blen)) <= ------------------ | | 54| 0|#define HENTRY_WORD(h) &(h->word[0]) ------------------ | Branch (2530:15): [True: 0, False: 0] ------------------ 2531| 0| cpdmaxsyllable))) && 2532| 0| ((!checkcompounddup || (rv != rv_first)))) { ------------------ | Branch (2532:15): [True: 0, False: 0] | Branch (2532:36): [True: 0, False: 0] ------------------ 2533| | // bad compound word 2534| 0| result.append(presult); 2535| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 2536| 0| result.append(MORPH_PART); ------------------ | | 116| 0|#define MORPH_PART "pa:" ------------------ 2537| 0| result.append(word, i, word.size()); 2538| | 2539| 0| if (HENTRY_DATA(rv)) { ------------------ | Branch (2539:15): [True: 0, False: 0] ------------------ 2540| 0| if (complexprefixes) ------------------ | Branch (2540:17): [True: 0, False: 0] ------------------ 2541| 0| result.append(HENTRY_DATA2(rv)); 2542| 0| if (!HENTRY_FIND(rv, MORPH_STEM)) { ------------------ | | 103| 0|#define MORPH_STEM "st:" ------------------ | Branch (2542:17): [True: 0, False: 0] ------------------ 2543| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 2544| 0| result.append(MORPH_STEM); ------------------ | | 103| 0|#define MORPH_STEM "st:" ------------------ 2545| 0| result.append(HENTRY_WORD(rv)); ------------------ | | 54| 0|#define HENTRY_WORD(h) &(h->word[0]) ------------------ 2546| 0| } 2547| | // store the pointer of the hash entry 2548| 0| if (!complexprefixes) { ------------------ | Branch (2548:17): [True: 0, False: 0] ------------------ 2549| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 2550| 0| result.append(HENTRY_DATA2(rv)); 2551| 0| } 2552| 0| } 2553| 0| result.push_back(MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 2554| 0| ok = 1; 2555| 0| } 2556| | 2557| 0| numsyllable = oldnumsyllable2; 2558| 0| wordnum = oldwordnum2; 2559| | 2560| | // perhaps second word has prefix or/and suffix 2561| 0| sfx = nullptr; 2562| 0| sfxflag = FLAG_NULL; ------------------ | | 96| 0|#define FLAG_NULL 0x00 ------------------ 2563| | 2564| 0| if (compoundflag && !onlycpdrule) ------------------ | Branch (2564:13): [True: 0, False: 0] | Branch (2564:29): [True: 0, False: 0] ------------------ 2565| 0| rv = affix_check(word, i, word.size() - i, scratch, compoundflag); 2566| 0| else 2567| 0| rv = nullptr; 2568| | 2569| 0| if (!rv && compoundend && !onlycpdrule) { ------------------ | Branch (2569:13): [True: 0, False: 0] | Branch (2569:20): [True: 0, False: 0] | Branch (2569:35): [True: 0, False: 0] ------------------ 2570| 0| sfx = nullptr; 2571| 0| pfx = nullptr; 2572| 0| rv = affix_check(word, i, word.size() - i, scratch, compoundend); 2573| 0| } 2574| | 2575| 0| if (!rv && !defcpdtable.empty() && words) { ------------------ | Branch (2575:13): [True: 0, False: 0] | Branch (2575:20): [True: 0, False: 0] | Branch (2575:44): [True: 0, False: 0] ------------------ 2576| 0| rv = affix_check(word, i, word.size() - i, scratch, 0, IN_CPD_END); ------------------ | | 74| 0|#define IN_CPD_END 2 ------------------ 2577| 0| if (rv && words && defcpd_check(&words, wnum + 1, maxwordnum, rv, nullptr, 1)) { ------------------ | Branch (2577:15): [True: 0, False: 0] | Branch (2577:21): [True: 0, False: 0] | Branch (2577:30): [True: 0, False: 0] ------------------ 2578| 0| std::string m; 2579| 0| if (compoundflag) ------------------ | Branch (2579:17): [True: 0, False: 0] ------------------ 2580| 0| m = affix_check_morph(word, i, word.size() - i, scratch, compoundflag); 2581| 0| if (m.empty() && compoundend) { ------------------ | Branch (2581:17): [True: 0, False: 0] | Branch (2581:30): [True: 0, False: 0] ------------------ 2582| 0| m = affix_check_morph(word, i, word.size() - i, scratch, compoundend); 2583| 0| } 2584| 0| result.append(presult); 2585| 0| if (!m.empty()) { ------------------ | Branch (2585:17): [True: 0, False: 0] ------------------ 2586| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 2587| 0| result.append(MORPH_PART); ------------------ | | 116| 0|#define MORPH_PART "pa:" ------------------ 2588| 0| result.append(word, i, word.size()); 2589| 0| line_uniq_app(m, MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 2590| 0| result.append(m); 2591| 0| } 2592| 0| result.push_back(MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 2593| 0| ok = 1; 2594| 0| } 2595| 0| } 2596| | 2597| | // check non_compound flag in suffix and prefix 2598| 0| if ((rv) && ------------------ | Branch (2598:13): [True: 0, False: 0] ------------------ 2599| 0| ((pfx && pfx->getCont() && ------------------ | Branch (2599:15): [True: 0, False: 0] | Branch (2599:22): [True: 0, False: 0] ------------------ 2600| 0| TESTAFF(pfx->getCont(), compoundforbidflag, pfx->getContLen())) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2601| 0| (sfx && sfx->getCont() && ------------------ | Branch (2601:15): [True: 0, False: 0] | Branch (2601:22): [True: 0, False: 0] ------------------ 2602| 0| TESTAFF(sfx->getCont(), compoundforbidflag, ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2603| 0| sfx->getContLen())))) { 2604| 0| rv = nullptr; 2605| 0| } 2606| | 2607| | // check forbiddenwords 2608| 0| if ((rv) && (rv->astr) && ------------------ | Branch (2608:13): [True: 0, False: 0] | Branch (2608:13): [True: 0, False: 0] | Branch (2608:21): [True: 0, False: 0] ------------------ 2609| 0| (TESTAFF(rv->astr, forbiddenword, rv->alen) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2610| 0| TESTAFF(rv->astr, ONLYUPCASEFLAG, rv->alen)) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2611| 0| (!TESTAFF(rv->astr, needaffix, rv->alen))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (2611:13): [True: 0, False: 0] ------------------ 2612| 0| st[i] = ch; 2613| 0| continue; 2614| 0| } 2615| | 2616| 0| if (langnum == LANG_hu) { ------------------ | Branch (2616:13): [True: 0, False: 0] ------------------ 2617| | // calculate syllable number of the word 2618| 0| numsyllable += get_syllable(word.c_str() + i); 2619| | 2620| | // - affix syllable num. 2621| | // XXX only second suffix (inflections, not derivations) 2622| 0| if (sfxappnd) { ------------------ | Branch (2622:15): [True: 0, False: 0] ------------------ 2623| 0| std::string tmp(sfxappnd); 2624| 0| reverseword(tmp); 2625| 0| numsyllable -= short(get_syllable(tmp) + sfxextra); 2626| 0| } else { 2627| 0| numsyllable -= short(sfxextra); 2628| 0| } 2629| | 2630| | // + 1 word, if syllable number of the prefix > 1 (hungarian 2631| | // convention) 2632| 0| if (pfx && (get_syllable(pfx->getKey()) > 1)) ------------------ | Branch (2632:15): [True: 0, False: 0] | Branch (2632:15): [True: 0, False: 0] | Branch (2632:22): [True: 0, False: 0] ------------------ 2633| 0| wordnum++; 2634| | 2635| | // increment syllable num, if last word has a SYLLABLENUM flag 2636| | // and the suffix is beginning `s' 2637| | 2638| 0| if (!cpdsyllablenum.empty()) { ------------------ | Branch (2638:15): [True: 0, False: 0] ------------------ 2639| 0| switch (sfxflag) { ------------------ | Branch (2639:21): [True: 0, False: 0] ------------------ 2640| 0| case 'c': { ------------------ | Branch (2640:15): [True: 0, False: 0] ------------------ 2641| 0| numsyllable += 2; 2642| 0| break; 2643| 0| } 2644| 0| case 'J': { ------------------ | Branch (2644:15): [True: 0, False: 0] ------------------ 2645| 0| numsyllable += 1; 2646| 0| break; 2647| 0| } 2648| 0| case 'I': { ------------------ | Branch (2648:15): [True: 0, False: 0] ------------------ 2649| 0| if (rv && TESTAFF(rv->astr, 'J', rv->alen)) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (2649:21): [True: 0, False: 0] | Branch (2649:21): [True: 0, False: 0] ------------------ 2650| 0| numsyllable += 1; 2651| 0| break; 2652| 0| } 2653| 0| } 2654| 0| } 2655| 0| } 2656| | 2657| | // increment word number, if the second word has a compoundroot flag 2658| 0| if ((rv) && (compoundroot) && ------------------ | Branch (2658:13): [True: 0, False: 0] | Branch (2658:21): [True: 0, False: 0] ------------------ 2659| 0| (TESTAFF(rv->astr, compoundroot, rv->alen))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (2659:13): [True: 0, False: 0] ------------------ 2660| 0| wordnum++; 2661| 0| } 2662| | // second word is acceptable, as a word with prefix or/and suffix? 2663| | // hungarian conventions: compounding is acceptable, 2664| | // when compound forms consist 2 word, otherwise 2665| | // the syllable number of root words is 6, or lesser. 2666| 0| if ((rv) && ------------------ | Branch (2666:13): [True: 0, False: 0] ------------------ 2667| 0| (((cpdwordmax == -1) || (wordnum + 1 < cpdwordmax)) || ------------------ | Branch (2667:15): [True: 0, False: 0] | Branch (2667:37): [True: 0, False: 0] ------------------ 2668| 0| ((cpdmaxsyllable != 0) && (numsyllable <= cpdmaxsyllable))) && ------------------ | Branch (2668:15): [True: 0, False: 0] | Branch (2668:40): [True: 0, False: 0] ------------------ 2669| 0| ((!checkcompounddup || (rv != rv_first)))) { ------------------ | Branch (2669:15): [True: 0, False: 0] | Branch (2669:36): [True: 0, False: 0] ------------------ 2670| 0| std::string m; 2671| 0| if (compoundflag) ------------------ | Branch (2671:15): [True: 0, False: 0] ------------------ 2672| 0| m = affix_check_morph(word, i, word.size() - i, scratch, compoundflag); 2673| 0| if (m.empty() && compoundend) { ------------------ | Branch (2673:15): [True: 0, False: 0] | Branch (2673:28): [True: 0, False: 0] ------------------ 2674| 0| m = affix_check_morph(word, i, word.size() - i, scratch, compoundend); 2675| 0| } 2676| 0| result.append(presult); 2677| 0| if (!m.empty()) { ------------------ | Branch (2677:15): [True: 0, False: 0] ------------------ 2678| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 2679| 0| result.append(MORPH_PART); ------------------ | | 116| 0|#define MORPH_PART "pa:" ------------------ 2680| 0| result.append(word, i, word.size()); 2681| 0| line_uniq_app(m, MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 2682| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 2683| 0| result.append(m); 2684| 0| } 2685| 0| result.push_back(MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 2686| 0| ok = 1; 2687| 0| } 2688| | 2689| 0| numsyllable = oldnumsyllable2; 2690| 0| wordnum = oldwordnum2; 2691| | 2692| | // perhaps second word is a compound word (recursive call) 2693| 0| if ((wordnum + 2 < maxwordnum) && (wnum + 1 < maxwordnum) && (ok == 0)) { ------------------ | Branch (2693:13): [True: 0, False: 0] | Branch (2693:43): [True: 0, False: 0] | Branch (2693:70): [True: 0, False: 0] ------------------ 2694| 0| compound_check_morph(word.substr(i), wordnum + 1, 2695| 0| numsyllable, maxwordnum, wnum + 1, words, rwords, 0, 2696| 0| result, &presult, scratch); 2697| 0| } else { 2698| 0| rv = nullptr; 2699| 0| } 2700| 0| } 2701| 30.3k| st[i] = ch; 2702| 30.3k| wordnum = oldwordnum; 2703| 30.3k| numsyllable = oldnumsyllable; 2704| | 2705| 30.3k| } while (!defcpdtable.empty() && oldwordnum == 0 && ------------------ | Branch (2705:14): [True: 0, False: 30.3k] | Branch (2705:38): [True: 0, False: 0] ------------------ 2706| 0| onlycpdrule++ < 1); // end of onlycpd loop ------------------ | Branch (2706:14): [True: 0, False: 0] ------------------ 2707| 30.4k| } 2708| 1.83k| return 0; 2709| 2.06k|} _ZN8AffixMgr12suffix_checkERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEiiiP8PfxEntryR12AffixScratchttc: 2732| 525M| char in_compound) { 2733| 525M| struct hentry* rv = nullptr; 2734| 525M| PfxEntry* ep = ppfx; 2735| | 2736| | // first handle the special case of 0 length suffixes 2737| 525M| SfxEntry* se = sStart[0]; 2738| | 2739| 525M| while (se) { ------------------ | Branch (2739:10): [True: 0, False: 525M] ------------------ 2740| 0| if (!cclass || se->getCont()) { ------------------ | Branch (2740:9): [True: 0, False: 0] | Branch (2740:20): [True: 0, False: 0] ------------------ 2741| | // suffixes are not allowed in beginning of compounds 2742| 0| if ((((in_compound != IN_CPD_BEGIN)) || // && !cclass ------------------ | | 73| 0|#define IN_CPD_BEGIN 1 ------------------ | Branch (2742:12): [True: 0, False: 0] ------------------ 2743| | // except when signed with compoundpermitflag flag 2744| 0| (se->getCont() && compoundpermitflag && ------------------ | Branch (2744:13): [True: 0, False: 0] | Branch (2744:30): [True: 0, False: 0] ------------------ 2745| 0| TESTAFF(se->getCont(), compoundpermitflag, se->getContLen()))) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2746| 0| (!circumfix || ------------------ | Branch (2746:12): [True: 0, False: 0] ------------------ 2747| | // no circumfix flag in prefix and suffix 2748| 0| ((!ppfx || !(ep->getCont()) || ------------------ | Branch (2748:14): [True: 0, False: 0] | Branch (2748:23): [True: 0, False: 0] ------------------ 2749| 0| !TESTAFF(ep->getCont(), circumfix, ep->getContLen())) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (2749:14): [True: 0, False: 0] ------------------ 2750| 0| (!se->getCont() || ------------------ | Branch (2750:14): [True: 0, False: 0] ------------------ 2751| 0| !(TESTAFF(se->getCont(), circumfix, se->getContLen())))) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (2751:14): [True: 0, False: 0] ------------------ 2752| | // circumfix flag in prefix AND suffix 2753| 0| ((ppfx && (ep->getCont()) && ------------------ | Branch (2753:14): [True: 0, False: 0] | Branch (2753:22): [True: 0, False: 0] ------------------ 2754| 0| TESTAFF(ep->getCont(), circumfix, ep->getContLen())) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2755| 0| (se->getCont() && ------------------ | Branch (2755:14): [True: 0, False: 0] ------------------ 2756| 0| (TESTAFF(se->getCont(), circumfix, se->getContLen()))))) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (2756:14): [True: 0, False: 0] ------------------ 2757| | // fogemorpheme 2758| 0| (in_compound || ------------------ | Branch (2758:12): [True: 0, False: 0] ------------------ 2759| 0| !(se->getCont() && ------------------ | Branch (2759:14): [True: 0, False: 0] ------------------ 2760| 0| (TESTAFF(se->getCont(), onlyincompound, se->getContLen())))) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (2760:14): [True: 0, False: 0] ------------------ 2761| | // needaffix on prefix or first suffix 2762| 0| (cclass || ------------------ | Branch (2762:12): [True: 0, False: 0] ------------------ 2763| 0| !(se->getCont() && ------------------ | Branch (2763:14): [True: 0, False: 0] ------------------ 2764| 0| TESTAFF(se->getCont(), needaffix, se->getContLen())) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2765| 0| (ppfx && ------------------ | Branch (2765:13): [True: 0, False: 0] ------------------ 2766| 0| !((ep->getCont()) && ------------------ | Branch (2766:15): [True: 0, False: 0] ------------------ 2767| 0| TESTAFF(ep->getCont(), needaffix, ep->getContLen()))))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2768| 0| rv = se->checkword(word, start, len, sfxopts, ppfx, 2769| 0| (FLAG)cclass, needflag, 2770| 0| (in_compound ? 0 : onlyincompound), ------------------ | Branch (2770:29): [True: 0, False: 0] ------------------ 2771| 0| scratch); 2772| 0| if (rv) { ------------------ | Branch (2772:13): [True: 0, False: 0] ------------------ 2773| 0| sfx = se; // BUG: sfx not stateless 2774| 0| return rv; 2775| 0| } 2776| 0| } 2777| 0| } 2778| 0| se = se->getNext(); 2779| 0| } 2780| | 2781| | // now handle the general case 2782| 525M| if (len == 0) ------------------ | Branch (2782:7): [True: 46.8k, False: 525M] ------------------ 2783| 46.8k| return nullptr; // FULLSTRIP 2784| 525M| unsigned char sp = word[start + len - 1]; 2785| 525M| SfxEntry* sptr = sStart[sp]; 2786| | 2787| 526M| while (sptr) { ------------------ | Branch (2787:10): [True: 378k, False: 525M] ------------------ 2788| 378k| if (isRevSubset(sptr->getKey(), word.c_str() + start + len - 1, len)) { ------------------ | Branch (2788:9): [True: 349k, False: 28.9k] ------------------ 2789| | // suffixes are not allowed in beginning of compounds 2790| 349k| if ((((in_compound != IN_CPD_BEGIN)) || // && !cclass ------------------ | | 73| 349k|#define IN_CPD_BEGIN 1 ------------------ | Branch (2790:12): [True: 65.1k, False: 284k] ------------------ 2791| | // except when signed with compoundpermitflag flag 2792| 284k| (sptr->getCont() && compoundpermitflag && ------------------ | Branch (2792:13): [True: 284k, False: 0] | Branch (2792:32): [True: 0, False: 284k] ------------------ 2793| 0| TESTAFF(sptr->getCont(), compoundpermitflag, ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2794| 284k| sptr->getContLen()))) && 2795| 65.1k| (!circumfix || ------------------ | Branch (2795:12): [True: 65.1k, False: 0] ------------------ 2796| | // no circumfix flag in prefix and suffix 2797| 0| ((!ppfx || !(ep->getCont()) || ------------------ | Branch (2797:14): [True: 0, False: 0] | Branch (2797:23): [True: 0, False: 0] ------------------ 2798| 0| !TESTAFF(ep->getCont(), circumfix, ep->getContLen())) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (2798:14): [True: 0, False: 0] ------------------ 2799| 0| (!sptr->getCont() || ------------------ | Branch (2799:14): [True: 0, False: 0] ------------------ 2800| 0| !(TESTAFF(sptr->getCont(), circumfix, sptr->getContLen())))) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (2800:14): [True: 0, False: 0] ------------------ 2801| | // circumfix flag in prefix AND suffix 2802| 0| ((ppfx && (ep->getCont()) && ------------------ | Branch (2802:14): [True: 0, False: 0] | Branch (2802:22): [True: 0, False: 0] ------------------ 2803| 0| TESTAFF(ep->getCont(), circumfix, ep->getContLen())) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2804| 0| (sptr->getCont() && ------------------ | Branch (2804:14): [True: 0, False: 0] ------------------ 2805| 0| (TESTAFF(sptr->getCont(), circumfix, sptr->getContLen()))))) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (2805:14): [True: 0, False: 0] ------------------ 2806| | // fogemorpheme 2807| 65.1k| (in_compound || ------------------ | Branch (2807:12): [True: 0, False: 65.1k] ------------------ 2808| 65.1k| !((sptr->getCont() && (TESTAFF(sptr->getCont(), onlyincompound, ------------------ | | 99| 32.4k|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (2808:15): [True: 32.4k, False: 32.6k] | Branch (2808:34): [True: 0, False: 32.4k] ------------------ 2809| 32.4k| sptr->getContLen()))))) && 2810| | // needaffix on prefix or first suffix 2811| 65.1k| (cclass || ------------------ | Branch (2811:12): [True: 60, False: 65.0k] ------------------ 2812| 65.0k| !(sptr->getCont() && ------------------ | Branch (2812:14): [True: 32.4k, False: 32.6k] ------------------ 2813| 32.4k| TESTAFF(sptr->getCont(), needaffix, sptr->getContLen())) || ------------------ | | 99| 32.4k|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 32.4k] | | ------------------ ------------------ 2814| 0| (ppfx && ------------------ | Branch (2814:13): [True: 0, False: 0] ------------------ 2815| 0| !((ep->getCont()) && ------------------ | Branch (2815:15): [True: 0, False: 0] ------------------ 2816| 0| TESTAFF(ep->getCont(), needaffix, ep->getContLen()))))) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2817| 65.1k| if (in_compound != IN_CPD_END || ppfx || ------------------ | | 74| 130k|#define IN_CPD_END 2 ------------------ | Branch (2817:13): [True: 65.1k, False: 0] | Branch (2817:42): [True: 0, False: 0] ------------------ 2818| 0| !(sptr->getCont() && ------------------ | Branch (2818:15): [True: 0, False: 0] ------------------ 2819| 65.1k| TESTAFF(sptr->getCont(), onlyincompound, sptr->getContLen()))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2820| 65.1k| rv = sptr->checkword(word, start, len, sfxopts, ppfx, 2821| 65.1k| cclass, needflag, 2822| 65.1k| (in_compound ? 0 : onlyincompound), ------------------ | Branch (2822:33): [True: 0, False: 65.1k] ------------------ 2823| 65.1k| scratch); 2824| 65.1k| if (rv) { ------------------ | Branch (2824:15): [True: 2, False: 65.1k] ------------------ 2825| 2| sfx = sptr; // BUG: sfx not stateless 2826| 2| sfxflag = sptr->getFlag(); // BUG: sfxflag not stateless 2827| 2| if (!sptr->getCont()) ------------------ | Branch (2827:17): [True: 1, False: 1] ------------------ 2828| 1| sfxappnd = sptr->getKey(); // BUG: sfxappnd not stateless 2829| | // LANG_hu section: spec. Hungarian rule 2830| 1| else if (langnum == LANG_hu && sptr->getKeyLen() && ------------------ | Branch (2830:22): [True: 0, False: 1] | Branch (2830:44): [True: 0, False: 0] ------------------ 2831| 0| sptr->getKey()[0] == 'i' && sptr->getKey()[1] != 'y' && ------------------ | Branch (2831:22): [True: 0, False: 0] | Branch (2831:50): [True: 0, False: 0] ------------------ 2832| 0| sptr->getKey()[1] != 't') { ------------------ | Branch (2832:22): [True: 0, False: 0] ------------------ 2833| 0| sfxextra = 1; 2834| 0| } 2835| | // END of LANG_hu section 2836| 2| return rv; 2837| 2| } 2838| 65.1k| } 2839| 349k| sptr = sptr->getNextEQ(); 2840| 349k| } else { 2841| 28.9k| sptr = sptr->getNextNE(); 2842| 28.9k| } 2843| 378k| } 2844| | 2845| 525M| return nullptr; 2846| 525M|} _ZN8AffixMgr19suffix_check_twosfxERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEiiiP8PfxEntryR12AffixScratcht: 2855| 14.7M| const FLAG needflag) { 2856| 14.7M| struct hentry* rv = nullptr; 2857| | 2858| | // first handle the special case of 0 length suffixes 2859| 14.7M| SfxEntry* se = sStart[0]; 2860| 14.7M| while (se) { ------------------ | Branch (2860:10): [True: 0, False: 14.7M] ------------------ 2861| 0| if (contclasses[se->getFlag()]) { ------------------ | Branch (2861:9): [True: 0, False: 0] ------------------ 2862| 0| rv = se->check_twosfx(word, start, len, sfxopts, ppfx, needflag, scratch); 2863| 0| if (rv) ------------------ | Branch (2863:11): [True: 0, False: 0] ------------------ 2864| 0| return rv; 2865| 0| } 2866| 0| se = se->getNext(); 2867| 0| } 2868| | 2869| | // now handle the general case 2870| 14.7M| if (len == 0) ------------------ | Branch (2870:7): [True: 14.8k, False: 14.7M] ------------------ 2871| 14.8k| return nullptr; // FULLSTRIP 2872| 14.7M| unsigned char sp = word[start + len - 1]; 2873| 14.7M| SfxEntry* sptr = sStart[sp]; 2874| | 2875| 14.8M| while (sptr) { ------------------ | Branch (2875:10): [True: 82.8k, False: 14.7M] ------------------ 2876| 82.8k| if (isRevSubset(sptr->getKey(), word.c_str() + start + len - 1, len)) { ------------------ | Branch (2876:9): [True: 58.2k, False: 24.5k] ------------------ 2877| 58.2k| if (contclasses[sptr->getFlag()]) { ------------------ | Branch (2877:11): [True: 25.7k, False: 32.4k] ------------------ 2878| 25.7k| rv = sptr->check_twosfx(word, start, len, sfxopts, ppfx, needflag, scratch); 2879| 25.7k| if (rv) { ------------------ | Branch (2879:13): [True: 0, False: 25.7k] ------------------ 2880| 0| sfxflag = sptr->getFlag(); // BUG: sfxflag not stateless 2881| 0| if (!sptr->getCont()) ------------------ | Branch (2881:15): [True: 0, False: 0] ------------------ 2882| 0| sfxappnd = sptr->getKey(); // BUG: sfxappnd not stateless 2883| 0| return rv; 2884| 0| } 2885| 25.7k| } 2886| 58.2k| sptr = sptr->getNextEQ(); 2887| 58.2k| } else { 2888| 24.5k| sptr = sptr->getNextNE(); 2889| 24.5k| } 2890| 82.8k| } 2891| | 2892| 14.7M| return nullptr; 2893| 14.7M|} _ZN8AffixMgr25suffix_check_twosfx_morphERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEiiiP8PfxEntryR12AffixScratcht: 2902| 3.96k| const FLAG needflag) { 2903| 3.96k| std::string result; 2904| 3.96k| std::string result2; 2905| 3.96k| std::string result3; 2906| | 2907| | // first handle the special case of 0 length suffixes 2908| 3.96k| SfxEntry* se = sStart[0]; 2909| 3.96k| while (se) { ------------------ | Branch (2909:10): [True: 0, False: 3.96k] ------------------ 2910| 0| if (contclasses[se->getFlag()]) { ------------------ | Branch (2910:9): [True: 0, False: 0] ------------------ 2911| 0| std::string st = se->check_twosfx_morph(word, start, len, sfxopts, ppfx, needflag, scratch); 2912| 0| if (!st.empty()) { ------------------ | Branch (2912:11): [True: 0, False: 0] ------------------ 2913| 0| if (ppfx) { ------------------ | Branch (2913:13): [True: 0, False: 0] ------------------ 2914| 0| if (ppfx->getMorph()) { ------------------ | Branch (2914:15): [True: 0, False: 0] ------------------ 2915| 0| result.append(ppfx->getMorph()); 2916| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 2917| 0| } else 2918| 0| debugflag(result, ppfx->getFlag()); 2919| 0| } 2920| 0| result.append(st); 2921| 0| if (se->getMorph()) { ------------------ | Branch (2921:13): [True: 0, False: 0] ------------------ 2922| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 2923| 0| result.append(se->getMorph()); 2924| 0| } else 2925| 0| debugflag(result, se->getFlag()); 2926| 0| result.push_back(MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 2927| 0| } 2928| 0| } 2929| 0| se = se->getNext(); 2930| 0| } 2931| | 2932| | // now handle the general case 2933| 3.96k| if (len == 0) ------------------ | Branch (2933:7): [True: 1, False: 3.96k] ------------------ 2934| 1| return { }; // FULLSTRIP 2935| 3.96k| unsigned char sp = word[start + len - 1]; 2936| 3.96k| SfxEntry* sptr = sStart[sp]; 2937| | 2938| 3.97k| while (sptr) { ------------------ | Branch (2938:10): [True: 7, False: 3.96k] ------------------ 2939| 7| if (isRevSubset(sptr->getKey(), word.c_str() + start + len - 1, len)) { ------------------ | Branch (2939:9): [True: 0, False: 7] ------------------ 2940| 0| if (contclasses[sptr->getFlag()]) { ------------------ | Branch (2940:11): [True: 0, False: 0] ------------------ 2941| 0| std::string st = sptr->check_twosfx_morph(word, start, len, sfxopts, ppfx, needflag, scratch); 2942| 0| if (!st.empty()) { ------------------ | Branch (2942:13): [True: 0, False: 0] ------------------ 2943| 0| sfxflag = sptr->getFlag(); // BUG: sfxflag not stateless 2944| 0| if (!sptr->getCont()) ------------------ | Branch (2944:15): [True: 0, False: 0] ------------------ 2945| 0| sfxappnd = sptr->getKey(); // BUG: sfxappnd not stateless 2946| 0| result2.assign(st); 2947| | 2948| 0| result3.clear(); 2949| | 2950| 0| if (sptr->getMorph()) { ------------------ | Branch (2950:15): [True: 0, False: 0] ------------------ 2951| 0| result3.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 2952| 0| result3.append(sptr->getMorph()); 2953| 0| } else 2954| 0| debugflag(result3, sptr->getFlag()); 2955| 0| strlinecat(result2, result3); 2956| 0| result2.push_back(MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 2957| 0| result.append(result2); 2958| 0| } 2959| 0| } 2960| 0| sptr = sptr->getNextEQ(); 2961| 7| } else { 2962| 7| sptr = sptr->getNextNE(); 2963| 7| } 2964| 7| } 2965| | 2966| 3.96k| return result; 2967| 3.96k|} _ZN8AffixMgr18suffix_check_morphERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEiiiP8PfxEntryR12AffixScratchttc: 2977| 12.7k| char in_compound) { 2978| 12.7k| std::string result; 2979| | 2980| 12.7k| struct hentry* rv = nullptr; 2981| | 2982| 12.7k| PfxEntry* ep = ppfx; 2983| | 2984| | // first handle the special case of 0 length suffixes 2985| 12.7k| SfxEntry* se = sStart[0]; 2986| 12.7k| while (se) { ------------------ | Branch (2986:10): [True: 0, False: 12.7k] ------------------ 2987| 0| if (!cclass || se->getCont()) { ------------------ | Branch (2987:9): [True: 0, False: 0] | Branch (2987:20): [True: 0, False: 0] ------------------ 2988| | // suffixes are not allowed in beginning of compounds 2989| 0| if (((((in_compound != IN_CPD_BEGIN)) || // && !cclass ------------------ | | 73| 0|#define IN_CPD_BEGIN 1 ------------------ | Branch (2989:13): [True: 0, False: 0] ------------------ 2990| | // except when signed with compoundpermitflag flag 2991| 0| (se->getCont() && compoundpermitflag && ------------------ | Branch (2991:14): [True: 0, False: 0] | Branch (2991:31): [True: 0, False: 0] ------------------ 2992| 0| TESTAFF(se->getCont(), compoundpermitflag, se->getContLen()))) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 2993| 0| (!circumfix || ------------------ | Branch (2993:13): [True: 0, False: 0] ------------------ 2994| | // no circumfix flag in prefix and suffix 2995| 0| ((!ppfx || !(ep->getCont()) || ------------------ | Branch (2995:15): [True: 0, False: 0] | Branch (2995:24): [True: 0, False: 0] ------------------ 2996| 0| !TESTAFF(ep->getCont(), circumfix, ep->getContLen())) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (2996:15): [True: 0, False: 0] ------------------ 2997| 0| (!se->getCont() || ------------------ | Branch (2997:15): [True: 0, False: 0] ------------------ 2998| 0| !(TESTAFF(se->getCont(), circumfix, se->getContLen())))) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (2998:15): [True: 0, False: 0] ------------------ 2999| | // circumfix flag in prefix AND suffix 3000| 0| ((ppfx && (ep->getCont()) && ------------------ | Branch (3000:15): [True: 0, False: 0] | Branch (3000:23): [True: 0, False: 0] ------------------ 3001| 0| TESTAFF(ep->getCont(), circumfix, ep->getContLen())) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 3002| 0| (se->getCont() && ------------------ | Branch (3002:15): [True: 0, False: 0] ------------------ 3003| 0| (TESTAFF(se->getCont(), circumfix, se->getContLen()))))) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (3003:15): [True: 0, False: 0] ------------------ 3004| | // fogemorpheme 3005| 0| (in_compound || ------------------ | Branch (3005:13): [True: 0, False: 0] ------------------ 3006| 0| !((se->getCont() && ------------------ | Branch (3006:16): [True: 0, False: 0] ------------------ 3007| 0| (TESTAFF(se->getCont(), onlyincompound, se->getContLen()))))) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (3007:16): [True: 0, False: 0] ------------------ 3008| | // needaffix on prefix or first suffix 3009| 0| (cclass || ------------------ | Branch (3009:13): [True: 0, False: 0] ------------------ 3010| 0| !(se->getCont() && ------------------ | Branch (3010:15): [True: 0, False: 0] ------------------ 3011| 0| TESTAFF(se->getCont(), needaffix, se->getContLen())) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 3012| 0| (ppfx && ------------------ | Branch (3012:14): [True: 0, False: 0] ------------------ 3013| 0| !((ep->getCont()) && ------------------ | Branch (3013:16): [True: 0, False: 0] ------------------ 3014| 0| TESTAFF(ep->getCont(), needaffix, ep->getContLen())))))) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 3015| 0| rv = se->checkword(word, start, len, sfxopts, ppfx, cclass, 3016| 0| needflag, FLAG_NULL, scratch); ------------------ | | 96| 0|#define FLAG_NULL 0x00 ------------------ 3017| 0| while (rv) { ------------------ | Branch (3017:14): [True: 0, False: 0] ------------------ 3018| 0| if (ppfx) { ------------------ | Branch (3018:13): [True: 0, False: 0] ------------------ 3019| 0| if (ppfx->getMorph()) { ------------------ | Branch (3019:15): [True: 0, False: 0] ------------------ 3020| 0| result.append(ppfx->getMorph()); 3021| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 3022| 0| } else 3023| 0| debugflag(result, ppfx->getFlag()); 3024| 0| } 3025| 0| if (complexprefixes && HENTRY_DATA(rv)) ------------------ | Branch (3025:13): [True: 0, False: 0] | Branch (3025:32): [True: 0, False: 0] ------------------ 3026| 0| result.append(HENTRY_DATA2(rv)); 3027| 0| if (!HENTRY_FIND(rv, MORPH_STEM)) { ------------------ | | 103| 0|#define MORPH_STEM "st:" ------------------ | Branch (3027:13): [True: 0, False: 0] ------------------ 3028| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 3029| 0| result.append(MORPH_STEM); ------------------ | | 103| 0|#define MORPH_STEM "st:" ------------------ 3030| 0| result.append(HENTRY_WORD(rv)); ------------------ | | 54| 0|#define HENTRY_WORD(h) &(h->word[0]) ------------------ 3031| 0| } 3032| | 3033| 0| if (!complexprefixes && HENTRY_DATA(rv)) { ------------------ | Branch (3033:13): [True: 0, False: 0] | Branch (3033:33): [True: 0, False: 0] ------------------ 3034| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 3035| 0| result.append(HENTRY_DATA2(rv)); 3036| 0| } 3037| 0| if (se->getMorph()) { ------------------ | Branch (3037:13): [True: 0, False: 0] ------------------ 3038| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 3039| 0| result.append(se->getMorph()); 3040| 0| } else 3041| 0| debugflag(result, se->getFlag()); 3042| 0| result.push_back(MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 3043| 0| rv = se->get_next_homonym(rv, sfxopts, ppfx, cclass, needflag); 3044| 0| } 3045| 0| } 3046| 0| se = se->getNext(); 3047| 0| } 3048| | 3049| | // now handle the general case 3050| 12.7k| if (len == 0) ------------------ | Branch (3050:7): [True: 1, False: 12.7k] ------------------ 3051| 1| return { }; // FULLSTRIP 3052| 12.7k| unsigned char sp = word[start + len - 1]; 3053| 12.7k| SfxEntry* sptr = sStart[sp]; 3054| | 3055| 12.7k| while (sptr) { ------------------ | Branch (3055:10): [True: 7, False: 12.7k] ------------------ 3056| 7| if (isRevSubset(sptr->getKey(), word.c_str() + start + len - 1, len)) { ------------------ | Branch (3056:9): [True: 0, False: 7] ------------------ 3057| | // suffixes are not allowed in beginning of compounds 3058| 0| if (((((in_compound != IN_CPD_BEGIN)) || // && !cclass ------------------ | | 73| 0|#define IN_CPD_BEGIN 1 ------------------ | Branch (3058:13): [True: 0, False: 0] ------------------ 3059| | // except when signed with compoundpermitflag flag 3060| 0| (sptr->getCont() && compoundpermitflag && ------------------ | Branch (3060:14): [True: 0, False: 0] | Branch (3060:33): [True: 0, False: 0] ------------------ 3061| 0| TESTAFF(sptr->getCont(), compoundpermitflag, ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 3062| 0| sptr->getContLen()))) && 3063| 0| (!circumfix || ------------------ | Branch (3063:13): [True: 0, False: 0] ------------------ 3064| | // no circumfix flag in prefix and suffix 3065| 0| ((!ppfx || !(ep->getCont()) || ------------------ | Branch (3065:15): [True: 0, False: 0] | Branch (3065:24): [True: 0, False: 0] ------------------ 3066| 0| !TESTAFF(ep->getCont(), circumfix, ep->getContLen())) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (3066:15): [True: 0, False: 0] ------------------ 3067| 0| (!sptr->getCont() || ------------------ | Branch (3067:15): [True: 0, False: 0] ------------------ 3068| 0| !(TESTAFF(sptr->getCont(), circumfix, sptr->getContLen())))) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (3068:15): [True: 0, False: 0] ------------------ 3069| | // circumfix flag in prefix AND suffix 3070| 0| ((ppfx && (ep->getCont()) && ------------------ | Branch (3070:15): [True: 0, False: 0] | Branch (3070:23): [True: 0, False: 0] ------------------ 3071| 0| TESTAFF(ep->getCont(), circumfix, ep->getContLen())) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 3072| 0| (sptr->getCont() && ------------------ | Branch (3072:15): [True: 0, False: 0] ------------------ 3073| 0| (TESTAFF(sptr->getCont(), circumfix, sptr->getContLen()))))) && ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (3073:15): [True: 0, False: 0] ------------------ 3074| | // fogemorpheme 3075| 0| (in_compound || ------------------ | Branch (3075:13): [True: 0, False: 0] ------------------ 3076| 0| !((sptr->getCont() && (TESTAFF(sptr->getCont(), onlyincompound, ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) ------------------ | Branch (3076:16): [True: 0, False: 0] | Branch (3076:35): [True: 0, False: 0] ------------------ 3077| 0| sptr->getContLen()))))) && 3078| | // needaffix on first suffix 3079| 0| (cclass || ------------------ | Branch (3079:13): [True: 0, False: 0] ------------------ 3080| 0| !(sptr->getCont() && ------------------ | Branch (3080:15): [True: 0, False: 0] ------------------ 3081| 0| TESTAFF(sptr->getCont(), needaffix, sptr->getContLen()))))) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 3082| 0| rv = sptr->checkword(word, start, len, sfxopts, ppfx, cclass, 3083| 0| needflag, FLAG_NULL, scratch); ------------------ | | 96| 0|#define FLAG_NULL 0x00 ------------------ 3084| 0| while (rv) { ------------------ | Branch (3084:14): [True: 0, False: 0] ------------------ 3085| 0| if (ppfx) { ------------------ | Branch (3085:13): [True: 0, False: 0] ------------------ 3086| 0| if (ppfx->getMorph()) { ------------------ | Branch (3086:15): [True: 0, False: 0] ------------------ 3087| 0| result.append(ppfx->getMorph()); 3088| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 3089| 0| } else 3090| 0| debugflag(result, ppfx->getFlag()); 3091| 0| } 3092| 0| if (complexprefixes && HENTRY_DATA(rv)) ------------------ | Branch (3092:13): [True: 0, False: 0] | Branch (3092:32): [True: 0, False: 0] ------------------ 3093| 0| result.append(HENTRY_DATA2(rv)); 3094| 0| if (!HENTRY_FIND(rv, MORPH_STEM)) { ------------------ | | 103| 0|#define MORPH_STEM "st:" ------------------ | Branch (3094:13): [True: 0, False: 0] ------------------ 3095| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 3096| 0| result.append(MORPH_STEM); ------------------ | | 103| 0|#define MORPH_STEM "st:" ------------------ 3097| 0| result.append(HENTRY_WORD(rv)); ------------------ | | 54| 0|#define HENTRY_WORD(h) &(h->word[0]) ------------------ 3098| 0| } 3099| | 3100| 0| if (!complexprefixes && HENTRY_DATA(rv)) { ------------------ | Branch (3100:13): [True: 0, False: 0] | Branch (3100:33): [True: 0, False: 0] ------------------ 3101| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 3102| 0| result.append(HENTRY_DATA2(rv)); 3103| 0| } 3104| | 3105| 0| if (sptr->getMorph()) { ------------------ | Branch (3105:13): [True: 0, False: 0] ------------------ 3106| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 3107| 0| result.append(sptr->getMorph()); 3108| 0| } else 3109| 0| debugflag(result, sptr->getFlag()); 3110| 0| result.push_back(MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 3111| 0| rv = sptr->get_next_homonym(rv, sfxopts, ppfx, cclass, needflag); 3112| 0| } 3113| 0| sptr = sptr->getNextEQ(); 3114| 7| } else { 3115| 7| sptr = sptr->getNextNE(); 3116| 7| } 3117| 7| } 3118| | 3119| 12.7k| return result; 3120| 12.7k|} _ZN8AffixMgr11affix_checkERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEiiR12AffixScratchtc: 3128| 11.5M| char in_compound) { 3129| | 3130| | // check all prefixes (also crossed with suffixes if allowed) 3131| 11.5M| struct hentry* rv = prefix_check(word, start, len, in_compound, scratch, needflag); 3132| 11.5M| if (rv) ------------------ | Branch (3132:7): [True: 0, False: 11.5M] ------------------ 3133| 0| return rv; 3134| | 3135| | // if still not found check all suffixes 3136| 11.5M| rv = suffix_check(word, start, len, 0, nullptr, scratch, FLAG_NULL, needflag, in_compound); ------------------ | | 96| 11.5M|#define FLAG_NULL 0x00 ------------------ 3137| | 3138| 11.5M| if (havecontclass) { ------------------ | Branch (3138:7): [True: 3.98M, False: 7.60M] ------------------ 3139| 3.98M| sfx = nullptr; 3140| 3.98M| pfx = nullptr; 3141| | 3142| 3.98M| if (rv) ------------------ | Branch (3142:9): [True: 0, False: 3.98M] ------------------ 3143| 0| return rv; 3144| | // if still not found check all two-level suffixes 3145| 3.98M| rv = suffix_check_twosfx(word, start, len, 0, nullptr, scratch, needflag); 3146| | 3147| 3.98M| if (rv) ------------------ | Branch (3147:9): [True: 0, False: 3.98M] ------------------ 3148| 0| return rv; 3149| | // if still not found check all two-level suffixes 3150| 3.98M| rv = prefix_check_twosfx(word, start, len, IN_CPD_NOT, scratch, needflag); ------------------ | | 72| 3.98M|#define IN_CPD_NOT 0 ------------------ 3151| 3.98M| } 3152| | 3153| 11.5M| return rv; 3154| 11.5M|} _ZN8AffixMgr17affix_check_morphERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEiiR12AffixScratchtc: 3162| 12.7k| char in_compound) { 3163| 12.7k| std::string result; 3164| | 3165| | // check all prefixes (also crossed with suffixes if allowed) 3166| 12.7k| std::string st = prefix_check_morph(word, start, len, in_compound, scratch); 3167| 12.7k| if (!st.empty()) { ------------------ | Branch (3167:7): [True: 0, False: 12.7k] ------------------ 3168| 0| result.append(st); 3169| 0| } 3170| | 3171| | // if still not found check all suffixes 3172| 12.7k| st = suffix_check_morph(word, start, len, 0, nullptr, scratch, '\0', needflag, in_compound); 3173| 12.7k| if (!st.empty()) { ------------------ | Branch (3173:7): [True: 0, False: 12.7k] ------------------ 3174| 0| result.append(st); 3175| 0| } 3176| | 3177| 12.7k| if (havecontclass) { ------------------ | Branch (3177:7): [True: 3.96k, False: 8.77k] ------------------ 3178| 3.96k| sfx = nullptr; 3179| 3.96k| pfx = nullptr; 3180| | // if still not found check all two-level suffixes 3181| 3.96k| st = suffix_check_twosfx_morph(word, start, len, 0, nullptr, scratch, needflag); 3182| 3.96k| if (!st.empty()) { ------------------ | Branch (3182:9): [True: 0, False: 3.96k] ------------------ 3183| 0| result.append(st); 3184| 0| } 3185| | 3186| | // if still not found check all two-level suffixes 3187| 3.96k| st = prefix_check_twosfx_morph(word, start, len, IN_CPD_NOT, scratch, needflag); ------------------ | | 72| 3.96k|#define IN_CPD_NOT 0 ------------------ 3188| 3.96k| if (!st.empty()) { ------------------ | Branch (3188:9): [True: 0, False: 3.96k] ------------------ 3189| 0| result.append(st); 3190| 0| } 3191| 3.96k| } 3192| | 3193| 12.7k| return result; 3194| 12.7k|} _ZN8AffixMgr15expand_rootwordEP9guesswordiPKciPKttS3_iS3_: 3382| 621k| const char* phon) { 3383| 621k| int nh = 0; 3384| | // first add root word to list 3385| 621k| if ((nh < maxn) && ------------------ | Branch (3385:7): [True: 621k, False: 0] ------------------ 3386| 621k| !(al && ((needaffix && TESTAFF(ap, needaffix, al)) || ------------------ | | 99| 4.66k|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 2.33k, False: 2.33k] | | ------------------ ------------------ | Branch (3386:9): [True: 98.8k, False: 522k] | Branch (3386:17): [True: 4.66k, False: 94.1k] ------------------ 3387| 618k| (onlyincompound && TESTAFF(ap, onlyincompound, al))))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (3387:17): [True: 0, False: 96.5k] ------------------ 3388| 618k| wlst[nh].word = mystrdup(ts); 3389| 618k| wlst[nh].allow = false; 3390| 618k| wlst[nh].orig = nullptr; 3391| 618k| nh++; 3392| | // add special phonetic version 3393| 618k| if (phon && (nh < maxn)) { ------------------ | Branch (3393:9): [True: 826, False: 617k] | Branch (3393:17): [True: 826, False: 0] ------------------ 3394| 826| wlst[nh].word = mystrdup(phon); 3395| 826| wlst[nh].allow = false; 3396| 826| wlst[nh].orig = mystrdup(ts); 3397| 826| nh++; 3398| 826| } 3399| 618k| } 3400| | 3401| | // handle suffixes 3402| 732k| for (int i = 0; i < al; i++) { ------------------ | Branch (3402:19): [True: 111k, False: 621k] ------------------ 3403| 111k| const auto c = (unsigned char)(ap[i] & 0x00FF); 3404| 111k| SfxEntry* sptr = sFlag[c]; 3405| 176k| while (sptr) { ------------------ | Branch (3405:12): [True: 65.5k, False: 111k] ------------------ 3406| 65.5k| if ((sptr->getFlag() == ap[i]) && ------------------ | Branch (3406:11): [True: 65.5k, False: 0] ------------------ 3407| 65.5k| (!sptr->getKeyLen() || ------------------ | Branch (3407:12): [True: 0, False: 65.5k] ------------------ 3408| 65.5k| ((badl > sptr->getKeyLen()) && ------------------ | Branch (3408:13): [True: 37.1k, False: 28.4k] ------------------ 3409| 37.1k| (strcmp(sptr->getAffix(), bad + badl - sptr->getKeyLen()) == 0))) && ------------------ | Branch (3409:13): [True: 53, False: 37.0k] ------------------ 3410| | // check needaffix flag 3411| 53| !(sptr->getCont() && ------------------ | Branch (3411:13): [True: 26, False: 27] ------------------ 3412| 26| ((needaffix && ------------------ | Branch (3412:15): [True: 15, False: 11] ------------------ 3413| 15| TESTAFF(sptr->getCont(), needaffix, sptr->getContLen())) || ------------------ | | 99| 15|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 15] | | ------------------ ------------------ 3414| 26| (circumfix && ------------------ | Branch (3414:15): [True: 0, False: 26] ------------------ 3415| 0| TESTAFF(sptr->getCont(), circumfix, sptr->getContLen())) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 3416| 26| (onlyincompound && ------------------ | Branch (3416:15): [True: 0, False: 26] ------------------ 3417| 53| TESTAFF(sptr->getCont(), onlyincompound, sptr->getContLen()))))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 3418| 53| std::string newword = sptr->add(ts, wl); 3419| 53| if (!newword.empty()) { ------------------ | Branch (3419:13): [True: 53, False: 0] ------------------ 3420| 53| if (nh < maxn) { ------------------ | Branch (3420:15): [True: 53, False: 0] ------------------ 3421| 53| wlst[nh].word = mystrdup(newword.c_str()); 3422| 53| wlst[nh].allow = sptr->allowCross(); 3423| 53| wlst[nh].orig = nullptr; 3424| 53| nh++; 3425| | // add special phonetic version 3426| 53| if (phon && (nh < maxn)) { ------------------ | Branch (3426:17): [True: 0, False: 53] | Branch (3426:25): [True: 0, False: 0] ------------------ 3427| 0| std::string prefix(phon); 3428| 0| std::string key(sptr->getKey()); 3429| 0| reverseword(key); 3430| 0| prefix.append(key); 3431| 0| wlst[nh].word = mystrdup(prefix.c_str()); 3432| 0| wlst[nh].allow = false; 3433| 0| wlst[nh].orig = mystrdup(newword.c_str()); 3434| 0| nh++; 3435| 0| } 3436| 53| } 3437| 53| } 3438| 53| } 3439| 65.5k| sptr = sptr->getFlgNxt(); 3440| 65.5k| } 3441| 111k| } 3442| | 3443| 621k| int n = nh; 3444| | 3445| | // handle cross products of prefixes and suffixes 3446| 622k| for (int j = 1; j < n; j++) ------------------ | Branch (3446:19): [True: 864, False: 621k] ------------------ 3447| 864| if (wlst[j].allow) { ------------------ | Branch (3447:9): [True: 38, False: 826] ------------------ 3448| 114| for (int k = 0; k < al; k++) { ------------------ | Branch (3448:23): [True: 76, False: 38] ------------------ 3449| 76| const auto c = (unsigned char)(ap[k] & 0x00FF); 3450| 76| PfxEntry* cptr = pFlag[c]; 3451| 76| while (cptr) { ------------------ | Branch (3451:16): [True: 0, False: 76] ------------------ 3452| 0| if ((cptr->getFlag() == ap[k]) && cptr->allowCross() && ------------------ | Branch (3452:15): [True: 0, False: 0] | Branch (3452:45): [True: 0, False: 0] ------------------ 3453| 0| (!cptr->getKeyLen() || ------------------ | Branch (3453:16): [True: 0, False: 0] ------------------ 3454| 0| ((badl > cptr->getKeyLen()) && ------------------ | Branch (3454:17): [True: 0, False: 0] ------------------ 3455| 0| (strncmp(cptr->getKey(), bad, cptr->getKeyLen()) == 0)))) { ------------------ | Branch (3455:17): [True: 0, False: 0] ------------------ 3456| 0| int l1 = strlen(wlst[j].word); 3457| 0| std::string newword = cptr->add(wlst[j].word, l1); 3458| 0| if (!newword.empty()) { ------------------ | Branch (3458:17): [True: 0, False: 0] ------------------ 3459| 0| if (nh < maxn) { ------------------ | Branch (3459:19): [True: 0, False: 0] ------------------ 3460| 0| wlst[nh].word = mystrdup(newword.c_str()); 3461| 0| wlst[nh].allow = cptr->allowCross(); 3462| 0| wlst[nh].orig = nullptr; 3463| 0| nh++; 3464| 0| } 3465| 0| } 3466| 0| } 3467| 0| cptr = cptr->getFlgNxt(); 3468| 0| } 3469| 76| } 3470| 38| } 3471| | 3472| | // now handle pure prefixes 3473| 732k| for (int m = 0; m < al; m++) { ------------------ | Branch (3473:19): [True: 111k, False: 621k] ------------------ 3474| 111k| const auto c = (unsigned char)(ap[m] & 0x00FF); 3475| 111k| PfxEntry* ptr = pFlag[c]; 3476| 112k| while (ptr) { ------------------ | Branch (3476:12): [True: 840, False: 111k] ------------------ 3477| 840| if ((ptr->getFlag() == ap[m]) && ------------------ | Branch (3477:11): [True: 840, False: 0] ------------------ 3478| 840| (!ptr->getKeyLen() || ------------------ | Branch (3478:12): [True: 0, False: 840] ------------------ 3479| 840| ((badl > ptr->getKeyLen()) && ------------------ | Branch (3479:13): [True: 840, False: 0] ------------------ 3480| 840| (strncmp(ptr->getKey(), bad, ptr->getKeyLen()) == 0))) && ------------------ | Branch (3480:13): [True: 1, False: 839] ------------------ 3481| | // check needaffix flag 3482| 1| !(ptr->getCont() && ------------------ | Branch (3482:13): [True: 0, False: 1] ------------------ 3483| 0| ((needaffix && ------------------ | Branch (3483:15): [True: 0, False: 0] ------------------ 3484| 0| TESTAFF(ptr->getCont(), needaffix, ptr->getContLen())) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 3485| 0| (circumfix && ------------------ | Branch (3485:15): [True: 0, False: 0] ------------------ 3486| 0| TESTAFF(ptr->getCont(), circumfix, ptr->getContLen())) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 3487| 0| (onlyincompound && ------------------ | Branch (3487:15): [True: 0, False: 0] ------------------ 3488| 1| TESTAFF(ptr->getCont(), onlyincompound, ptr->getContLen()))))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 3489| 1| std::string newword = ptr->add(ts, wl); 3490| 1| if (!newword.empty()) { ------------------ | Branch (3490:13): [True: 1, False: 0] ------------------ 3491| 1| if (nh < maxn) { ------------------ | Branch (3491:15): [True: 1, False: 0] ------------------ 3492| 1| wlst[nh].word = mystrdup(newword.c_str()); 3493| 1| wlst[nh].allow = ptr->allowCross(); 3494| 1| wlst[nh].orig = nullptr; 3495| 1| nh++; 3496| 1| } 3497| 1| } 3498| 1| } 3499| 840| ptr = ptr->getFlgNxt(); 3500| 840| } 3501| 111k| } 3502| | 3503| 621k| return nh; 3504| 621k|} _ZNK8AffixMgr12get_reptableEv: 3507| 376k|const std::vector& AffixMgr::get_reptable() const { 3508| 376k| return pHMgr->get_reptable(); 3509| 376k|} _ZNK8AffixMgr14get_iconvtableEv: 3512| 10.3M|RepList* AffixMgr::get_iconvtable() const { 3513| 10.3M| if (!iconvtable) ------------------ | Branch (3513:7): [True: 10.0M, False: 344k] ------------------ 3514| 10.0M| return nullptr; 3515| 344k| return iconvtable; 3516| 10.3M|} _ZNK8AffixMgr14get_oconvtableEv: 3519| 98.5k|RepList* AffixMgr::get_oconvtable() const { 3520| 98.5k| if (!oconvtable) ------------------ | Branch (3520:7): [True: 93.2k, False: 5.25k] ------------------ 3521| 93.2k| return nullptr; 3522| 5.25k| return oconvtable; 3523| 98.5k|} _ZNK8AffixMgr14get_phonetableEv: 3526| 62.3k|struct phonetable* AffixMgr::get_phonetable() const { 3527| 62.3k| if (!phone) ------------------ | Branch (3527:7): [True: 58.7k, False: 3.54k] ------------------ 3528| 58.7k| return nullptr; 3529| 3.54k| return phone; 3530| 62.3k|} _ZNK8AffixMgr12get_maptableEv: 3533| 376k|const std::vector& AffixMgr::get_maptable() const { 3534| 376k| return maptable; 3535| 376k|} _ZNK8AffixMgr14get_breaktableEv: 3538| 38|const std::vector& AffixMgr::get_breaktable() const { 3539| 38| return breaktable; 3540| 38|} _ZN8AffixMgr12get_encodingEv: 3543| 78|const std::string& AffixMgr::get_encoding() { 3544| 78| if (encoding.empty()) ------------------ | Branch (3544:7): [True: 20, False: 58] ------------------ 3545| 20| encoding = SPELL_ENCODING; ------------------ | | 99| 20|#define SPELL_ENCODING "ISO8859-1" ------------------ 3546| 78| return encoding; 3547| 78|} _ZNK8AffixMgr11get_langnumEv: 3550| 76|int AffixMgr::get_langnum() const { 3551| 76| return langnum; 3552| 76|} _ZNK8AffixMgr19get_complexprefixesEv: 3555| 76|int AffixMgr::get_complexprefixes() const { 3556| 76| return complexprefixes; 3557| 76|} _ZNK8AffixMgr13get_fullstripEv: 3560| 7.81k|int AffixMgr::get_fullstrip() const { 3561| 7.81k| return fullstrip; 3562| 7.81k|} _ZNK8AffixMgr12get_keepcaseEv: 3564| 102k|FLAG AffixMgr::get_keepcase() const { 3565| 102k| return keepcase; 3566| 102k|} _ZNK8AffixMgr14get_forceucaseEv: 3568| 20.5k|FLAG AffixMgr::get_forceucase() const { 3569| 20.5k| return forceucase; 3570| 20.5k|} _ZNK8AffixMgr8get_warnEv: 3572| 27.8k|FLAG AffixMgr::get_warn() const { 3573| 27.8k| return warn; 3574| 27.8k|} _ZNK8AffixMgr14get_forbidwarnEv: 3576| 4|int AffixMgr::get_forbidwarn() const { 3577| 4| return forbidwarn; 3578| 4|} _ZNK8AffixMgr15get_checksharpsEv: 3580| 196k|int AffixMgr::get_checksharps() const { 3581| 196k| return checksharps; 3582| 196k|} _ZNK8AffixMgr10get_ignoreEv: 3589| 21.8M|const char* AffixMgr::get_ignore() const { 3590| 21.8M| if (ignorechars.empty()) ------------------ | Branch (3590:7): [True: 19.6M, False: 2.17M] ------------------ 3591| 19.6M| return nullptr; 3592| 2.17M| return ignorechars.c_str(); 3593| 21.8M|} _ZNK8AffixMgr16get_ignore_utf16Ev: 3596| 512k|const std::vector& AffixMgr::get_ignore_utf16() const { 3597| 512k| return ignorechars_utf16; 3598| 512k|} _ZN8AffixMgr14get_key_stringEv: 3601| 38|const std::string& AffixMgr::get_key_string() { 3602| 38| if (keystring.empty()) ------------------ | Branch (3602:7): [True: 38, False: 0] ------------------ 3603| 38| keystring = SPELL_KEYSTRING; ------------------ | | 100| 38|#define SPELL_KEYSTRING "qwertyuiop|asdfghjkl|zxcvbnm" ------------------ 3604| 38| return keystring; 3605| 38|} _ZNK8AffixMgr14get_try_stringEv: 3608| 38|const std::string& AffixMgr::get_try_string() const { 3609| 38| return trystring; 3610| 38|} _ZNK8AffixMgr12get_compoundEv: 3622| 67.9M|int AffixMgr::get_compound() const { 3623| 67.9M| return compoundflag || compoundbegin || !defcpdtable.empty(); ------------------ | Branch (3623:10): [True: 17.0M, False: 50.8M] | Branch (3623:26): [True: 0, False: 50.8M] | Branch (3623:43): [True: 0, False: 50.8M] ------------------ 3624| 67.9M|} _ZNK8AffixMgr16get_compoundflagEv: 3627| 39.2k|FLAG AffixMgr::get_compoundflag() const { 3628| 39.2k| return compoundflag; 3629| 39.2k|} _ZNK8AffixMgr17get_forbiddenwordEv: 3632| 161k|FLAG AffixMgr::get_forbiddenword() const { 3633| 161k| return forbiddenword; 3634| 161k|} _ZNK8AffixMgr13get_nosuggestEv: 3637| 62.7k|FLAG AffixMgr::get_nosuggest() const { 3638| 62.7k| return nosuggest; 3639| 62.7k|} _ZNK8AffixMgr18get_nongramsuggestEv: 3642| 62.3k|FLAG AffixMgr::get_nongramsuggest() const { 3643| 62.3k| return nongramsuggest; 3644| 62.3k|} _ZNK8AffixMgr15get_substandardEv: 3647| 372|FLAG AffixMgr::get_substandard() const { 3648| 372| return substandard; 3649| 372|} _ZNK8AffixMgr13get_needaffixEv: 3652| 2.89k|FLAG AffixMgr::get_needaffix() const { 3653| 2.89k| return needaffix; 3654| 2.89k|} _ZNK8AffixMgr18get_onlyincompoundEv: 3657| 64.7k|FLAG AffixMgr::get_onlyincompound() const { 3658| 64.7k| return onlyincompound; 3659| 64.7k|} _ZN8AffixMgr6lookupEPKcm: 3667| 514M|struct hentry* AffixMgr::lookup(const char* word, size_t len) { 3668| 514M| struct hentry* he = nullptr; 3669| 1.02G| for (size_t i = 0; i < alldic.size() && !he; ++i) { ------------------ | Branch (3669:22): [True: 514M, False: 514M] | Branch (3669:43): [True: 514M, False: 0] ------------------ 3670| 514M| he = alldic[i]->lookup(word, len); 3671| 514M| } 3672| 514M| return he; 3673| 514M|} _ZNK8AffixMgr14have_contclassEv: 3676| 29.6M|int AffixMgr::have_contclass() const { 3677| 29.6M| return havecontclass; 3678| 29.6M|} _ZNK8AffixMgr8get_utf8Ev: 3681| 76|int AffixMgr::get_utf8() const { 3682| 76| return utf8; 3683| 76|} _ZNK8AffixMgr16get_maxngramsugsEv: 3685| 69.9k|int AffixMgr::get_maxngramsugs() const { 3686| 69.9k| return maxngramsugs; 3687| 69.9k|} _ZNK8AffixMgr14get_maxcpdsugsEv: 3689| 38|int AffixMgr::get_maxcpdsugs() const { 3690| 38| return maxcpdsugs; 3691| 38|} _ZNK8AffixMgr11get_maxdiffEv: 3693| 48.4k|int AffixMgr::get_maxdiff() const { 3694| 48.4k| return maxdiff; 3695| 48.4k|} _ZNK8AffixMgr15get_onlymaxdiffEv: 3697| 7.56k|int AffixMgr::get_onlymaxdiff() const { 3698| 7.56k| return onlymaxdiff; 3699| 7.56k|} _ZNK8AffixMgr15get_nosplitsugsEv: 3702| 38|int AffixMgr::get_nosplitsugs() const { 3703| 38| return nosplitsugs; 3704| 38|} _ZNK8AffixMgr16get_sugswithdotsEv: 3707| 3.70k|int AffixMgr::get_sugswithdots() const { 3708| 3.70k| return sugswithdots; 3709| 3.70k|} _ZN8AffixMgr10parse_flagERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEPtP7FileMgr: 3712| 22|bool AffixMgr::parse_flag(const std::string& line, unsigned short* out, FileMgr* af) { 3713| 22| if (*out != FLAG_NULL && !(*out >= DEFAULTFLAGS)) { ------------------ | | 96| 44|#define FLAG_NULL 0x00 ------------------ if (*out != FLAG_NULL && !(*out >= DEFAULTFLAGS)) { ------------------ | | 126| 2|#define DEFAULTFLAGS 65510 ------------------ | Branch (3713:7): [True: 2, False: 20] | Branch (3713:28): [True: 0, False: 2] ------------------ 3714| 0| HUNSPELL_WARNING( 3715| 0| stderr, 3716| 0| "error: line %d: multiple definitions of an affix file parameter\n", 3717| 0| af->getlinenum()); 3718| 0| return false; 3719| 0| } 3720| 22| std::string s; 3721| 22| if (!parse_string(line, s, af->getlinenum())) ------------------ | Branch (3721:7): [True: 0, False: 22] ------------------ 3722| 0| return false; 3723| 22| *out = pHMgr->decode_flag(s); 3724| 22| return true; 3725| 22|} _ZN8AffixMgr9parse_numERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEPiP7FileMgr: 3728| 10|bool AffixMgr::parse_num(const std::string& line, int* out, FileMgr* af) { 3729| 10| if (*out != -1) { ------------------ | Branch (3729:7): [True: 0, False: 10] ------------------ 3730| 0| HUNSPELL_WARNING( 3731| 0| stderr, 3732| 0| "error: line %d: multiple definitions of an affix file parameter\n", 3733| 0| af->getlinenum()); 3734| 0| return false; 3735| 0| } 3736| 10| std::string s; 3737| 10| if (!parse_string(line, s, af->getlinenum())) ------------------ | Branch (3737:7): [True: 0, False: 10] ------------------ 3738| 0| return false; 3739| 10| *out = atoi(s.c_str()); 3740| 10| return true; 3741| 10|} _ZN8AffixMgr17parse_cpdsyllableERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEP7FileMgr: 3744| 2|bool AffixMgr::parse_cpdsyllable(const std::string& line, FileMgr* af) { 3745| 2| int i = 0; 3746| 2| int np = 0; 3747| 2| auto iter = line.begin(), start_piece = mystrsep(line, iter); 3748| 8| while (start_piece != line.end()) { ------------------ | Branch (3748:10): [True: 6, False: 2] ------------------ 3749| 6| switch (i) { 3750| 2| case 0: { ------------------ | Branch (3750:7): [True: 2, False: 4] ------------------ 3751| 2| np++; 3752| 2| break; 3753| 0| } 3754| 2| case 1: { ------------------ | Branch (3754:7): [True: 2, False: 4] ------------------ 3755| 2| cpdmaxsyllable = atoi(std::string(start_piece, iter).c_str()); 3756| 2| np++; 3757| 2| break; 3758| 0| } 3759| 2| case 2: { ------------------ | Branch (3759:7): [True: 2, False: 4] ------------------ 3760| 2| if (!utf8) { ------------------ | Branch (3760:13): [True: 0, False: 2] ------------------ 3761| 0| cpdvowels.assign(start_piece, iter); 3762| 0| std::sort(cpdvowels.begin(), cpdvowels.end()); 3763| 2| } else { 3764| 2| std::string piece(start_piece, iter); 3765| 2| u8_u16(cpdvowels_utf16, piece); 3766| 2| std::sort(cpdvowels_utf16.begin(), cpdvowels_utf16.end()); 3767| 2| } 3768| 2| np++; 3769| 2| break; 3770| 0| } 3771| 0| default: ------------------ | Branch (3771:7): [True: 0, False: 6] ------------------ 3772| 0| break; 3773| 6| } 3774| 6| ++i; 3775| 6| start_piece = mystrsep(line, iter); 3776| 6| } 3777| 2| if (np < 2) { ------------------ | Branch (3777:7): [True: 0, False: 2] ------------------ 3778| 0| HUNSPELL_WARNING(stderr, 3779| 0| "error: line %d: missing compoundsyllable information\n", 3780| 0| af->getlinenum()); 3781| 0| return false; 3782| 0| } 3783| 2| if (np == 2) ------------------ | Branch (3783:7): [True: 0, False: 2] ------------------ 3784| 0| cpdvowels = "AEIOUaeiou"; 3785| 2| return true; 3786| 2|} _ZN8AffixMgr15parse_convtableERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEP7FileMgrPP7RepListS8_: 3791| 4| const std::string& keyword) { 3792| 4| if (*rl) { ------------------ | Branch (3792:7): [True: 0, False: 4] ------------------ 3793| 0| HUNSPELL_WARNING(stderr, "error: line %d: multiple table definitions\n", 3794| 0| af->getlinenum()); 3795| 0| return false; 3796| 0| } 3797| 4| int i = 0; 3798| 4| int np = 0; 3799| 4| int numrl = 0; 3800| 4| auto iter = line.begin(), start_piece = mystrsep(line, iter); 3801| 18| while (start_piece != line.end()) { ------------------ | Branch (3801:10): [True: 14, False: 4] ------------------ 3802| 14| switch (i) { 3803| 4| case 0: { ------------------ | Branch (3803:7): [True: 4, False: 10] ------------------ 3804| 4| np++; 3805| 4| break; 3806| 0| } 3807| 4| case 1: { ------------------ | Branch (3807:7): [True: 4, False: 10] ------------------ 3808| 4| numrl = atoi(std::string(start_piece, iter).c_str()); 3809| 4| if (numrl < 1) { ------------------ | Branch (3809:13): [True: 0, False: 4] ------------------ 3810| 0| HUNSPELL_WARNING(stderr, "error: line %d: incorrect entry number\n", 3811| 0| af->getlinenum()); 3812| 0| return false; 3813| 0| } 3814| 4| *rl = new RepList(numrl); 3815| 4| if (!*rl) ------------------ | Branch (3815:13): [True: 0, False: 4] ------------------ 3816| 0| return false; 3817| 4| np++; 3818| 4| break; 3819| 4| } 3820| 6| default: ------------------ | Branch (3820:7): [True: 6, False: 8] ------------------ 3821| 6| break; 3822| 14| } 3823| 14| ++i; 3824| 14| start_piece = mystrsep(line, iter); 3825| 14| } 3826| 4| if (np != 2) { ------------------ | Branch (3826:7): [True: 0, False: 4] ------------------ 3827| 0| HUNSPELL_WARNING(stderr, "error: line %d: missing data\n", 3828| 0| af->getlinenum()); 3829| 0| return false; 3830| 0| } 3831| | 3832| | /* now parse the num lines to read in the remainder of the table */ 3833| 26| for (int j = 0; j < numrl; j++) { ------------------ | Branch (3833:19): [True: 22, False: 4] ------------------ 3834| 22| std::string nl; 3835| 22| if (!af->getline(nl)) ------------------ | Branch (3835:9): [True: 0, False: 22] ------------------ 3836| 0| return false; 3837| 22| mychomp(nl); 3838| 22| i = 0; 3839| 22| std::string pattern; 3840| 22| std::string pattern2; 3841| 22| iter = nl.begin(); 3842| 22| start_piece = mystrsep(nl, iter); 3843| 130| while (start_piece != nl.end()) { ------------------ | Branch (3843:12): [True: 108, False: 22] ------------------ 3844| 108| { 3845| 108| switch (i) { 3846| 22| case 0: { ------------------ | Branch (3846:11): [True: 22, False: 86] ------------------ 3847| 22| if (nl.compare(start_piece - nl.begin(), keyword.size(), keyword, 0, keyword.size()) != 0) { ------------------ | Branch (3847:17): [True: 0, False: 22] ------------------ 3848| 0| HUNSPELL_WARNING(stderr, "error: line %d: table is corrupt\n", 3849| 0| af->getlinenum()); 3850| 0| delete *rl; 3851| 0| *rl = nullptr; 3852| 0| return false; 3853| 0| } 3854| 22| break; 3855| 22| } 3856| 22| case 1: { ------------------ | Branch (3856:11): [True: 22, False: 86] ------------------ 3857| 22| pattern.assign(start_piece, iter); 3858| 22| break; 3859| 22| } 3860| 22| case 2: { ------------------ | Branch (3860:11): [True: 22, False: 86] ------------------ 3861| 22| pattern2.assign(start_piece, iter); 3862| 22| break; 3863| 22| } 3864| 42| default: ------------------ | Branch (3864:11): [True: 42, False: 66] ------------------ 3865| 42| break; 3866| 108| } 3867| 108| ++i; 3868| 108| } 3869| 0| start_piece = mystrsep(nl, iter); 3870| 108| } 3871| 22| if (pattern.empty() || pattern2.empty()) { ------------------ | Branch (3871:9): [True: 0, False: 22] | Branch (3871:28): [True: 0, False: 22] ------------------ 3872| 0| HUNSPELL_WARNING(stderr, "error: line %d: table is corrupt\n", 3873| 0| af->getlinenum()); 3874| 0| return false; 3875| 0| } 3876| | 3877| 22| (*rl)->add(pattern, pattern2); 3878| 22| } 3879| 4| return true; 3880| 4|} _ZN8AffixMgr16parse_phonetableERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEP7FileMgr: 3883| 2|bool AffixMgr::parse_phonetable(const std::string& line, FileMgr* af) { 3884| 2| if (phone) { ------------------ | Branch (3884:7): [True: 0, False: 2] ------------------ 3885| 0| HUNSPELL_WARNING(stderr, "error: line %d: multiple table definitions\n", 3886| 0| af->getlinenum()); 3887| 0| return false; 3888| 0| } 3889| 2| std::unique_ptr new_phone; 3890| 2| int num = -1; 3891| 2| int i = 0; 3892| 2| int np = 0; 3893| 2| auto iter = line.begin(), start_piece = mystrsep(line, iter); 3894| 6| while (start_piece != line.end()) { ------------------ | Branch (3894:10): [True: 4, False: 2] ------------------ 3895| 4| switch (i) { 3896| 2| case 0: { ------------------ | Branch (3896:7): [True: 2, False: 2] ------------------ 3897| 2| np++; 3898| 2| break; 3899| 0| } 3900| 2| case 1: { ------------------ | Branch (3900:7): [True: 2, False: 2] ------------------ 3901| 2| num = atoi(std::string(start_piece, iter).c_str()); 3902| 2| if (num < 1) { ------------------ | Branch (3902:13): [True: 0, False: 2] ------------------ 3903| 0| HUNSPELL_WARNING(stderr, "error: line %d: bad entry number\n", 3904| 0| af->getlinenum()); 3905| 0| return false; 3906| 0| } 3907| 2| new_phone = std::make_unique(); 3908| 2| new_phone->utf8 = (char)utf8; 3909| 2| np++; 3910| 2| break; 3911| 2| } 3912| 0| default: ------------------ | Branch (3912:7): [True: 0, False: 4] ------------------ 3913| 0| break; 3914| 4| } 3915| 4| ++i; 3916| 4| start_piece = mystrsep(line, iter); 3917| 4| } 3918| 2| if (np != 2) { ------------------ | Branch (3918:7): [True: 0, False: 2] ------------------ 3919| 0| HUNSPELL_WARNING(stderr, "error: line %d: missing data\n", 3920| 0| af->getlinenum()); 3921| 0| return false; 3922| 0| } 3923| | 3924| | /* now parse the phone->num lines to read in the remainder of the table */ 3925| 212| for (int j = 0; j < num; ++j) { ------------------ | Branch (3925:19): [True: 210, False: 2] ------------------ 3926| 210| std::string nl; 3927| 210| if (!af->getline(nl)) ------------------ | Branch (3927:9): [True: 0, False: 210] ------------------ 3928| 0| return false; 3929| 210| mychomp(nl); 3930| 210| i = 0; 3931| 210| const size_t old_size = new_phone->rules.size(); 3932| 210| iter = nl.begin(); 3933| 210| start_piece = mystrsep(nl, iter); 3934| 840| while (start_piece != nl.end()) { ------------------ | Branch (3934:12): [True: 630, False: 210] ------------------ 3935| 630| { 3936| 630| switch (i) { 3937| 210| case 0: { ------------------ | Branch (3937:11): [True: 210, False: 420] ------------------ 3938| 210| if (nl.compare(start_piece - nl.begin(), 5, "PHONE", 5) != 0) { ------------------ | Branch (3938:17): [True: 0, False: 210] ------------------ 3939| 0| HUNSPELL_WARNING(stderr, "error: line %d: table is corrupt\n", 3940| 0| af->getlinenum()); 3941| 0| return false; 3942| 0| } 3943| 210| break; 3944| 210| } 3945| 210| case 1: { ------------------ | Branch (3945:11): [True: 210, False: 420] ------------------ 3946| 210| new_phone->rules.emplace_back(start_piece, iter); 3947| 210| break; 3948| 210| } 3949| 210| case 2: { ------------------ | Branch (3949:11): [True: 210, False: 420] ------------------ 3950| 210| new_phone->rules.emplace_back(start_piece, iter); 3951| 210| mystrrep(new_phone->rules.back(), "_", ""); 3952| 210| break; 3953| 210| } 3954| 0| default: ------------------ | Branch (3954:11): [True: 0, False: 630] ------------------ 3955| 0| break; 3956| 630| } 3957| 630| ++i; 3958| 630| } 3959| 0| start_piece = mystrsep(nl, iter); 3960| 630| } 3961| 210| if (new_phone->rules.size() != old_size + 2) { ------------------ | Branch (3961:9): [True: 0, False: 210] ------------------ 3962| 0| HUNSPELL_WARNING(stderr, "error: line %d: table is corrupt\n", 3963| 0| af->getlinenum()); 3964| 0| return false; 3965| 0| } 3966| 210| } 3967| 2| new_phone->rules.emplace_back(""); 3968| 2| new_phone->rules.emplace_back(""); 3969| 2| init_phonet_hash(*new_phone); 3970| 2| phone = new_phone.release(); 3971| 2| return true; 3972| 2|} _ZN8AffixMgr19parse_checkcpdtableERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEP7FileMgr: 3975| 2|bool AffixMgr::parse_checkcpdtable(const std::string& line, FileMgr* af) { 3976| 2| if (parsedcheckcpd) { ------------------ | Branch (3976:7): [True: 0, False: 2] ------------------ 3977| 0| HUNSPELL_WARNING(stderr, "error: line %d: multiple table definitions\n", 3978| 0| af->getlinenum()); 3979| 0| return false; 3980| 0| } 3981| 2| parsedcheckcpd = true; 3982| 2| int numcheckcpd = -1; 3983| 2| int i = 0; 3984| 2| int np = 0; 3985| 2| auto iter = line.begin(), start_piece = mystrsep(line, iter); 3986| 6| while (start_piece != line.end()) { ------------------ | Branch (3986:10): [True: 4, False: 2] ------------------ 3987| 4| switch (i) { 3988| 2| case 0: { ------------------ | Branch (3988:7): [True: 2, False: 2] ------------------ 3989| 2| np++; 3990| 2| break; 3991| 0| } 3992| 2| case 1: { ------------------ | Branch (3992:7): [True: 2, False: 2] ------------------ 3993| 2| numcheckcpd = atoi(std::string(start_piece, iter).c_str()); 3994| 2| if (numcheckcpd < 1) { ------------------ | Branch (3994:13): [True: 0, False: 2] ------------------ 3995| 0| HUNSPELL_WARNING(stderr, "error: line %d: bad entry number\n", 3996| 0| af->getlinenum()); 3997| 0| return false; 3998| 0| } 3999| 2| checkcpdtable.reserve(std::min(numcheckcpd, 16384)); 4000| 2| np++; 4001| 2| break; 4002| 2| } 4003| 0| default: ------------------ | Branch (4003:7): [True: 0, False: 4] ------------------ 4004| 0| break; 4005| 4| } 4006| 4| ++i; 4007| 4| start_piece = mystrsep(line, iter); 4008| 4| } 4009| 2| if (np != 2) { ------------------ | Branch (4009:7): [True: 0, False: 2] ------------------ 4010| 0| HUNSPELL_WARNING(stderr, "error: line %d: missing data\n", 4011| 0| af->getlinenum()); 4012| 0| return false; 4013| 0| } 4014| | 4015| | /* now parse the numcheckcpd lines to read in the remainder of the table */ 4016| 6| for (int j = 0; j < numcheckcpd; ++j) { ------------------ | Branch (4016:19): [True: 4, False: 2] ------------------ 4017| 4| std::string nl; 4018| 4| if (!af->getline(nl)) ------------------ | Branch (4018:9): [True: 0, False: 4] ------------------ 4019| 0| return false; 4020| 4| mychomp(nl); 4021| 4| i = 0; 4022| 4| checkcpdtable.emplace_back(); 4023| 4| iter = nl.begin(); 4024| 4| start_piece = mystrsep(nl, iter); 4025| 16| while (start_piece != nl.end()) { ------------------ | Branch (4025:12): [True: 12, False: 4] ------------------ 4026| 12| switch (i) { 4027| 4| case 0: { ------------------ | Branch (4027:9): [True: 4, False: 8] ------------------ 4028| 4| if (nl.compare(start_piece - nl.begin(), 20, "CHECKCOMPOUNDPATTERN", 20) != 0) { ------------------ | Branch (4028:15): [True: 0, False: 4] ------------------ 4029| 0| HUNSPELL_WARNING(stderr, "error: line %d: table is corrupt\n", 4030| 0| af->getlinenum()); 4031| 0| checkcpdtable.clear(); 4032| 0| return false; 4033| 0| } 4034| 4| break; 4035| 4| } 4036| 4| case 1: { ------------------ | Branch (4036:9): [True: 4, False: 8] ------------------ 4037| 4| checkcpdtable.back().pattern.assign(start_piece, iter); 4038| 4| size_t slash_pos = checkcpdtable.back().pattern.find('/'); 4039| 4| if (slash_pos != std::string::npos) { ------------------ | Branch (4039:15): [True: 0, False: 4] ------------------ 4040| 0| std::string chunk(checkcpdtable.back().pattern, slash_pos + 1); 4041| 0| checkcpdtable.back().pattern.resize(slash_pos); 4042| 0| checkcpdtable.back().cond = pHMgr->decode_flag(chunk); 4043| 0| } 4044| 4| break; 4045| 4| } 4046| 4| case 2: { ------------------ | Branch (4046:9): [True: 4, False: 8] ------------------ 4047| 4| checkcpdtable.back().pattern2.assign(start_piece, iter); 4048| 4| size_t slash_pos = checkcpdtable.back().pattern2.find('/'); 4049| 4| if (slash_pos != std::string::npos) { ------------------ | Branch (4049:15): [True: 0, False: 4] ------------------ 4050| 0| std::string chunk(checkcpdtable.back().pattern2, slash_pos + 1); 4051| 0| checkcpdtable.back().pattern2.resize(slash_pos); 4052| 0| checkcpdtable.back().cond2 = pHMgr->decode_flag(chunk); 4053| 0| } 4054| 4| break; 4055| 4| } 4056| 0| case 3: { ------------------ | Branch (4056:9): [True: 0, False: 12] ------------------ 4057| 0| checkcpdtable.back().pattern3.assign(start_piece, iter); 4058| 0| simplifiedcpd = 1; 4059| 0| break; 4060| 4| } 4061| 0| default: ------------------ | Branch (4061:9): [True: 0, False: 12] ------------------ 4062| 0| break; 4063| 12| } 4064| 12| i++; 4065| 12| start_piece = mystrsep(nl, iter); 4066| 12| } 4067| 4| } 4068| 2| return true; 4069| 2|} _ZN8AffixMgr14parse_maptableERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEP7FileMgr: 4173| 2|bool AffixMgr::parse_maptable(const std::string& line, FileMgr* af) { 4174| 2| if (parsedmaptable) { ------------------ | Branch (4174:7): [True: 0, False: 2] ------------------ 4175| 0| HUNSPELL_WARNING(stderr, "error: line %d: multiple table definitions\n", 4176| 0| af->getlinenum()); 4177| 0| return false; 4178| 0| } 4179| 2| parsedmaptable = true; 4180| 2| int nummap = -1; 4181| 2| int i = 0; 4182| 2| int np = 0; 4183| 2| auto iter = line.begin(), start_piece = mystrsep(line, iter); 4184| 6| while (start_piece != line.end()) { ------------------ | Branch (4184:10): [True: 4, False: 2] ------------------ 4185| 4| switch (i) { 4186| 2| case 0: { ------------------ | Branch (4186:7): [True: 2, False: 2] ------------------ 4187| 2| np++; 4188| 2| break; 4189| 0| } 4190| 2| case 1: { ------------------ | Branch (4190:7): [True: 2, False: 2] ------------------ 4191| 2| nummap = atoi(std::string(start_piece, iter).c_str()); 4192| 2| if (nummap < 1) { ------------------ | Branch (4192:13): [True: 0, False: 2] ------------------ 4193| 0| HUNSPELL_WARNING(stderr, "error: line %d: bad entry number\n", 4194| 0| af->getlinenum()); 4195| 0| return false; 4196| 0| } 4197| 2| maptable.reserve(std::min(nummap, 16384)); 4198| 2| np++; 4199| 2| break; 4200| 2| } 4201| 0| default: ------------------ | Branch (4201:7): [True: 0, False: 4] ------------------ 4202| 0| break; 4203| 4| } 4204| 4| ++i; 4205| 4| start_piece = mystrsep(line, iter); 4206| 4| } 4207| 2| if (np != 2) { ------------------ | Branch (4207:7): [True: 0, False: 2] ------------------ 4208| 0| HUNSPELL_WARNING(stderr, "error: line %d: missing data\n", 4209| 0| af->getlinenum()); 4210| 0| return false; 4211| 0| } 4212| | 4213| | /* now parse the nummap lines to read in the remainder of the table */ 4214| 8| for (int j = 0; j < nummap; ++j) { ------------------ | Branch (4214:19): [True: 6, False: 2] ------------------ 4215| 6| std::string nl; 4216| 6| if (!af->getline(nl)) ------------------ | Branch (4216:9): [True: 0, False: 6] ------------------ 4217| 0| return false; 4218| 6| mychomp(nl); 4219| 6| i = 0; 4220| 6| maptable.emplace_back(); 4221| 6| iter = nl.begin(); 4222| 6| start_piece = mystrsep(nl, iter); 4223| 18| while (start_piece != nl.end()) { ------------------ | Branch (4223:12): [True: 12, False: 6] ------------------ 4224| 12| switch (i) { 4225| 6| case 0: { ------------------ | Branch (4225:9): [True: 6, False: 6] ------------------ 4226| 6| if (nl.compare(start_piece - nl.begin(), 3, "MAP", 3) != 0) { ------------------ | Branch (4226:15): [True: 0, False: 6] ------------------ 4227| 0| HUNSPELL_WARNING(stderr, "error: line %d: table is corrupt\n", 4228| 0| af->getlinenum()); 4229| 0| nummap = 0; 4230| 0| return false; 4231| 0| } 4232| 6| break; 4233| 6| } 4234| 6| case 1: { ------------------ | Branch (4234:9): [True: 6, False: 6] ------------------ 4235| 22| for (auto k = start_piece; k != iter; ++k) { ------------------ | Branch (4235:38): [True: 16, False: 6] ------------------ 4236| 16| auto chb = k, che = k + 1; 4237| 16| if (*k == '(') { ------------------ | Branch (4237:17): [True: 2, False: 14] ------------------ 4238| 2| auto parpos = std::find(k, iter, ')'); 4239| 2| if (parpos != iter) { ------------------ | Branch (4239:19): [True: 2, False: 0] ------------------ 4240| 2| chb = k + 1; 4241| 2| che = parpos; 4242| 2| k = parpos; 4243| 2| } 4244| 14| } else { 4245| 14| if (utf8 && (*k & 0xc0) == 0xc0) { ------------------ | Branch (4245:19): [True: 14, False: 0] | Branch (4245:27): [True: 10, False: 4] ------------------ 4246| 10| ++k; 4247| 20| while (k != iter && is_utf8_cont(*k)) ------------------ | Branch (4247:24): [True: 18, False: 2] | Branch (4247:37): [True: 10, False: 8] ------------------ 4248| 10| ++k; 4249| 10| che = k; 4250| 10| --k; 4251| 10| } 4252| 14| } 4253| 16| if (chb == che) { ------------------ | Branch (4253:17): [True: 0, False: 16] ------------------ 4254| 0| HUNSPELL_WARNING(stderr, "error: line %d: table is corrupt\n", 4255| 0| af->getlinenum()); 4256| 0| } 4257| | 4258| 16| maptable.back().emplace_back(chb, che); 4259| 16| } 4260| 6| break; 4261| 6| } 4262| 0| default: ------------------ | Branch (4262:9): [True: 0, False: 12] ------------------ 4263| 0| break; 4264| 12| } 4265| 12| ++i; 4266| 12| start_piece = mystrsep(nl, iter); 4267| 12| } 4268| 6| if (maptable.back().empty()) { ------------------ | Branch (4268:9): [True: 0, False: 6] ------------------ 4269| 0| HUNSPELL_WARNING(stderr, "error: line %d: table is corrupt\n", 4270| 0| af->getlinenum()); 4271| 0| return false; 4272| 0| } 4273| 6| } 4274| 2| return true; 4275| 2|} _ZN8AffixMgr16parse_breaktableERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEP7FileMgr: 4278| 4|bool AffixMgr::parse_breaktable(const std::string& line, FileMgr* af) { 4279| 4| if (parsedbreaktable) { ------------------ | Branch (4279:7): [True: 0, False: 4] ------------------ 4280| 0| HUNSPELL_WARNING(stderr, "error: line %d: multiple table definitions\n", 4281| 0| af->getlinenum()); 4282| 0| return false; 4283| 0| } 4284| 4| parsedbreaktable = true; 4285| 4| int numbreak = -1; 4286| 4| int i = 0; 4287| 4| int np = 0; 4288| 4| auto iter = line.begin(), start_piece = mystrsep(line, iter); 4289| 12| while (start_piece != line.end()) { ------------------ | Branch (4289:10): [True: 8, False: 4] ------------------ 4290| 8| switch (i) { 4291| 4| case 0: { ------------------ | Branch (4291:7): [True: 4, False: 4] ------------------ 4292| 4| np++; 4293| 4| break; 4294| 0| } 4295| 4| case 1: { ------------------ | Branch (4295:7): [True: 4, False: 4] ------------------ 4296| 4| numbreak = atoi(std::string(start_piece, iter).c_str()); 4297| 4| if (numbreak < 0) { ------------------ | Branch (4297:13): [True: 0, False: 4] ------------------ 4298| 0| HUNSPELL_WARNING(stderr, "error: line %d: bad entry number\n", 4299| 0| af->getlinenum()); 4300| 0| return false; 4301| 0| } 4302| 4| if (numbreak == 0) ------------------ | Branch (4302:13): [True: 0, False: 4] ------------------ 4303| 0| return true; 4304| 4| breaktable.reserve(std::min(numbreak, 16384)); 4305| 4| np++; 4306| 4| break; 4307| 4| } 4308| 0| default: ------------------ | Branch (4308:7): [True: 0, False: 8] ------------------ 4309| 0| break; 4310| 8| } 4311| 8| ++i; 4312| 8| start_piece = mystrsep(line, iter); 4313| 8| } 4314| 4| if (np != 2) { ------------------ | Branch (4314:7): [True: 0, False: 4] ------------------ 4315| 0| HUNSPELL_WARNING(stderr, "error: line %d: missing data\n", 4316| 0| af->getlinenum()); 4317| 0| return false; 4318| 0| } 4319| | 4320| | /* now parse the numbreak lines to read in the remainder of the table */ 4321| 10| for (int j = 0; j < numbreak; ++j) { ------------------ | Branch (4321:19): [True: 6, False: 4] ------------------ 4322| 6| std::string nl; 4323| 6| if (!af->getline(nl)) ------------------ | Branch (4323:9): [True: 0, False: 6] ------------------ 4324| 0| return false; 4325| 6| mychomp(nl); 4326| 6| i = 0; 4327| 6| iter = nl.begin(); 4328| 6| start_piece = mystrsep(nl, iter); 4329| 18| while (start_piece != nl.end()) { ------------------ | Branch (4329:12): [True: 12, False: 6] ------------------ 4330| 12| switch (i) { 4331| 6| case 0: { ------------------ | Branch (4331:9): [True: 6, False: 6] ------------------ 4332| 6| if (nl.compare(start_piece - nl.begin(), 5, "BREAK", 5) != 0) { ------------------ | Branch (4332:15): [True: 0, False: 6] ------------------ 4333| 0| HUNSPELL_WARNING(stderr, "error: line %d: table is corrupt\n", 4334| 0| af->getlinenum()); 4335| 0| numbreak = 0; 4336| 0| return false; 4337| 0| } 4338| 6| break; 4339| 6| } 4340| 6| case 1: { ------------------ | Branch (4340:9): [True: 6, False: 6] ------------------ 4341| 6| breaktable.emplace_back(start_piece, iter); 4342| 6| break; 4343| 6| } 4344| 0| default: ------------------ | Branch (4344:9): [True: 0, False: 12] ------------------ 4345| 0| break; 4346| 12| } 4347| 12| ++i; 4348| 12| start_piece = mystrsep(nl, iter); 4349| 12| } 4350| 6| } 4351| | 4352| 4| if (breaktable.size() != static_cast(numbreak)) { ------------------ | Branch (4352:7): [True: 0, False: 4] ------------------ 4353| 0| HUNSPELL_WARNING(stderr, "error: line %d: table is corrupt\n", 4354| 0| af->getlinenum()); 4355| 0| return false; 4356| 0| } 4357| | 4358| 4| return true; 4359| 4|} _ZN8AffixMgr17reverse_conditionERNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEE: 4361| 38|void AffixMgr::reverse_condition(std::string& piece) { 4362| 38| if (piece.empty()) ------------------ | Branch (4362:7): [True: 0, False: 38] ------------------ 4363| 0| return; 4364| | 4365| 38| int neg = 0; 4366| | // iterate backwards; k wraps to npos (SIZE_MAX) when decremented past 0 4367| 92| for (size_t k = piece.size() - 1; k != std::string::npos; --k) { ------------------ | Branch (4367:37): [True: 54, False: 38] ------------------ 4368| 54| switch (piece[k]) { 4369| 2| case '[': { ------------------ | Branch (4369:7): [True: 2, False: 52] ------------------ 4370| 2| if (neg) ------------------ | Branch (4370:13): [True: 0, False: 2] ------------------ 4371| 0| piece[k + 1] = '['; 4372| 2| else 4373| 2| piece[k] = ']'; 4374| 2| break; 4375| 0| } 4376| 2| case ']': { ------------------ | Branch (4376:7): [True: 2, False: 52] ------------------ 4377| 2| piece[k] = '['; 4378| 2| if (neg) ------------------ | Branch (4378:13): [True: 0, False: 2] ------------------ 4379| 0| piece[k + 1] = '^'; 4380| 2| neg = 0; 4381| 2| break; 4382| 0| } 4383| 0| case '^': { ------------------ | Branch (4383:7): [True: 0, False: 54] ------------------ 4384| 0| if (piece[k + 1] == ']') ------------------ | Branch (4384:13): [True: 0, False: 0] ------------------ 4385| 0| neg = 1; 4386| 0| else if (neg) ------------------ | Branch (4386:18): [True: 0, False: 0] ------------------ 4387| 0| piece[k + 1] = piece[k]; 4388| 0| break; 4389| 0| } 4390| 50| default: { ------------------ | Branch (4390:7): [True: 50, False: 4] ------------------ 4391| 50| if (neg) ------------------ | Branch (4391:13): [True: 0, False: 50] ------------------ 4392| 0| piece[k + 1] = piece[k]; 4393| 50| } 4394| 54| } 4395| 54| } 4396| 38|} _ZN8AffixMgr11parse_affixERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEcP7FileMgrPc: 4450| 26| char* dupflags) { 4451| 26| int numents = 0; // number of AffEntry structures to parse 4452| | 4453| 26| unsigned short aflag = 0; // affix char identifier 4454| | 4455| 26| char ff = 0; 4456| 26| entries_container affentries(at, this); 4457| | 4458| 26| int i = 0; 4459| | 4460| |// checking lines with bad syntax 4461| |#ifdef DEBUG 4462| | int basefieldnum = 0; 4463| |#endif 4464| | 4465| | // split affix header line into pieces 4466| | 4467| 26| int np = 0; 4468| 26| auto iter = line.begin(), start_piece = mystrsep(line, iter); 4469| 130| while (start_piece != line.end()) { ------------------ | Branch (4469:10): [True: 104, False: 26] ------------------ 4470| 104| switch (i) { 4471| | // piece 1 - is type of affix 4472| 26| case 0: { ------------------ | Branch (4472:7): [True: 26, False: 78] ------------------ 4473| 26| np++; 4474| 26| break; 4475| 0| } 4476| | 4477| | // piece 2 - is affix char 4478| 26| case 1: { ------------------ | Branch (4478:7): [True: 26, False: 78] ------------------ 4479| 26| np++; 4480| 26| aflag = pHMgr->decode_flag(std::string(start_piece, iter)); 4481| 26| if (((at == 'S') && (dupflags[aflag] & dupSFX)) || ------------------ | | 88| 18|#define dupSFX (1 << 0) ------------------ | Branch (4481:14): [True: 18, False: 8] | Branch (4481:29): [True: 0, False: 18] ------------------ 4482| 26| ((at == 'P') && (dupflags[aflag] & dupPFX))) { ------------------ | | 89| 8|#define dupPFX (1 << 1) ------------------ | Branch (4482:14): [True: 8, False: 18] | Branch (4482:29): [True: 0, False: 8] ------------------ 4483| 0| HUNSPELL_WARNING( 4484| 0| stderr, 4485| 0| "error: line %d: multiple definitions of an affix flag\n", 4486| 0| af->getlinenum()); 4487| 0| } 4488| 26| dupflags[aflag] += (char)((at == 'S') ? dupSFX : dupPFX); ------------------ | | 88| 18|#define dupSFX (1 << 0) ------------------ dupflags[aflag] += (char)((at == 'S') ? dupSFX : dupPFX); ------------------ | | 89| 8|#define dupPFX (1 << 1) ------------------ | Branch (4488:35): [True: 18, False: 8] ------------------ 4489| 26| break; 4490| 0| } 4491| | // piece 3 - is cross product indicator 4492| 26| case 2: { ------------------ | Branch (4492:7): [True: 26, False: 78] ------------------ 4493| 26| np++; 4494| 26| if (*start_piece == 'Y') ------------------ | Branch (4494:13): [True: 26, False: 0] ------------------ 4495| 26| ff = aeXPRODUCT; ------------------ | | 65| 26|#define aeXPRODUCT (1 << 0) ------------------ 4496| 26| break; 4497| 0| } 4498| | 4499| | // piece 4 - is number of affentries 4500| 26| case 3: { ------------------ | Branch (4500:7): [True: 26, False: 78] ------------------ 4501| 26| np++; 4502| 26| numents = atoi(std::string(start_piece, iter).c_str()); 4503| 26| if ((numents <= 0) || ((std::numeric_limits::max() / ------------------ | Branch (4503:13): [True: 0, False: 26] | Branch (4503:31): [True: 0, False: 26] ------------------ 4504| 26| sizeof(AffEntry)) < static_cast(numents))) { 4505| 0| std::string err = pHMgr->encode_flag(aflag); 4506| 0| HUNSPELL_WARNING(stderr, "error: line %d: affix %s: bad entry number\n", 4507| 0| af->getlinenum(), err.c_str()); 4508| 0| return false; 4509| 0| } 4510| | 4511| 26| char opts = ff; 4512| 26| if (utf8) ------------------ | Branch (4512:13): [True: 6, False: 20] ------------------ 4513| 6| opts |= aeUTF8; ------------------ | | 66| 6|#define aeUTF8 (1 << 1) ------------------ 4514| 26| if (pHMgr->is_aliasf()) ------------------ | Branch (4514:13): [True: 4, False: 22] ------------------ 4515| 4| opts |= aeALIASF; ------------------ | | 67| 4|#define aeALIASF (1 << 2) ------------------ 4516| 26| if (pHMgr->is_aliasm()) ------------------ | Branch (4516:13): [True: 0, False: 26] ------------------ 4517| 0| opts |= aeALIASM; ------------------ | | 68| 0|#define aeALIASM (1 << 3) ------------------ 4518| 26| affentries.initialize(numents, opts, aflag); 4519| 26| } 4520| | 4521| 26| default: ------------------ | Branch (4521:7): [True: 0, False: 104] ------------------ 4522| 26| break; 4523| 104| } 4524| 104| ++i; 4525| 104| start_piece = mystrsep(line, iter); 4526| 104| } 4527| | // check to make sure we parsed enough pieces 4528| 26| if (np != 4) { ------------------ | Branch (4528:7): [True: 0, False: 26] ------------------ 4529| 0| std::string err = pHMgr->encode_flag(aflag); 4530| 0| HUNSPELL_WARNING(stderr, "error: line %d: affix %s: missing data\n", 4531| 0| af->getlinenum(), err.c_str()); 4532| 0| return false; 4533| 0| } 4534| | 4535| | // now parse numents affentries for this affix 4536| 26| AffEntry* entry = affentries.first_entry(); 4537| 68| for (int ent = 0; ent < numents; ++ent) { ------------------ | Branch (4537:21): [True: 42, False: 26] ------------------ 4538| 42| std::string nl; 4539| 42| if (!af->getline(nl)) ------------------ | Branch (4539:9): [True: 0, False: 42] ------------------ 4540| 0| return false; 4541| 42| mychomp(nl); 4542| | 4543| 42| iter = nl.begin(); 4544| 42| i = 0; 4545| 42| np = 0; 4546| | 4547| | // split line into pieces 4548| 42| start_piece = mystrsep(nl, iter); 4549| 258| while (start_piece != nl.end()) { ------------------ | Branch (4549:12): [True: 216, False: 42] ------------------ 4550| 216| switch (i) { 4551| | // piece 1 - is type 4552| 42| case 0: { ------------------ | Branch (4552:9): [True: 42, False: 174] ------------------ 4553| 42| np++; 4554| 42| if (ent != 0) ------------------ | Branch (4554:15): [True: 16, False: 26] ------------------ 4555| 16| entry = affentries.add_entry((char)(aeXPRODUCT | aeUTF8 | aeALIASF | aeALIASM)); ------------------ | | 65| 16|#define aeXPRODUCT (1 << 0) ------------------ entry = affentries.add_entry((char)(aeXPRODUCT | aeUTF8 | aeALIASF | aeALIASM)); ------------------ | | 66| 16|#define aeUTF8 (1 << 1) ------------------ entry = affentries.add_entry((char)(aeXPRODUCT | aeUTF8 | aeALIASF | aeALIASM)); ------------------ | | 67| 16|#define aeALIASF (1 << 2) ------------------ entry = affentries.add_entry((char)(aeXPRODUCT | aeUTF8 | aeALIASF | aeALIASM)); ------------------ | | 68| 16|#define aeALIASM (1 << 3) ------------------ 4556| 42| break; 4557| 0| } 4558| | 4559| | // piece 2 - is affix char 4560| 42| case 1: { ------------------ | Branch (4560:9): [True: 42, False: 174] ------------------ 4561| 42| np++; 4562| 42| std::string chunk(start_piece, iter); 4563| 42| if (pHMgr->decode_flag(chunk) != aflag) { ------------------ | Branch (4563:15): [True: 0, False: 42] ------------------ 4564| 0| std::string err = pHMgr->encode_flag(aflag); 4565| 0| HUNSPELL_WARNING(stderr, 4566| 0| "error: line %d: affix %s is corrupt\n", 4567| 0| af->getlinenum(), err.c_str()); 4568| 0| return false; 4569| 0| } 4570| | 4571| 42| if (ent != 0) { ------------------ | Branch (4571:15): [True: 16, False: 26] ------------------ 4572| 16| AffEntry* start_entry = affentries.first_entry(); 4573| 16| entry->aflag = start_entry->aflag; 4574| 16| } 4575| 42| break; 4576| 42| } 4577| | 4578| | // piece 3 - is string to strip or 0 for null 4579| 42| case 2: { ------------------ | Branch (4579:9): [True: 42, False: 174] ------------------ 4580| 42| np++; 4581| 42| entry->strip = std::string(start_piece, iter); 4582| 42| if (complexprefixes) { ------------------ | Branch (4582:15): [True: 4, False: 38] ------------------ 4583| 4| if (utf8) ------------------ | Branch (4583:17): [True: 0, False: 4] ------------------ 4584| 0| reverseword_utf(entry->strip); 4585| 4| else 4586| 4| reverseword(entry->strip); 4587| 4| } 4588| 42| if (entry->strip.compare("0") == 0) { ------------------ | Branch (4588:15): [True: 36, False: 6] ------------------ 4589| 36| entry->strip.clear(); 4590| 36| } 4591| 42| break; 4592| 42| } 4593| | 4594| | // piece 4 - is affix string or 0 for null 4595| 42| case 3: { ------------------ | Branch (4595:9): [True: 42, False: 174] ------------------ 4596| 42| entry->morphcode = nullptr; 4597| 42| entry->contclass = nullptr; 4598| 42| entry->contclasslen = 0; 4599| 42| np++; 4600| 42| std::string::const_iterator dash = std::find(start_piece, iter, '/'); 4601| 42| if (dash != iter) { ------------------ | Branch (4601:15): [True: 18, False: 24] ------------------ 4602| 18| entry->appnd = std::string(start_piece, dash); 4603| 18| std::string dash_str(dash + 1, iter); 4604| | 4605| 18| if (!ignorechars.empty() && !has_no_ignored_chars(entry->appnd, ignorechars)) { ------------------ | Branch (4605:17): [True: 2, False: 16] | Branch (4605:41): [True: 0, False: 2] ------------------ 4606| 0| if (utf8) { ------------------ | Branch (4606:19): [True: 0, False: 0] ------------------ 4607| 0| remove_ignored_chars_utf(entry->appnd, ignorechars_utf16); 4608| 0| } else { 4609| 0| remove_ignored_chars(entry->appnd, ignorechars); 4610| 0| } 4611| 0| } 4612| | 4613| 18| if (complexprefixes) { ------------------ | Branch (4613:17): [True: 2, False: 16] ------------------ 4614| 2| if (utf8) ------------------ | Branch (4614:19): [True: 0, False: 2] ------------------ 4615| 0| reverseword_utf(entry->appnd); 4616| 2| else 4617| 2| reverseword(entry->appnd); 4618| 2| } 4619| | 4620| 18| if (pHMgr->is_aliasf()) { ------------------ | Branch (4620:17): [True: 2, False: 16] ------------------ 4621| 2| int index = atoi(dash_str.c_str()); 4622| 2| entry->contclasslen = (unsigned short)pHMgr->get_aliasf( 4623| 2| index, &(entry->contclass), af); 4624| 2| if (!entry->contclasslen) ------------------ | Branch (4624:19): [True: 0, False: 2] ------------------ 4625| 0| HUNSPELL_WARNING(stderr, 4626| 0| "error: bad affix flag alias: \"%s\"\n", 4627| 0| dash_str.c_str()); 4628| 16| } else { 4629| 16| entry->contclasslen = (unsigned short)pHMgr->decode_flags( 4630| 16| &(entry->contclass), dash_str, af); 4631| 16| std::sort(entry->contclass, entry->contclass + entry->contclasslen); 4632| 16| } 4633| | 4634| 18| havecontclass = 1; 4635| 44| for (unsigned short _i = 0; _i < entry->contclasslen; _i++) { ------------------ | Branch (4635:41): [True: 26, False: 18] ------------------ 4636| 26| contclasses[(entry->contclass)[_i]] = 1; 4637| 26| } 4638| 24| } else { 4639| 24| entry->appnd = std::string(start_piece, iter); 4640| | 4641| 24| if (!ignorechars.empty() && !has_no_ignored_chars(entry->appnd, ignorechars)) { ------------------ | Branch (4641:17): [True: 2, False: 22] | Branch (4641:41): [True: 2, False: 0] ------------------ 4642| 2| if (utf8) { ------------------ | Branch (4642:19): [True: 0, False: 2] ------------------ 4643| 0| remove_ignored_chars_utf(entry->appnd, ignorechars_utf16); 4644| 2| } else { 4645| 2| remove_ignored_chars(entry->appnd, ignorechars); 4646| 2| } 4647| 2| } 4648| | 4649| 24| if (complexprefixes) { ------------------ | Branch (4649:17): [True: 2, False: 22] ------------------ 4650| 2| if (utf8) ------------------ | Branch (4650:19): [True: 0, False: 2] ------------------ 4651| 0| reverseword_utf(entry->appnd); 4652| 2| else 4653| 2| reverseword(entry->appnd); 4654| 2| } 4655| 24| } 4656| | 4657| 42| if (entry->appnd.compare("0") == 0) { ------------------ | Branch (4657:15): [True: 2, False: 40] ------------------ 4658| 2| entry->appnd.clear(); 4659| 2| } 4660| 42| break; 4661| 42| } 4662| | 4663| | // piece 5 - is the conditions descriptions 4664| 42| case 4: { ------------------ | Branch (4664:9): [True: 42, False: 174] ------------------ 4665| 42| std::string chunk(start_piece, iter); 4666| 42| np++; 4667| 42| if (complexprefixes) { ------------------ | Branch (4667:15): [True: 4, False: 38] ------------------ 4668| 4| if (utf8) ------------------ | Branch (4668:17): [True: 0, False: 4] ------------------ 4669| 0| reverseword_utf(chunk); 4670| 4| else 4671| 4| reverseword(chunk); 4672| 4| reverse_condition(chunk); 4673| 4| } 4674| 42| if (!entry->strip.empty() && chunk != "." && ------------------ | Branch (4674:15): [True: 6, False: 36] | Branch (4674:40): [True: 6, False: 0] ------------------ 4675| 6| redundant_condition(at, entry->strip, chunk, ------------------ | Branch (4675:15): [True: 6, False: 0] ------------------ 4676| 6| af->getlinenum())) 4677| 6| chunk = "."; 4678| 42| if (at == 'S') { ------------------ | Branch (4678:15): [True: 34, False: 8] ------------------ 4679| 34| reverseword(chunk); 4680| 34| reverse_condition(chunk); 4681| 34| } 4682| 42| if (encodeit(*entry, chunk)) ------------------ | Branch (4682:15): [True: 0, False: 42] ------------------ 4683| 0| return false; 4684| 42| break; 4685| 42| } 4686| | 4687| 42| case 5: { ------------------ | Branch (4687:9): [True: 6, False: 210] ------------------ 4688| 6| std::string chunk(start_piece, iter); 4689| 6| np++; 4690| 6| if (pHMgr->is_aliasm()) { ------------------ | Branch (4690:15): [True: 0, False: 6] ------------------ 4691| 0| int index = atoi(chunk.c_str()); 4692| 0| entry->morphcode = pHMgr->get_aliasm(index); 4693| 6| } else { 4694| 6| if (complexprefixes) { // XXX - fix me for morph. gen. ------------------ | Branch (4694:17): [True: 0, False: 6] ------------------ 4695| 0| if (utf8) ------------------ | Branch (4695:19): [True: 0, False: 0] ------------------ 4696| 0| reverseword_utf(chunk); 4697| 0| else 4698| 0| reverseword(chunk); 4699| 0| } 4700| | // add the remaining of the line 4701| 6| std::string::const_iterator end = nl.end(); 4702| 6| if (iter != end) { ------------------ | Branch (4702:17): [True: 0, False: 6] ------------------ 4703| 0| chunk.append(iter, end); 4704| 0| } 4705| 6| entry->morphcode = mystrdup(chunk.c_str()); 4706| 6| } 4707| 6| break; 4708| 42| } 4709| 0| default: ------------------ | Branch (4709:9): [True: 0, False: 216] ------------------ 4710| 0| break; 4711| 216| } 4712| 216| i++; 4713| 216| start_piece = mystrsep(nl, iter); 4714| 216| } 4715| | // check to make sure we parsed enough pieces 4716| 42| if (np < 4) { ------------------ | Branch (4716:9): [True: 0, False: 42] ------------------ 4717| 0| std::string err = pHMgr->encode_flag(aflag); 4718| 0| HUNSPELL_WARNING(stderr, "error: line %d: affix %s is corrupt\n", 4719| 0| af->getlinenum(), err.c_str()); 4720| 0| return false; 4721| 0| } 4722| | 4723| |#ifdef DEBUG 4724| | // detect unnecessary fields, excepting comments 4725| | if (basefieldnum) { 4726| | int fieldnum = 4727| | !(entry->morphcode) ? 5 : ((*(entry->morphcode) == '#') ? 5 : 6); 4728| | if (fieldnum != basefieldnum) 4729| | HUNSPELL_WARNING(stderr, "warning: line %d: bad field number\n", 4730| | af->getlinenum()); 4731| | } else { 4732| | basefieldnum = 4733| | !(entry->morphcode) ? 5 : ((*(entry->morphcode) == '#') ? 5 : 6); 4734| | } 4735| |#endif 4736| 42| } 4737| | 4738| | // now create SfxEntry or PfxEntry objects and use links to 4739| | // build an ordered (sorted by affix string) list 4740| 26| auto start = affentries.begin(), end = affentries.end(); 4741| 68| for (auto affentry = start; affentry != end; ++affentry) { ------------------ | Branch (4741:31): [True: 42, False: 26] ------------------ 4742| 42| if (at == 'P') { ------------------ | Branch (4742:9): [True: 8, False: 34] ------------------ 4743| 8| build_pfxtree(static_cast(*affentry)); 4744| 34| } else { 4745| 34| build_sfxtree(static_cast(*affentry)); 4746| 34| } 4747| 42| } 4748| | 4749| | //contents belong to AffixMgr now 4750| 26| affentries.release(); 4751| | 4752| 26| return true; 4753| 26|} _ZN8AffixMgr19redundant_conditionEcRKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEES8_i: 4758| 6| int linenum) { 4759| 6| int stripl = strip.size(), condl = cond.size(), i, j, neg, in; 4760| 6| if (ft == 'P') { // prefix ------------------ | Branch (4760:7): [True: 0, False: 6] ------------------ 4761| 0| if (strip.compare(0, condl, cond) == 0) ------------------ | Branch (4761:9): [True: 0, False: 0] ------------------ 4762| 0| return 1; 4763| 0| if (utf8) { ------------------ | Branch (4763:9): [True: 0, False: 0] ------------------ 4764| 0| } else { 4765| 0| for (i = 0, j = 0; (i < stripl) && (j < condl); i++, j++) { ------------------ | Branch (4765:26): [True: 0, False: 0] | Branch (4765:42): [True: 0, False: 0] ------------------ 4766| 0| if (cond[j] != '[') { ------------------ | Branch (4766:13): [True: 0, False: 0] ------------------ 4767| 0| if (cond[j] != strip[i]) { ------------------ | Branch (4767:15): [True: 0, False: 0] ------------------ 4768| 0| HUNSPELL_WARNING(stderr, 4769| 0| "warning: line %d: incompatible stripping " 4770| 0| "characters and condition\n", 4771| 0| linenum); 4772| 0| return 0; 4773| 0| } 4774| 0| } else { 4775| 0| neg = (cond[j + 1] == '^') ? 1 : 0; ------------------ | Branch (4775:17): [True: 0, False: 0] ------------------ 4776| 0| in = 0; 4777| 0| do { 4778| 0| j++; 4779| 0| if (strip[i] == cond[j]) ------------------ | Branch (4779:17): [True: 0, False: 0] ------------------ 4780| 0| in = 1; 4781| 0| } while ((j < (condl - 1)) && (cond[j] != ']')); ------------------ | Branch (4781:20): [True: 0, False: 0] | Branch (4781:41): [True: 0, False: 0] ------------------ 4782| 0| if (j == (condl - 1) && (cond[j] != ']')) { ------------------ | Branch (4782:15): [True: 0, False: 0] | Branch (4782:35): [True: 0, False: 0] ------------------ 4783| 0| HUNSPELL_WARNING(stderr, 4784| 0| "error: line %d: missing ] in condition:\n%s\n", 4785| 0| linenum, cond.c_str()); 4786| 0| return 0; 4787| 0| } 4788| 0| if ((!neg && !in) || (neg && in)) { ------------------ | Branch (4788:16): [True: 0, False: 0] | Branch (4788:24): [True: 0, False: 0] | Branch (4788:33): [True: 0, False: 0] | Branch (4788:40): [True: 0, False: 0] ------------------ 4789| 0| HUNSPELL_WARNING(stderr, 4790| 0| "warning: line %d: incompatible stripping " 4791| 0| "characters and condition\n", 4792| 0| linenum); 4793| 0| return 0; 4794| 0| } 4795| 0| } 4796| 0| } 4797| 0| if (j >= condl) ------------------ | Branch (4797:11): [True: 0, False: 0] ------------------ 4798| 0| return 1; 4799| 0| } 4800| 6| } else { // suffix 4801| 6| if ((stripl >= condl) && strip.compare(stripl - condl, std::string::npos, cond) == 0) ------------------ | Branch (4801:9): [True: 6, False: 0] | Branch (4801:30): [True: 6, False: 0] ------------------ 4802| 6| return 1; 4803| 0| if (utf8) { ------------------ | Branch (4803:9): [True: 0, False: 0] ------------------ 4804| 0| } else { 4805| 0| for (i = stripl - 1, j = condl - 1; (i >= 0) && (j >= 0); i--, j--) { ------------------ | Branch (4805:43): [True: 0, False: 0] | Branch (4805:55): [True: 0, False: 0] ------------------ 4806| 0| if (cond[j] != ']') { ------------------ | Branch (4806:13): [True: 0, False: 0] ------------------ 4807| 0| if (cond[j] != strip[i]) { ------------------ | Branch (4807:15): [True: 0, False: 0] ------------------ 4808| 0| HUNSPELL_WARNING(stderr, 4809| 0| "warning: line %d: incompatible stripping " 4810| 0| "characters and condition\n", 4811| 0| linenum); 4812| 0| return 0; 4813| 0| } 4814| 0| } else if (j > 0) { ------------------ | Branch (4814:20): [True: 0, False: 0] ------------------ 4815| 0| in = 0; 4816| 0| do { 4817| 0| j--; 4818| 0| if (strip[i] == cond[j]) ------------------ | Branch (4818:17): [True: 0, False: 0] ------------------ 4819| 0| in = 1; 4820| 0| } while ((j > 0) && (cond[j] != '[')); ------------------ | Branch (4820:20): [True: 0, False: 0] | Branch (4820:31): [True: 0, False: 0] ------------------ 4821| 0| if ((j == 0) && (cond[j] != '[')) { ------------------ | Branch (4821:15): [True: 0, False: 0] | Branch (4821:27): [True: 0, False: 0] ------------------ 4822| 0| HUNSPELL_WARNING(stderr, 4823| 0| "error: line: %d: missing ] in condition:\n%s\n", 4824| 0| linenum, cond.c_str()); 4825| 0| return 0; 4826| 0| } 4827| 0| neg = (cond[j + 1] == '^') ? 1 : 0; ------------------ | Branch (4827:17): [True: 0, False: 0] ------------------ 4828| 0| if ((!neg && !in) || (neg && in)) { ------------------ | Branch (4828:16): [True: 0, False: 0] | Branch (4828:24): [True: 0, False: 0] | Branch (4828:33): [True: 0, False: 0] | Branch (4828:40): [True: 0, False: 0] ------------------ 4829| 0| HUNSPELL_WARNING(stderr, 4830| 0| "warning: line %d: incompatible stripping " 4831| 0| "characters and condition\n", 4832| 0| linenum); 4833| 0| return 0; 4834| 0| } 4835| 0| } 4836| 0| } 4837| 0| if (j < 0) ------------------ | Branch (4837:11): [True: 0, False: 0] ------------------ 4838| 0| return 1; 4839| 0| } 4840| 0| } 4841| 0| return 0; 4842| 6|} _ZN8AffixMgr8isSubsetEPKcS1_: 1069| 115k|inline int AffixMgr::isSubset(const char* s1, const char* s2) { 1070| 159k| while (((*s1 == *s2) || (*s1 == '.')) && (*s1 != '\0') && (*s2 != '\0')) { ------------------ | Branch (1070:11): [True: 116k, False: 43.1k] | Branch (1070:27): [True: 0, False: 43.1k] | Branch (1070:44): [True: 44.0k, False: 72.5k] | Branch (1070:61): [True: 44.0k, False: 0] ------------------ 1071| 44.0k| s1++; 1072| 44.0k| s2++; 1073| 44.0k| } 1074| 115k| return (*s1 == '\0'); 1075| 115k|} _ZN8AffixMgr11isRevSubsetEPKcS1_i: 2714| 461k| int len) { 2715| 923k| while ((len > 0) && (*s1 != '\0') && ((*s1 == *end_of_s2) || (*s1 == '.'))) { ------------------ | Branch (2715:10): [True: 919k, False: 3.58k] | Branch (2715:23): [True: 514k, False: 405k] | Branch (2715:41): [True: 462k, False: 52.2k] | Branch (2715:64): [True: 0, False: 52.2k] ------------------ 2716| 462k| s1++; 2717| 462k| end_of_s2--; 2718| 462k| len--; 2719| 462k| } 2720| 461k| return (*s1 == '\0'); 2721| 461k|} affixmgr.cxx:_ZN12_GLOBAL__N_18mystrdupEPKc: 3363| 620k| char* mystrdup(const char* s) { 3364| 620k| char* d = nullptr; 3365| 620k| if (s) { ------------------ | Branch (3365:9): [True: 620k, False: 0] ------------------ 3366| 620k| size_t sl = strlen(s) + 1; 3367| 620k| d = new char[sl]; 3368| 620k| memcpy(d, s, sl); 3369| 620k| } 3370| 620k| return d; 3371| 620k| } _ZN17entries_containerC2EcP8AffixMgr: 4404| 26| : m_mgr(mgr) 4405| 26| , m_at(at) { 4406| 26| } _ZN17entries_container10initializeEict: 4411| 26| char opts, unsigned short aflag) { 4412| 26| entries.reserve(std::min(numents, 16384)); 4413| | 4414| 26| if (m_at == 'P') { ------------------ | Branch (4414:9): [True: 8, False: 18] ------------------ 4415| 8| entries.push_back(new PfxEntry(m_mgr)); 4416| 18| } else { 4417| 18| entries.push_back(new SfxEntry(m_mgr)); 4418| 18| } 4419| | 4420| 26| entries.back()->opts = opts; 4421| 26| entries.back()->aflag = aflag; 4422| 26| } _ZN17entries_container11first_entryEv: 4435| 42| AffEntry* first_entry() { return entries.empty() ? nullptr : entries[0]; } ------------------ | Branch (4435:36): [True: 0, False: 42] ------------------ _ZN17entries_container9add_entryEc: 4424| 16| AffEntry* add_entry(char opts) { 4425| 16| if (m_at == 'P') { ------------------ | Branch (4425:9): [True: 0, False: 16] ------------------ 4426| 0| entries.push_back(new PfxEntry(m_mgr)); 4427| 16| } else { 4428| 16| entries.push_back(new SfxEntry(m_mgr)); 4429| 16| } 4430| 16| AffEntry* ret = entries.back(); 4431| 16| ret->opts = entries[0]->opts & opts; 4432| 16| return ret; 4433| 16| } _ZN17entries_container5beginEv: 4443| 26| std::vector::iterator begin() { return entries.begin(); } _ZN17entries_container3endEv: 4444| 26| std::vector::iterator end() { return entries.end(); } _ZN17entries_container7releaseEv: 4407| 26| void release() { 4408| 26| entries.clear(); 4409| 26| } _ZN17entries_containerD2Ev: 4437| 26| ~entries_container() { 4438| 26| for (auto& entry : entries) { ------------------ | Branch (4438:22): [True: 0, False: 26] ------------------ 4439| 0| delete entry; 4440| 0| } 4441| 26| } affixmgr.cxx:_ZZN8AffixMgr12get_syllableERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEENK3$_1clE6w_char: 1568| 1.68k| [&](w_char wc) { 1569| 1.68k| return std::binary_search(cpdvowels_utf16.begin(), cpdvowels_utf16.end(), wc); 1570| 1.68k| }); _ZN9guesswordC2Ev: 115| 4.84M| : word(nullptr) 116| 4.84M| , allow(false) 117| 4.84M| , orig(nullptr) 118| 4.84M| { 119| 4.84M| } _ZN8patentryC2Ev: 132| 4| : cond(FLAG_NULL) ------------------ | | 96| 4|#define FLAG_NULL 0x00 ------------------ 133| 4| , cond2(FLAG_NULL) { ------------------ | | 96| 4|#define FLAG_NULL 0x00 ------------------ 134| 4| } csutil.cxx:_ZL16HUNSPELL_WARNINGP8_IO_FILEPKcz: 48| 1.46M|static inline void HUNSPELL_WARNING(FILE*, const char*, ...) {} _ZN8AffEntryC2Ev: 46| 42| : morphcode(nullptr) 47| 42| , contclass(nullptr) 48| 42| , aflag(0) 49| 42| , contclasslen(0) 50| 42| , numconds(0) 51| 42| , opts(0) {} _Z6myopenRNSt3__114basic_ifstreamIcNS_11char_traitsIcEEEEPKcj: 110| 114|{ 111| |#if defined(_WIN32) && defined(_MSC_VER) 112| |#define WIN32_LONG_PATH_PREFIX "\\\\?\\" 113| | if (strncmp(path, WIN32_LONG_PATH_PREFIX, 4) == 0) { 114| | int len = MultiByteToWideChar(CP_UTF8, 0, path, -1, NULL, 0); 115| | wchar_t* buff = new wchar_t[len]; 116| | wchar_t* buff2 = new wchar_t[len]; 117| | MultiByteToWideChar(CP_UTF8, 0, path, -1, buff, len); 118| | if (_wfullpath(buff2, buff, len) != NULL) { 119| | stream.open(buff2, mode); 120| | } 121| | delete [] buff; 122| | delete [] buff2; 123| | } 124| | else 125| |#endif 126| 114| stream.open(path, mode); 127| 114|} _Z6u16_u8RNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEERKNS_6vectorI6w_charNS3_IS8_EEEE: 134| 20.9M|std::string& u16_u8(std::string& dest, const std::vector& src) { 135| 20.9M| dest.clear(); 136| 20.9M| dest.reserve(src.size()); 137| 555M| for (const w_char& wc : src) { ------------------ | Branch (137:25): [True: 555M, False: 20.9M] ------------------ 138| 555M| uint16_t cp = (static_cast(wc.h) << 8) | wc.l; 139| 555M| if (cp < 0x80) { ------------------ | Branch (139:9): [True: 480M, False: 75.0M] ------------------ 140| 480M| dest.push_back(static_cast(cp)); 141| 480M| } else if (cp < 0x800) { ------------------ | Branch (141:16): [True: 2.13M, False: 72.9M] ------------------ 142| 2.13M| dest.push_back(static_cast(0xc0 | (cp >> 6))); 143| 2.13M| dest.push_back(static_cast(0x80 | (cp & 0x3f))); 144| 72.9M| } else { 145| 72.9M| dest.push_back(static_cast(0xe0 | (cp >> 12))); 146| 72.9M| dest.push_back(static_cast(0x80 | ((cp >> 6) & 0x3f))); 147| 72.9M| dest.push_back(static_cast(0x80 | (cp & 0x3f))); 148| 72.9M| } 149| 555M| } 150| 20.9M| return dest; 151| 20.9M|} _Z6u8_u16RNSt3__16vectorI6w_charNS_9allocatorIS1_EEEERKNS_12basic_stringIcNS_11char_traitsIcEENS2_IcEEEEb: 169| 3.64M|int u8_u16(std::vector& dest, const std::string& src, bool only_convert_first_letter) { 170| | // faster to oversize initially, assign to elements and resize to what's used 171| | // than to reserve and push_back 172| 3.64M| dest.resize(only_convert_first_letter ? 1 : src.size()); ------------------ | Branch (172:15): [True: 0, False: 3.64M] ------------------ 173| 3.64M| auto out = dest.begin(); 174| 3.64M| auto p = src.begin(), end = src.end(); 175| | 176| 25.8M| while (p < end) { ------------------ | Branch (176:10): [True: 22.4M, False: 3.42M] ------------------ 177| 22.4M| uint8_t b0 = static_cast(*p); 178| 22.4M| uint16_t cp; 179| | 180| 22.4M| if (b0 < 0x80) { ------------------ | Branch (180:9): [True: 19.2M, False: 3.19M] ------------------ 181| | // 1-byte ASCII 182| 19.2M| cp = b0; 183| 19.2M| } else if (b0 < 0xc0) { ------------------ | Branch (183:16): [True: 471k, False: 2.71M] ------------------ 184| | // continuation byte at lead position 185| 471k| HUNSPELL_WARNING(stderr, 186| 471k| "UTF-8 encoding error. Unexpected continuation bytes " 187| 471k| "in %ld. character position\n%s\n", 188| 471k| static_cast(std::distance(src.begin(), p)), 189| 471k| src.c_str()); 190| 471k| cp = 0xfffd; 191| 2.71M| } else if (b0 < 0xe0) { ------------------ | Branch (191:16): [True: 912k, False: 1.80M] ------------------ 192| | // 2-byte sequence: 110xxxxx 10yyyyyy 193| 912k| if (p + 1 < end && is_utf8_cont(p[1])) { ------------------ | Branch (193:11): [True: 795k, False: 117k] | Branch (193:11): [True: 212k, False: 699k] | Branch (193:26): [True: 212k, False: 582k] ------------------ 194| 212k| cp = ((b0 & 0x1f) << 6) | (static_cast(p[1]) & 0x3f); 195| 212k| ++p; // step past lead; loop bottom steps past cont 196| 699k| } else { 197| 699k| warn_missing_cont(src, p); 198| 699k| cp = 0xfffd; 199| 699k| } 200| 1.80M| } else if (b0 < 0xf0) { ------------------ | Branch (200:16): [True: 1.58M, False: 221k] ------------------ 201| | // 3-byte sequence: 1110xxxx 10yyyyyy 10zzzzzz 202| 1.58M| if (p + 1 < end && is_utf8_cont(p[1])) { ------------------ | Branch (202:11): [True: 1.56M, False: 16.7k] | Branch (202:11): [True: 1.52M, False: 63.9k] | Branch (202:26): [True: 1.52M, False: 47.2k] ------------------ 203| 1.52M| uint8_t b1 = static_cast(p[1]); 204| 1.52M| ++p; // step past lead 205| 1.52M| if (p + 1 < end && is_utf8_cont(p[1])) { ------------------ | Branch (205:13): [True: 1.51M, False: 2.49k] | Branch (205:13): [True: 1.51M, False: 10.5k] | Branch (205:28): [True: 1.51M, False: 8.01k] ------------------ 206| 1.51M| cp = ((b0 & 0x0f) << 12) | ((b1 & 0x3f) << 6) | (static_cast(p[1]) & 0x3f); 207| 1.51M| ++p; // step past first cont; loop bottom steps past second cont 208| 1.51M| } else { 209| 10.5k| warn_missing_cont(src, p); 210| 10.5k| cp = 0xfffd; 211| 10.5k| } 212| 1.52M| } else { 213| 63.9k| warn_missing_cont(src, p); 214| 63.9k| cp = 0xfffd; 215| 63.9k| } 216| 1.58M| } else { 217| | // 4+ byte lead: codepoint >= U+10000, can't fit in w_char 218| 221k| HUNSPELL_WARNING(stderr, 219| 221k| "This UTF-8 encoding can't convert to UTF-16:\n%s\n", 220| 221k| src.c_str()); 221| 221k| out->h = 0xff; 222| 221k| out->l = 0xfd; 223| 221k| ++out; 224| 221k| dest.resize(out - dest.begin()); 225| 221k| return -1; 226| 221k| } 227| | 228| 22.2M| out->h = static_cast(cp >> 8); 229| 22.2M| out->l = static_cast(cp); 230| 22.2M| ++out; 231| 22.2M| if (only_convert_first_letter) ------------------ | Branch (231:9): [True: 0, False: 22.2M] ------------------ 232| 0| break; 233| 22.2M| ++p; // consume lead byte 234| 22.2M| } 235| | 236| 3.42M| int size = static_cast(out - dest.begin()); 237| 3.42M| dest.resize(size); 238| 3.42M| return size; 239| 3.64M|} _Z8mystrsepRKNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEERNS_11__wrap_iterIPKcEE: 254| 1.80k| std::string::const_iterator& start) { 255| 1.80k| auto end = str.end(); 256| | 257| 1.80k| is_any_of op(" \t"); 258| | // don't use isspace() here, the string can be in some random charset 259| | // that's way different than the locale's 260| 1.80k| auto sp = start; 261| 5.87k| while (sp != end && op(*sp)) ------------------ | Branch (261:10): [True: 5.43k, False: 442] | Branch (261:23): [True: 4.07k, False: 1.35k] ------------------ 262| 4.07k| ++sp; 263| | 264| 1.80k| auto dp = sp; 265| 6.69k| while (dp != end && !op(*dp)) ------------------ | Branch (265:10): [True: 5.81k, False: 882] | Branch (265:23): [True: 4.89k, False: 918] ------------------ 266| 4.89k| ++dp; 267| | 268| 1.80k| start = dp; 269| 1.80k| return sp; 270| 1.80k|} _Z7mychompRNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEE: 273| 1.99k|void mychomp(std::string& s) { 274| 1.99k| size_t k = s.size(); 275| 1.99k| size_t newsize = k; 276| 1.99k| if ((k > 0) && ((s[k - 1] == '\r') || (s[k - 1] == '\n'))) ------------------ | Branch (276:7): [True: 1.80k, False: 190] | Branch (276:19): [True: 0, False: 1.80k] | Branch (276:41): [True: 0, False: 1.80k] ------------------ 277| 0| --newsize; 278| 1.99k| if ((k > 1) && (s[k - 2] == '\r')) ------------------ | Branch (278:7): [True: 1.60k, False: 384] | Branch (278:18): [True: 0, False: 1.60k] ------------------ 279| 0| --newsize; 280| 1.99k| s.resize(newsize); 281| 1.99k|} _Z8line_tokRKNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEEc: 284| 293k|std::vector line_tok(const std::string& text, char breakchar) { 285| 293k| std::vector ret; 286| 293k| if (text.empty()) { ------------------ | Branch (286:7): [True: 10.1k, False: 283k] ------------------ 287| 10.1k| return ret; 288| 10.1k| } 289| | 290| 283k| std::stringstream ss(text); 291| 283k| std::string tok; 292| 567k| while(std::getline(ss, tok, breakchar)) { ------------------ | Branch (292:9): [True: 283k, False: 283k] ------------------ 293| 283k| if (!tok.empty()) { ------------------ | Branch (293:9): [True: 283k, False: 128] ------------------ 294| 283k| ret.push_back(tok); 295| 283k| } 296| 283k| } 297| | 298| 283k| return ret; 299| 293k|} _Z9line_uniqRNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEEc: 303| 12.7k|{ 304| 12.7k| std::vector lines = line_tok(text, breakchar); 305| 12.7k| text.clear(); 306| 12.7k| if (lines.empty()) { ------------------ | Branch (306:7): [True: 10.1k, False: 2.58k] ------------------ 307| 10.1k| return; 308| 10.1k| } 309| 2.58k| text = lines[0]; 310| 2.75k| for (size_t i = 1; i < lines.size(); ++i) { ------------------ | Branch (310:22): [True: 172, False: 2.58k] ------------------ 311| 172| bool dup = false; 312| 192| for (size_t j = 0; j < i; ++j) { ------------------ | Branch (312:24): [True: 172, False: 20] ------------------ 313| 172| if (lines[i] == lines[j]) { ------------------ | Branch (313:11): [True: 152, False: 20] ------------------ 314| 152| dup = true; 315| 152| break; 316| 152| } 317| 172| } 318| 172| if (!dup) { ------------------ | Branch (318:9): [True: 20, False: 152] ------------------ 319| 20| if (!text.empty()) ------------------ | Branch (319:11): [True: 20, False: 0] ------------------ 320| 20| text.push_back(breakchar); 321| 20| text.append(lines[i]); 322| 20| } 323| 172| } 324| 2.58k|} _Z10copy_fieldRNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEERKS5_S8_: 388| 265k| const std::string& var) { 389| 265k| if (morph.empty()) ------------------ | Branch (389:7): [True: 0, False: 265k] ------------------ 390| 0| return false; 391| 265k| size_t pos = morph.find(var); 392| 265k| if (pos == std::string::npos) ------------------ | Branch (392:7): [True: 8, False: 265k] ------------------ 393| 8| return false; 394| 265k| dest.clear(); 395| 265k| std::string beg(morph.substr(pos + MORPH_TAG_LEN, std::string::npos)); ------------------ | | 119| 265k|#define MORPH_TAG_LEN strlen(MORPH_STEM) | | ------------------ | | | | 103| 265k|#define MORPH_STEM "st:" | | ------------------ ------------------ 396| | 397| 465k| for (const char c : beg) { ------------------ | Branch (397:21): [True: 465k, False: 265k] ------------------ 398| 465k| if (c == ' ' || c == '\t' || c == '\n') ------------------ | Branch (398:9): [True: 0, False: 465k] | Branch (398:21): [True: 0, False: 465k] | Branch (398:34): [True: 0, False: 465k] ------------------ 399| 0| break; 400| 465k| dest.push_back(c); 401| 465k| } 402| | 403| 265k| return true; 404| 265k|} _Z8mystrrepRNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEERKS5_S8_: 408| 999k| const std::string& replace) { 409| 999k| size_t pos = 0; 410| 1.19M| while ((pos = str.find(search, pos)) != std::string::npos) { ------------------ | Branch (410:10): [True: 192k, False: 999k] ------------------ 411| 192k| str.replace(pos, search.length(), replace); 412| 192k| pos += replace.length(); 413| 192k| } 414| 999k| return str; 415| 999k|} _Z11reversewordRNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEE: 418| 919k|size_t reverseword(std::string& word) { 419| 919k| std::reverse(word.begin(), word.end()); 420| 919k| return word.size(); 421| 919k|} _Z8uniqlistRNSt3__16vectorINS_12basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEENS4_IS6_EEEE: 498| 6.20k|void uniqlist(std::vector& list) { 499| 6.20k| if (list.size() < 2) ------------------ | Branch (499:7): [True: 6.20k, False: 0] ------------------ 500| 6.20k| return; 501| | 502| 0| std::vector ret; 503| 0| ret.push_back(list[0]); 504| | 505| 0| for (size_t i = 1; i < list.size(); ++i) { ------------------ | Branch (505:22): [True: 0, False: 0] ------------------ 506| 0| if (std::find(ret.begin(), ret.end(), list[i]) == ret.end()) ------------------ | Branch (506:9): [True: 0, False: 0] ------------------ 507| 0| ret.push_back(list[i]); 508| 0| } 509| | 510| 0| list.swap(ret); 511| 0|} _Z9upper_utf6w_chari: 530| 3.92M|w_char upper_utf(w_char u, int langnum) { 531| | 532| 3.92M|#if defined(_WIN32) || (defined(__BYTE_ORDER__) && (__BYTE_ORDER__==__ORDER_LITTLE_ENDIAN__)) || defined(__LITTLE_ENDIAN__) 533| | 534| |//with these optimizations, msvc can optimize this function to one jmp instruction 535| |//but g++ remains in five instructions 536| |//maybe use inline asm for g++? 537| | 538| |#if (__cplusplus >= 202002L || (defined(_MSVC_LANG) && _MSVC_LANG >= 202002L)) && defined __cpp_lib_bit_cast && __cpp_lib_bit_cast >= 201806L 539| | return std::bit_cast(unicodetoupper((unsigned short)u, langnum)); 540| |#else 541| 3.92M| const auto us = unicodetoupper((unsigned short)u, langnum); 542| 3.92M| memcpy(&u, &us, sizeof(unsigned short)); 543| 3.92M| return u; 544| 3.92M|#endif 545| | 546| |#else 547| | const auto us = unicodetoupper((unsigned short)u, langnum); 548| | u.h = (unsigned char)(us >> 8); 549| | u.l = (unsigned char)(us & 0xff); 550| | return u; 551| |#endif 552| 3.92M|} _Z9lower_utf6w_chari: 554| 5.38M|w_char lower_utf(w_char u, int langnum) { 555| | 556| 5.38M|#if defined(_WIN32) || (defined(__BYTE_ORDER__) && (__BYTE_ORDER__==__ORDER_LITTLE_ENDIAN__)) || defined(__LITTLE_ENDIAN__) 557| | 558| |//with these optimizations, msvc can optimize this function to one jmp instruction 559| |//but g++ remains in five instructions 560| |//maybe use inline asm for g++? 561| | 562| |#if (__cplusplus >= 202002L || (defined(_MSVC_LANG) && _MSVC_LANG >= 202002L)) && defined __cpp_lib_bit_cast && __cpp_lib_bit_cast >= 201806L 563| | return std::bit_cast(unicodetolower((unsigned short)u, langnum)); 564| |#else 565| 5.38M| const auto us = unicodetolower((unsigned short)u, langnum); 566| 5.38M| memcpy(&u, &us, sizeof(unsigned short)); 567| 5.38M| return u; 568| 5.38M|#endif 569| | 570| |#else 571| | const auto us = unicodetolower((unsigned short)u, langnum); 572| | u.h = (unsigned char)(us >> 8); 573| | u.l = (unsigned char)(us & 0xff); 574| | return u; 575| |#endif 576| 5.38M|} _Z8mkallcapRNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEEPK7cs_info: 579| 268k|std::string& mkallcap(std::string& s, const struct cs_info* csconv) { 580| 5.26M| for (char& aI : s) { ------------------ | Branch (580:17): [True: 5.26M, False: 268k] ------------------ 581| 5.26M| aI = cupper(csconv, static_cast(aI)); 582| 5.26M| } 583| 268k| return s; 584| 268k|} _Z10mkallsmallRNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEEPK7cs_info: 587| 1.61M|std::string& mkallsmall(std::string& s, const struct cs_info* csconv) { 588| 14.6M| for (char& aI : s) { ------------------ | Branch (588:17): [True: 14.6M, False: 1.61M] ------------------ 589| 14.6M| aI = clower(csconv, static_cast(aI)); 590| 14.6M| } 591| 1.61M| return s; 592| 1.61M|} _Z14mkallsmall_utfRNSt3__16vectorI6w_charNS_9allocatorIS1_EEEEi: 594| 919k|std::vector& mkallsmall_utf(std::vector& u, int langnum) { 595| 5.35M| for (auto& i : u) { ------------------ | Branch (595:16): [True: 5.35M, False: 919k] ------------------ 596| 5.35M| i = lower_utf(i, langnum); 597| 5.35M| } 598| 919k| return u; 599| 919k|} _Z12mkallcap_utfRNSt3__16vectorI6w_charNS_9allocatorIS1_EEEEi: 601| 156k|std::vector& mkallcap_utf(std::vector& u, int langnum) { 602| 1.83M| for (auto& i : u) { ------------------ | Branch (602:16): [True: 1.83M, False: 156k] ------------------ 603| 1.83M| i = upper_utf(i, langnum); 604| 1.83M| } 605| 156k| return u; 606| 156k|} _Z9mkinitcapRNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEEPK7cs_info: 608| 717k|std::string& mkinitcap(std::string& s, const struct cs_info* csconv) { 609| 717k| if (!s.empty()) { ------------------ | Branch (609:7): [True: 717k, False: 93] ------------------ 610| 717k| s[0] = cupper(csconv, static_cast(s[0])); 611| 717k| } 612| 717k| return s; 613| 717k|} _Z13mkinitcap_utfRNSt3__16vectorI6w_charNS_9allocatorIS1_EEEEi: 615| 262k|std::vector& mkinitcap_utf(std::vector& u, int langnum) { 616| 262k| if (!u.empty()) { ------------------ | Branch (616:7): [True: 262k, False: 0] ------------------ 617| 262k| u[0] = upper_utf(u[0], langnum); 618| 262k| } 619| 262k| return u; 620| 262k|} _Z11mkinitsmallRNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEEPK7cs_info: 622| 9.75k|std::string& mkinitsmall(std::string& s, const struct cs_info* csconv) { 623| 9.75k| if (!s.empty()) { ------------------ | Branch (623:7): [True: 9.75k, False: 0] ------------------ 624| 9.75k| s[0] = clower(csconv, static_cast(s[0])); 625| 9.75k| } 626| 9.75k| return s; 627| 9.75k|} _Z15mkinitsmall_utfRNSt3__16vectorI6w_charNS_9allocatorIS1_EEEEi: 629| 2.87k|std::vector& mkinitsmall_utf(std::vector& u, int langnum) { 630| 2.87k| if (!u.empty()) { ------------------ | Branch (630:7): [True: 2.87k, False: 0] ------------------ 631| 2.87k| u[0] = lower_utf(u[0], langnum); 632| 2.87k| } 633| 2.87k| return u; 634| 2.87k|} _Z14get_current_csRKNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEE: 2292| 1.85k|const struct cs_info* get_current_cs(const std::string& es) { 2293| 1.85k| char* normalized_encoding = new char[es.size() + 1]; 2294| 1.85k| toAsciiLowerAndRemoveNonAlphanumeric(es.c_str(), normalized_encoding); 2295| | 2296| 1.85k| const struct cs_info* ccs = nullptr; 2297| 1.85k| for (const auto& encd : encds) { ------------------ | Branch (2297:25): [True: 1.85k, False: 0] ------------------ 2298| 1.85k| if (strcmp(normalized_encoding, encd.enc_name) == 0) { ------------------ | Branch (2298:9): [True: 1.85k, False: 0] ------------------ 2299| 1.85k| ccs = encd.cs_table; 2300| 1.85k| break; 2301| 1.85k| } 2302| 1.85k| } 2303| | 2304| 1.85k| delete[] normalized_encoding; 2305| | 2306| 1.85k| if (!ccs) { ------------------ | Branch (2306:7): [True: 0, False: 1.85k] ------------------ 2307| 0| HUNSPELL_WARNING(stderr, 2308| 0| "error: unknown encoding %s: using %s as fallback\n", es.c_str(), 2309| 0| encds[0].enc_name); 2310| 0| ccs = encds[0].cs_table; 2311| 0| } 2312| | 2313| 1.85k| return ccs; 2314| 1.85k|} _Z12get_lang_numRKNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEE: 2444| 4|int get_lang_num(const std::string& lang) { 2445| 72| for (const auto& i : lang2enc) { ------------------ | Branch (2445:22): [True: 72, False: 0] ------------------ 2446| 72| if (strcmp(lang.c_str(), i.lang) == 0) { ------------------ | Branch (2446:9): [True: 4, False: 68] ------------------ 2447| 4| return i.num; 2448| 4| } 2449| 72| } 2450| 0| return LANG_xx; 2451| 4|} _Z14unicodetoupperti: 2453| 19.1M|unsigned short unicodetoupper(unsigned short c, int langnum) { 2454| | // In Azeri and Turkish, I and i dictinct letters: 2455| | // There are a dotless lower case i pair of upper `I', 2456| | // and an upper I with dot pair of lower `i'. 2457| 19.1M| if (c == 0x0069 && ((langnum == LANG_az) || (langnum == LANG_tr) || (langnum == LANG_crh))) ------------------ | Branch (2457:7): [True: 23.3k, False: 19.1M] | Branch (2457:23): [True: 0, False: 23.3k] | Branch (2457:47): [True: 0, False: 23.3k] | Branch (2457:71): [True: 0, False: 23.3k] ------------------ 2458| 0| return 0x0130; 2459| |#ifdef OPENOFFICEORG 2460| | return static_cast(u_toupper(c)); 2461| |#else 2462| |#ifdef MOZILLA_CLIENT 2463| | return ToUpperCase((char16_t)c); 2464| |#else 2465| 19.1M| unsigned short up = utf_pages[utf_page_index[c >> 8]][c & 0xFF].cupper; 2466| 19.1M| return up ? up : c; ------------------ | Branch (2466:10): [True: 3.25M, False: 15.8M] ------------------ 2467| 19.1M|#endif 2468| 19.1M|#endif 2469| 19.1M|} _Z14unicodetolowerti: 2471| 21.5M|unsigned short unicodetolower(unsigned short c, int langnum) { 2472| | // In Azeri and Turkish, I and i dictinct letters: 2473| | // There are a dotless lower case i pair of upper `I', 2474| | // and an upper I with dot pair of lower `i'. 2475| 21.5M| if (c == 0x0049 && ((langnum == LANG_az) || (langnum == LANG_tr) || (langnum == LANG_crh))) ------------------ | Branch (2475:7): [True: 6.22k, False: 21.4M] | Branch (2475:23): [True: 0, False: 6.22k] | Branch (2475:47): [True: 0, False: 6.22k] | Branch (2475:71): [True: 0, False: 6.22k] ------------------ 2476| 0| return 0x0131; 2477| |#ifdef OPENOFFICEORG 2478| | return static_cast(u_tolower(c)); 2479| |#else 2480| |#ifdef MOZILLA_CLIENT 2481| | return ToLowerCase((char16_t)c); 2482| |#else 2483| 21.5M| unsigned short lo = utf_pages[utf_page_index[c >> 8]][c & 0xFF].clower; 2484| 21.5M| return lo ? lo : c; ------------------ | Branch (2484:10): [True: 2.23M, False: 19.2M] ------------------ 2485| 21.5M|#endif 2486| 21.5M|#endif 2487| 21.5M|} _Z11get_captypeRKNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEEPK7cs_info: 2498| 7.58M|int get_captype(const std::string& word, const cs_info* csconv) { 2499| | // now determine the capitalization type of the first nl letters 2500| 7.58M| size_t ncap = 0; 2501| 7.58M| size_t nneutral = 0; 2502| 7.58M| size_t firstcap = 0; 2503| 7.58M| if (csconv == nullptr) ------------------ | Branch (2503:7): [True: 0, False: 7.58M] ------------------ 2504| 0| return NOCAP; ------------------ | | 92| 0|#define NOCAP 0 ------------------ 2505| 89.8M| for (char q : word) { ------------------ | Branch (2505:15): [True: 89.8M, False: 7.58M] ------------------ 2506| 89.8M| const auto nIndex = static_cast(q); 2507| 89.8M| if (ccase(csconv, nIndex)) ------------------ | Branch (2507:9): [True: 3.79M, False: 86.0M] ------------------ 2508| 3.79M| ncap++; 2509| 89.8M| if (cupper(csconv, nIndex) == clower(csconv, nIndex)) ------------------ | Branch (2509:9): [True: 71.7M, False: 18.0M] ------------------ 2510| 71.7M| nneutral++; 2511| 89.8M| } 2512| 7.58M| if (ncap) { ------------------ | Branch (2512:7): [True: 1.36M, False: 6.22M] ------------------ 2513| 1.36M| const auto nIndex = static_cast(word[0]); 2514| 1.36M| firstcap = csconv[nIndex].ccase; 2515| 1.36M| } 2516| | 2517| | // now finally set the captype 2518| 7.58M| if (ncap == 0) { ------------------ | Branch (2518:7): [True: 6.22M, False: 1.36M] ------------------ 2519| 6.22M| return NOCAP; ------------------ | | 92| 6.22M|#define NOCAP 0 ------------------ 2520| 6.22M| } else if ((ncap == 1) && firstcap) { ------------------ | Branch (2520:14): [True: 632k, False: 729k] | Branch (2520:29): [True: 311k, False: 320k] ------------------ 2521| 311k| return INITCAP; ------------------ | | 93| 311k|#define INITCAP 1 ------------------ 2522| 1.05M| } else if ((ncap == word.size()) || ((ncap + nneutral) == word.size())) { ------------------ | Branch (2522:14): [True: 22.9k, False: 1.02M] | Branch (2522:39): [True: 125k, False: 902k] ------------------ 2523| 148k| return ALLCAP; ------------------ | | 94| 148k|#define ALLCAP 2 ------------------ 2524| 902k| } else if ((ncap > 1) && firstcap) { ------------------ | Branch (2524:14): [True: 625k, False: 276k] | Branch (2524:28): [True: 320k, False: 304k] ------------------ 2525| 320k| return HUHINITCAP; ------------------ | | 96| 320k|#define HUHINITCAP 4 ------------------ 2526| 320k| } 2527| 581k| return HUHCAP; ------------------ | | 95| 581k|#define HUHCAP 3 ------------------ 2528| 7.58M|} _Z16get_captype_utf8RKNSt3__16vectorI6w_charNS_9allocatorIS1_EEEEi: 2530| 2.27M|int get_captype_utf8(const std::vector& word, int langnum) { 2531| | // now determine the capitalization type of the first nl letters 2532| 2.27M| size_t ncap = 0; 2533| 2.27M| size_t nneutral = 0; 2534| 2.27M| size_t firstcap = 0; 2535| | 2536| 2.27M| auto it = word.begin(), it_end = word.end(); 2537| 17.3M| while (it != it_end) { ------------------ | Branch (2537:10): [True: 15.0M, False: 2.27M] ------------------ 2538| 15.0M| const auto idx = (unsigned short)*it; 2539| 15.0M| const auto lwridx = unicodetolower(idx, langnum); 2540| 15.0M| if (idx != lwridx) ------------------ | Branch (2540:9): [True: 891k, False: 14.1M] ------------------ 2541| 891k| ncap++; 2542| 15.0M| if (unicodetoupper(idx, langnum) == lwridx) ------------------ | Branch (2542:9): [True: 11.8M, False: 3.18M] ------------------ 2543| 11.8M| nneutral++; 2544| 15.0M| ++it; 2545| 15.0M| } 2546| 2.27M| if (ncap) { ------------------ | Branch (2546:7): [True: 367k, False: 1.90M] ------------------ 2547| 367k| const auto idx = (unsigned short)word[0]; 2548| 367k| firstcap = (idx != unicodetolower(idx, langnum)); 2549| 367k| } 2550| | 2551| | // now finally set the captype 2552| 2.27M| if (ncap == 0) { ------------------ | Branch (2552:7): [True: 1.90M, False: 367k] ------------------ 2553| 1.90M| return NOCAP; ------------------ | | 92| 1.90M|#define NOCAP 0 ------------------ 2554| 1.90M| } else if ((ncap == 1) && firstcap) { ------------------ | Branch (2554:14): [True: 215k, False: 151k] | Branch (2554:29): [True: 122k, False: 92.9k] ------------------ 2555| 122k| return INITCAP; ------------------ | | 93| 122k|#define INITCAP 1 ------------------ 2556| 244k| } else if ((ncap == word.size()) || ((ncap + nneutral) == word.size())) { ------------------ | Branch (2556:14): [True: 4.00k, False: 240k] | Branch (2556:39): [True: 55.8k, False: 184k] ------------------ 2557| 59.8k| return ALLCAP; ------------------ | | 94| 59.8k|#define ALLCAP 2 ------------------ 2558| 184k| } else if ((ncap > 1) && firstcap) { ------------------ | Branch (2558:14): [True: 109k, False: 75.5k] | Branch (2558:28): [True: 54.4k, False: 54.8k] ------------------ 2559| 54.4k| return HUHINITCAP; ------------------ | | 96| 54.4k|#define HUHINITCAP 4 ------------------ 2560| 54.4k| } 2561| 130k| return HUHCAP; ------------------ | | 95| 130k|#define HUHCAP 3 ------------------ 2562| 2.27M|} _Z24remove_ignored_chars_utfRNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEERKNS_6vectorI6w_charNS3_IS8_EEEE: 2566| 512k| const std::vector& ignored_chars) { 2567| 512k| std::vector w; 2568| 512k| std::vector w2; 2569| 512k| u8_u16(w, word); 2570| | 2571| 512k| std::copy_if(w.begin(), w.end(), std::back_inserter(w2), 2572| 512k| [&ignored_chars](w_char wc) { 2573| 512k| return !std::binary_search(ignored_chars.begin(), ignored_chars.end(), wc); 2574| 512k| }); 2575| | 2576| 512k| u16_u8(word, w2); 2577| 512k| return w2.size(); 2578| 512k|} _Z20remove_ignored_charsRNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEERKS5_: 2582| 1.66M| const std::string& ignored_chars) { 2583| 1.66M| word.erase( 2584| 1.66M| std::remove_if(word.begin(), word.end(), is_any_of(ignored_chars)), 2585| 1.66M| word.end()); 2586| 1.66M| return word.size(); 2587| 1.66M|} _Z12parse_stringRKNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEERS5_i: 2589| 90|bool parse_string(const std::string& line, std::string& out, int ln) { 2590| 90| if (!out.empty()) { ------------------ | Branch (2590:7): [True: 0, False: 90] ------------------ 2591| 0| HUNSPELL_WARNING(stderr, "error: line %d: multiple definitions\n", ln); 2592| 0| return false; 2593| 0| } 2594| 90| int i = 0; 2595| 90| int np = 0; 2596| 90| auto iter = line.begin(), start_piece = mystrsep(line, iter); 2597| 270| while (start_piece != line.end()) { ------------------ | Branch (2597:10): [True: 180, False: 90] ------------------ 2598| 180| switch (i) { 2599| 90| case 0: { ------------------ | Branch (2599:7): [True: 90, False: 90] ------------------ 2600| 90| np++; 2601| 90| break; 2602| 0| } 2603| 90| case 1: { ------------------ | Branch (2603:7): [True: 90, False: 90] ------------------ 2604| 90| out.assign(start_piece, iter); 2605| 90| np++; 2606| 90| break; 2607| 0| } 2608| 0| default: ------------------ | Branch (2608:7): [True: 0, False: 180] ------------------ 2609| 0| break; 2610| 180| } 2611| 180| ++i; 2612| 180| start_piece = mystrsep(line, iter); 2613| 180| } 2614| 90| if (np != 2) { ------------------ | Branch (2614:7): [True: 0, False: 90] ------------------ 2615| 0| HUNSPELL_WARNING(stderr, "error: line %d: missing data\n", ln); 2616| 0| return false; 2617| 0| } 2618| 90| return true; 2619| 90|} _Z11parse_arrayRKNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEERS5_RNS_6vectorI6w_charNS3_ISA_EEEEii: 2625| 14| int ln) { 2626| 14| if (!parse_string(line, out, ln)) ------------------ | Branch (2626:7): [True: 0, False: 14] ------------------ 2627| 0| return false; 2628| 14| if (utf8) { ------------------ | Branch (2628:7): [True: 8, False: 6] ------------------ 2629| 8| u8_u16(out_utf16, out); 2630| 8| std::sort(out_utf16.begin(), out_utf16.end()); 2631| 8| } 2632| 14| return true; 2633| 14|} csutil.cxx:_ZL17warn_missing_contRKNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEENS_11__wrap_iterIPKcEE: 153| 774k|static void warn_missing_cont(const std::string& src, std::string::const_iterator p) { 154| | HUNSPELL_WARNING(stderr, 155| 774k| "UTF-8 encoding error. Missing continuation byte in " 156| 774k| "%ld. character position:\n%s\n", 157| 774k| static_cast(std::distance(src.begin(), p)), 158| 774k| src.c_str()); 159| 774k|} csutil.cxx:_ZN12_GLOBAL__N_19is_any_ofC2ERKNSt3__112basic_stringIcNS1_11char_traitsIcEENS1_9allocatorIcEEEE: 244| 1.66M| explicit is_any_of(const std::string& in) : chars(in) {} csutil.cxx:_ZNK12_GLOBAL__N_19is_any_ofclEc: 246| 20.4M| bool operator()(char c) const { return chars.find(c) != std::string::npos; } csutil.cxx:_ZN12_GLOBAL__N_16cupperEPK7cs_infoi: 514| 95.7M|unsigned char cupper(const struct cs_info* csconv, int nIndex) { 515| 95.7M| assert(nIndex >= 0 && nIndex <= 255); ------------------ | Branch (515:3): [True: 95.7M, False: 0] | Branch (515:3): [True: 95.7M, False: 0] | Branch (515:3): [True: 95.7M, False: 0] ------------------ 516| 95.7M| return csconv[nIndex].cupper; 517| 95.7M|} csutil.cxx:_ZN12_GLOBAL__N_16clowerEPK7cs_infoi: 519| 104M|unsigned char clower(const struct cs_info* csconv, int nIndex) { 520| 104M| assert(nIndex >= 0 && nIndex <= 255); ------------------ | Branch (520:3): [True: 104M, False: 0] | Branch (520:3): [True: 104M, False: 0] | Branch (520:3): [True: 104M, False: 0] ------------------ 521| 104M| return csconv[nIndex].clower; 522| 104M|} csutil.cxx:_ZL36toAsciiLowerAndRemoveNonAlphanumericPKcPc: 2272| 1.85k| char* pBuf) { 2273| 16.8k| while (*pName) { ------------------ | Branch (2273:10): [True: 14.9k, False: 1.85k] ------------------ 2274| | /* A-Z */ 2275| 14.9k| if ((*pName >= 0x41) && (*pName <= 0x5A)) { ------------------ | Branch (2275:9): [True: 5.56k, False: 9.37k] | Branch (2275:29): [True: 294, False: 5.27k] ------------------ 2276| 294| *pBuf = (*pName) + 0x20; /* toAsciiLower */ 2277| 294| pBuf++; 2278| 294| } 2279| | /* a-z, 0-9 */ 2280| 14.6k| else if (((*pName >= 0x61) && (*pName <= 0x7A)) || ------------------ | Branch (2280:15): [True: 5.27k, False: 9.37k] | Branch (2280:35): [True: 5.27k, False: 0] ------------------ 2281| 14.5k| ((*pName >= 0x30) && (*pName <= 0x39))) { ------------------ | Branch (2281:15): [True: 9.28k, False: 98] | Branch (2281:35): [True: 9.28k, False: 0] ------------------ 2282| 14.5k| *pBuf = *pName; 2283| 14.5k| pBuf++; 2284| 14.5k| } 2285| | 2286| 14.9k| pName++; 2287| 14.9k| } 2288| | 2289| 1.85k| *pBuf = '\0'; 2290| 1.85k|} csutil.cxx:_ZN12_GLOBAL__N_15ccaseEPK7cs_infoi: 524| 89.8M|unsigned char ccase(const struct cs_info* csconv, int nIndex) { 525| 89.8M| assert(nIndex >= 0 && nIndex <= 255); ------------------ | Branch (525:3): [True: 89.8M, False: 0] | Branch (525:3): [True: 89.8M, False: 0] | Branch (525:3): [True: 89.8M, False: 0] ------------------ 526| 89.8M| return csconv[nIndex].ccase; 527| 89.8M|} csutil.cxx:_ZZ24remove_ignored_chars_utfRNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEERKNS_6vectorI6w_charNS3_IS8_EEEEENK3$_0clES8_: 2572| 3.34M| [&ignored_chars](w_char wc) { 2573| 3.34M| return !std::binary_search(ignored_chars.begin(), ignored_chars.end(), wc); 2574| 3.34M| }); _Z12is_utf8_contc: 143| 57.4M|inline bool is_utf8_cont(char c) { 144| 57.4M| return (static_cast(c) & 0xc0) == 0x80; 145| 57.4M|} _Z9utf8_nextRKNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEEm: 148| 23.3k|inline size_t utf8_next(const std::string& s, size_t pos) { 149| 23.3k| if (pos < s.size()) { ------------------ | Branch (149:7): [True: 23.3k, False: 0] ------------------ 150| 23.3k| ++pos; 151| 27.2k| while (pos < s.size() && is_utf8_cont(s[pos])) ------------------ | Branch (151:12): [True: 24.9k, False: 2.24k] | Branch (151:30): [True: 3.88k, False: 21.0k] ------------------ 152| 3.88k| ++pos; 153| 23.3k| } 154| 23.3k| return pos; 155| 23.3k|} _Z20has_no_ignored_charsRKNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEES7_: 290| 753| const std::string& ignored_chars) { 291| 753| return std::all_of(ignored_chars.begin(), ignored_chars.end(), 292| 753| [&word](char ic) { return word.find(ic) == std::string::npos; }); 293| 753|} _Z11HENTRY_DATAP6hentry: 296| 1.68M|inline char* HENTRY_DATA(struct hentry* h) { 297| 1.68M| char* ret; 298| 1.68M| if (!(h->var & H_OPT)) ------------------ | | 47| 1.68M|#define H_OPT (1 << 0) // is there optional morphological data? ------------------ | Branch (298:7): [True: 1.68M, False: 1.65k] ------------------ 299| 1.68M| ret = nullptr; 300| 1.65k| else if (h->var & H_OPT_ALIASM) ------------------ | | 48| 1.65k|#define H_OPT_ALIASM (1 << 1) // using alias compression? ------------------ | Branch (300:12): [True: 0, False: 1.65k] ------------------ 301| 0| ret = get_stored_pointer(HENTRY_WORD(h) + h->blen + 1); ------------------ | | 54| 0|#define HENTRY_WORD(h) &(h->word[0]) ------------------ 302| 1.65k| else 303| 1.65k| ret = HENTRY_WORD(h) + h->blen + 1; ------------------ | | 54| 1.65k|#define HENTRY_WORD(h) &(h->word[0]) ------------------ 304| 1.68M| return ret; 305| 1.68M|} _Z11HENTRY_FINDP6hentryPKc: 332| 2.73k|inline char* HENTRY_FIND(struct hentry* h, const char* p) { 333| 2.73k| char* data = HENTRY_DATA(h); 334| 2.73k| return data ? strstr(data, p) : nullptr; ------------------ | Branch (334:10): [True: 4, False: 2.73k] ------------------ 335| 2.73k|} _ZZ20has_no_ignored_charsRKNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEES7_ENKUlcE_clEc: 292| 5.62k| [&word](char ic) { return word.find(ic) == std::string::npos; }); _ZN7FileMgrC2EPKcS1_: 84| 114| : hin(nullptr) 85| 114| , linenum(0) { 86| 114| in[0] = '\0'; 87| | 88| 114| if (!file || !strlen(file)) ------------------ | Branch (88:7): [True: 0, False: 114] | Branch (88:16): [True: 0, False: 114] ------------------ 89| 0| return; 90| 114| myopen(fin, file, std::ios_base::in); 91| 114| if (!fin.is_open()) { ------------------ | Branch (91:7): [True: 0, False: 114] ------------------ 92| | // check hzipped file 93| 0| std::string st(file); 94| 0| st.append(HZIP_EXTENSION); ------------------ | | 51| 0|#define HZIP_EXTENSION ".hz" ------------------ 95| 0| hin = new Hunzip(st.c_str(), key); 96| 0| } 97| 114| if (!fin.is_open() && !hin->is_open()) ------------------ | Branch (97:7): [True: 0, False: 114] | Branch (97:25): [True: 0, False: 0] ------------------ 98| 0| fail(MSG_OPEN, file); ------------------ | | 53| 0|#define MSG_OPEN "error: %s: cannot open\n" ------------------ 99| 114|} _ZN7FileMgrD2Ev: 101| 114|FileMgr::~FileMgr() { 102| 114| delete hin; 103| 114|} _ZN7FileMgr7getlineERNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEE: 105| 2.10k|bool FileMgr::getline(std::string& dest) { 106| 2.10k| bool ret = false; 107| 2.10k| ++linenum; 108| 2.10k| if (fin.is_open()) { ------------------ | Branch (108:7): [True: 2.10k, False: 0] ------------------ 109| 2.10k| ret = static_cast(std::getline(fin, dest)); 110| 2.10k| } else if (hin && hin->is_open()) { ------------------ | Branch (110:14): [True: 0, False: 0] | Branch (110:21): [True: 0, False: 0] ------------------ 111| 0| ret = hin->getline(dest); 112| 0| } 113| 2.10k| if (!ret) { ------------------ | Branch (113:7): [True: 112, False: 1.99k] ------------------ 114| 112| --linenum; 115| 112| } 116| 2.10k| return ret; 117| 2.10k|} _ZN7FileMgr10getlinenumEv: 119| 96|int FileMgr::getlinenum() { 120| 96| return linenum; 121| 96|} _ZN7HashMgrC2EPKcS1_S1_: 93| 38| : flag_mode(FLAG_CHAR) 94| 38| , complexprefixes(0) 95| 38| , utf8(0) 96| 38| , forbiddenword(FORBIDDENWORD) // forbidden word signing flag ------------------ | | 127| 38|#define FORBIDDENWORD 65510 ------------------ 97| 38| , langnum(0) 98| 38| , csconv(nullptr) { 99| 38| load_config(apath, key); 100| 38| if (!csconv) ------------------ | Branch (100:7): [True: 38, False: 0] ------------------ 101| 38| csconv = get_current_cs(SPELL_ENCODING); ------------------ | | 99| 38|#define SPELL_ENCODING "ISO8859-1" ------------------ 102| 38| int ec = load_tables(tpath, key); 103| 38| if (ec) { ------------------ | Branch (103:7): [True: 0, False: 38] ------------------ 104| | /* error condition - what should we do here */ 105| 0| fprintf(stderr, "Hash Manager Error : %d\n", ec); 106| 0| free_table(); 107| | //keep table size to 1 to fix possible division with zero 108| 0| tableptr.resize(1, nullptr); 109| 0| } 110| 38|} _ZN7HashMgr13release_flagsEPtb: 112| 213|void HashMgr::release_flags(unsigned short* astr, bool owned) { 113| 213| if (owned) ------------------ | Branch (113:7): [True: 139, False: 74] ------------------ 114| 139| delete[] astr; 115| 213|} _ZNK7HashMgr6lookupEPKcm: 154| 526M|struct hentry* HashMgr::lookup(const char* word, size_t len) const { 155| 526M| struct hentry* dp = tableptr[hash(word, len)]; 156| 526M| if (!dp) ------------------ | Branch (156:7): [True: 472M, False: 53.8M] ------------------ 157| 472M| return nullptr; 158| 113M| for (; dp != nullptr; dp = dp->next) { ------------------ | Branch (158:10): [True: 60.5M, False: 52.5M] ------------------ 159| 60.5M| if (strcmp(word, dp->word) == 0) ------------------ | Branch (159:9): [True: 1.29M, False: 59.2M] ------------------ 160| 1.29M| return dp; 161| 60.5M| } 162| 52.5M| return nullptr; 163| 53.8M|} _ZN7HashMgr8add_wordERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEiPtiPS7_bib: 173| 7.35k| bool own_aff) { 174| | 175| 7.35k| if (al > std::numeric_limits::max()) { ------------------ | Branch (175:7): [True: 0, False: 7.35k] ------------------ 176| 0| HUNSPELL_WARNING(stderr, "error: affix len %d is over max limit\n", al); 177| 0| release_flags(aff, own_aff); 178| 0| return 1; 179| 0| } 180| | 181| 7.35k| const std::string* word = &in_word; 182| 7.35k| const std::string* desc = in_desc; 183| | 184| 7.35k| std::string* word_copy = nullptr; 185| 7.35k| std::string* desc_copy = nullptr; 186| 7.35k| if ((!ignorechars.empty() && !has_no_ignored_chars(in_word, ignorechars)) || complexprefixes) { ------------------ | Branch (186:8): [True: 749, False: 6.60k] | Branch (186:32): [True: 198, False: 551] | Branch (186:80): [True: 451, False: 6.70k] ------------------ 187| 649| word_copy = new std::string(in_word); 188| | 189| 649| if (!ignorechars.empty()) { ------------------ | Branch (189:9): [True: 198, False: 451] ------------------ 190| 198| if (utf8) { ------------------ | Branch (190:11): [True: 57, False: 141] ------------------ 191| 57| wcl = remove_ignored_chars_utf(*word_copy, ignorechars_utf16); 192| 141| } else { 193| 141| remove_ignored_chars(*word_copy, ignorechars); 194| 141| } 195| 198| } 196| | 197| 649| if (complexprefixes) { ------------------ | Branch (197:9): [True: 451, False: 198] ------------------ 198| 451| if (utf8) ------------------ | Branch (198:11): [True: 0, False: 451] ------------------ 199| 0| wcl = reverseword_utf(*word_copy); 200| 451| else 201| 451| reverseword(*word_copy); 202| | 203| 451| if (in_desc && aliasm.empty()) { ------------------ | Branch (203:11): [True: 0, False: 451] | Branch (203:22): [True: 0, False: 0] ------------------ 204| 0| desc_copy = new std::string(*in_desc); 205| | 206| 0| if (complexprefixes) { ------------------ | Branch (206:13): [True: 0, False: 0] ------------------ 207| 0| if (utf8) ------------------ | Branch (207:15): [True: 0, False: 0] ------------------ 208| 0| reverseword_utf(*desc_copy); 209| 0| else 210| 0| reverseword(*desc_copy); 211| 0| } 212| 0| desc = desc_copy; 213| 0| } 214| 451| } 215| | 216| 649| word = word_copy; 217| 649| } 218| | 219| | // limit of hp->blen 220| 7.35k| if (word->size() > std::numeric_limits::max()) { ------------------ | Branch (220:7): [True: 0, False: 7.35k] ------------------ 221| 0| HUNSPELL_WARNING(stderr, "error: word len %ld is over max limit\n", word->size()); 222| 0| delete desc_copy; 223| 0| delete word_copy; 224| 0| release_flags(aff, own_aff); 225| 0| return 1; 226| 0| } 227| | 228| 7.35k| bool upcasehomonym = false; 229| 7.35k| int descl = desc ? (!aliasm.empty() ? sizeof(char*) : desc->size() + 1) : 0; ------------------ | Branch (229:15): [True: 4, False: 7.35k] | Branch (229:23): [True: 0, False: 4] ------------------ 230| | // variable-length hash record with word and optional fields 231| 7.35k| auto hp = 232| 7.35k| (struct hentry*)arena_alloc(sizeof(struct hentry) + word->size() + descl, 233| 7.35k| alignof(struct hentry)); 234| 7.35k| if (!hp) { ------------------ | Branch (234:7): [True: 0, False: 7.35k] ------------------ 235| 0| delete desc_copy; 236| 0| delete word_copy; 237| 0| release_flags(aff, own_aff); 238| 0| return 1; 239| 0| } 240| | 241| 7.35k| char* hpw = hp->word; 242| 7.35k| memcpy(hpw, word->data(), word->size()); 243| 7.35k| hpw[word->size()] = 0; 244| | 245| 7.35k| int i = hash(hpw, word->size()); 246| | 247| 7.35k| hp->blen = (unsigned short)word->size(); 248| 7.35k| hp->clen = (unsigned short)wcl; 249| 7.35k| hp->alen = (short)al; 250| 7.35k| hp->astr = aff; 251| 7.35k| hp->next = nullptr; 252| 7.35k| hp->next_homonym = nullptr; 253| 7.35k| hp->var = (captype == INITCAP) ? H_OPT_INITCAP : 0; ------------------ | | 93| 7.35k|#define INITCAP 1 ------------------ hp->var = (captype == INITCAP) ? H_OPT_INITCAP : 0; ------------------ | | 50| 4.59k|#define H_OPT_INITCAP (1 << 3) // is dictionary word capitalized? ------------------ | Branch (253:13): [True: 4.59k, False: 2.76k] ------------------ 254| 7.35k| if (own_aff) ------------------ | Branch (254:7): [True: 5.04k, False: 2.31k] ------------------ 255| 5.04k| hp->var |= H_OPT_OWNFLAGS; ------------------ | | 51| 5.04k|#define H_OPT_OWNFLAGS (1 << 4) // astr is independently allocated? ------------------ 256| | 257| | // store the description string or its pointer 258| 7.35k| if (desc) { ------------------ | Branch (258:7): [True: 4, False: 7.35k] ------------------ 259| 4| hp->var |= H_OPT; ------------------ | | 47| 4|#define H_OPT (1 << 0) // is there optional morphological data? ------------------ 260| 4| if (!aliasm.empty()) { ------------------ | Branch (260:9): [True: 0, False: 4] ------------------ 261| 0| hp->var |= H_OPT_ALIASM; ------------------ | | 48| 0|#define H_OPT_ALIASM (1 << 1) // using alias compression? ------------------ 262| 0| store_pointer(hpw + word->size() + 1, get_aliasm(atoi(desc->c_str()))); 263| 4| } else { 264| 4| strcpy(hpw + word->size() + 1, desc->c_str()); 265| 4| } 266| 4| if (HENTRY_FIND(hp, MORPH_PHON)) { ------------------ | | 114| 4|#define MORPH_PHON "ph:" ------------------ | Branch (266:9): [True: 2, False: 2] ------------------ 267| 2| hp->var |= H_OPT_PHON; ------------------ | | 49| 2|#define H_OPT_PHON (1 << 2) // is there ph: field in the morphological data? ------------------ 268| | // store ph: fields (pronounciation, misspellings, old orthography etc.) 269| | // of a morphological description in reptable to use in REP replacements. 270| 2| size_t predicted = tableptr.size() / MORPH_PHON_RATIO; ------------------ | | 90| 2|#define MORPH_PHON_RATIO 500 ------------------ 271| 2| if (reptable.capacity() < predicted) ------------------ | Branch (271:11): [True: 2, False: 0] ------------------ 272| 2| reptable.reserve(predicted); 273| 2| std::string fields = HENTRY_DATA(hp); 274| 2| std::string::const_iterator iter = fields.begin(), start_piece = mystrsep(fields, iter); 275| 4| while (start_piece != fields.end()) { ------------------ | Branch (275:14): [True: 2, False: 2] ------------------ 276| 2| if (std::string(start_piece, iter).find(MORPH_PHON) == 0) { ------------------ | | 114| 2|#define MORPH_PHON "ph:" ------------------ | Branch (276:13): [True: 2, False: 0] ------------------ 277| 2| std::string ph = std::string(start_piece, iter).substr(sizeof MORPH_PHON - 1); ------------------ | | 114| 2|#define MORPH_PHON "ph:" ------------------ 278| 2| if (!ph.empty()) { ------------------ | Branch (278:15): [True: 2, False: 0] ------------------ 279| 2| std::vector w; 280| 2| size_t strippatt; 281| 2| std::string wordpart; 282| | // dictionary based REP replacement, separated by "->" 283| | // for example "pretty ph:prity ph:priti->pretti" to handle 284| | // both prity -> pretty and pritier -> prettiest suggestions. 285| 2| if (((strippatt = ph.find("->")) != std::string::npos) && ------------------ | Branch (285:17): [True: 0, False: 2] ------------------ 286| 0| (strippatt > 0) && (strippatt < ph.size() - 2)) { ------------------ | Branch (286:21): [True: 0, False: 0] | Branch (286:40): [True: 0, False: 0] ------------------ 287| 0| wordpart = ph.substr(strippatt + 2); 288| 0| ph.erase(ph.begin() + strippatt, ph.end()); 289| 0| } else 290| 2| wordpart = in_word; 291| | // when the ph: field ends with the character *, 292| | // strip last character of the pattern and the replacement 293| | // to match in REP suggestions also at character changes, 294| | // for example, "pretty ph:prity*" results "prit->prett" 295| | // REP replacement instead of "prity->pretty", to get 296| | // prity->pretty and pritiest->prettiest suggestions. 297| 2| if (ph.at(ph.size()-1) == '*') { ------------------ | Branch (297:17): [True: 0, False: 2] ------------------ 298| 0| strippatt = 1; 299| 0| size_t stripword = 0; 300| 0| if (utf8) { ------------------ | Branch (300:19): [True: 0, False: 0] ------------------ 301| 0| while ((strippatt < ph.size()) && ------------------ | Branch (301:24): [True: 0, False: 0] ------------------ 302| 0| is_utf8_cont(ph.at(ph.size()-strippatt-1))) ------------------ | Branch (302:19): [True: 0, False: 0] ------------------ 303| 0| ++strippatt; 304| 0| while ((stripword < wordpart.size()) && ------------------ | Branch (304:24): [True: 0, False: 0] ------------------ 305| 0| is_utf8_cont(wordpart.at(wordpart.size()-stripword-1))) ------------------ | Branch (305:19): [True: 0, False: 0] ------------------ 306| 0| ++stripword; 307| 0| } 308| 0| ++strippatt; 309| 0| ++stripword; 310| 0| if ((ph.size() > strippatt) && (wordpart.size() > stripword)) { ------------------ | Branch (310:19): [True: 0, False: 0] | Branch (310:46): [True: 0, False: 0] ------------------ 311| 0| ph.erase(ph.size()-strippatt, strippatt); 312| 0| wordpart.erase(wordpart.size()-stripword, stripword); 313| 0| } 314| 0| } 315| | // capitalize lowercase pattern for capitalized words to support 316| | // good suggestions also for capitalized misspellings, eg. 317| | // Wednesday ph:wendsay 318| | // results wendsay -> Wednesday and Wendsay -> Wednesday, too. 319| 2| if (captype == INITCAP) { ------------------ | | 93| 2|#define INITCAP 1 ------------------ | Branch (319:17): [True: 0, False: 2] ------------------ 320| 0| std::string ph_capitalized; 321| 0| if (utf8) { ------------------ | Branch (321:19): [True: 0, False: 0] ------------------ 322| 0| u8_u16(w, ph); 323| 0| if (get_captype_utf8(w, langnum) == NOCAP) { ------------------ | | 92| 0|#define NOCAP 0 ------------------ | Branch (323:21): [True: 0, False: 0] ------------------ 324| 0| mkinitcap_utf(w, langnum); 325| 0| u16_u8(ph_capitalized, w); 326| 0| } 327| 0| } else if (get_captype(ph, csconv) == NOCAP) ------------------ | | 92| 0|#define NOCAP 0 ------------------ | Branch (327:26): [True: 0, False: 0] ------------------ 328| 0| mkinitcap(ph_capitalized, csconv); 329| | 330| 0| if (!ph_capitalized.empty()) { ------------------ | Branch (330:19): [True: 0, False: 0] ------------------ 331| | // add also lowercase word in the case of German or 332| | // Hungarian to support lowercase suggestions lowercased by 333| | // compound word generation or derivational suffixes 334| | // (for example by adjectival suffix "-i" of geographical 335| | // names in Hungarian: 336| | // Massachusetts ph:messzecsuzec 337| | // messzecsuzeci -> massachusettsi (adjective) 338| | // For lowercasing by conditional PFX rules, see 339| | // tests/germancompounding test example or the 340| | // Hungarian dictionary.) 341| 0| if (langnum == LANG_de || langnum == LANG_hu) { ------------------ | Branch (341:21): [True: 0, False: 0] | Branch (341:43): [True: 0, False: 0] ------------------ 342| 0| std::string wordpart_lower(wordpart); 343| 0| if (utf8) { ------------------ | Branch (343:23): [True: 0, False: 0] ------------------ 344| 0| u8_u16(w, wordpart_lower); 345| 0| mkallsmall_utf(w, langnum); 346| 0| u16_u8(wordpart_lower, w); 347| 0| } else { 348| 0| mkallsmall(wordpart_lower, csconv); 349| 0| } 350| 0| reptable.emplace_back(); 351| 0| reptable.back().pattern.assign(ph); 352| 0| reptable.back().outstrings[0].assign(wordpart_lower); 353| 0| } 354| 0| reptable.emplace_back(); 355| 0| reptable.back().pattern.assign(ph_capitalized); 356| 0| reptable.back().outstrings[0].assign(wordpart); 357| 0| } 358| 0| } 359| 2| reptable.emplace_back(); 360| 2| reptable.back().pattern.assign(ph); 361| 2| reptable.back().outstrings[0].assign(wordpart); 362| 2| } 363| 2| } 364| 2| start_piece = mystrsep(fields, iter); 365| 2| } 366| 2| } 367| 4| } 368| | 369| 7.35k| struct hentry* dp = tableptr[i]; 370| 7.35k| if (!dp) { ------------------ | Branch (370:7): [True: 5.89k, False: 1.46k] ------------------ 371| 5.89k| tableptr[i] = hp; 372| 5.89k| delete desc_copy; 373| 5.89k| delete word_copy; 374| 5.89k| return 0; 375| 5.89k| } 376| 1.77k| while (dp->next != nullptr) { ------------------ | Branch (376:10): [True: 338, False: 1.44k] ------------------ 377| 338| if ((!dp->next_homonym) && (strcmp(hp->word, dp->word) == 0)) { ------------------ | Branch (377:9): [True: 338, False: 0] | Branch (377:32): [True: 27, False: 311] ------------------ 378| | // remove hidden onlyupcase homonym 379| 27| if (!onlyupcase) { ------------------ | Branch (379:11): [True: 25, False: 2] ------------------ 380| 25| if ((dp->astr) && TESTAFF(dp->astr, ONLYUPCASEFLAG, dp->alen)) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ | Branch (380:13): [True: 0, False: 25] | Branch (380:13): [True: 0, False: 25] ------------------ 381| 0| release_flags(dp->astr, dp->var & H_OPT_OWNFLAGS); ------------------ | | 51| 0|#define H_OPT_OWNFLAGS (1 << 4) // astr is independently allocated? ------------------ 382| 0| dp->astr = hp->astr; 383| 0| dp->alen = hp->alen; 384| 0| dp->var &= ~H_OPT_OWNFLAGS; ------------------ | | 51| 0|#define H_OPT_OWNFLAGS (1 << 4) // astr is independently allocated? ------------------ 385| 0| dp->var |= (hp->var & H_OPT_OWNFLAGS); ------------------ | | 51| 0|#define H_OPT_OWNFLAGS (1 << 4) // astr is independently allocated? ------------------ 386| 0| arena_free(hp); 387| 0| delete desc_copy; 388| 0| delete word_copy; 389| 0| return 0; 390| 25| } else if (!dp->astr && dp->alen == 0 && ------------------ | Branch (390:20): [True: 25, False: 0] | Branch (390:33): [True: 25, False: 0] ------------------ 391| 25| !hp->astr && hp->alen == 0) { ------------------ | Branch (391:20): [True: 25, False: 0] | Branch (391:33): [True: 25, False: 0] ------------------ 392| | // word already exists with no flags, skip duplicate 393| 25| release_flags(hp->astr, hp->var & H_OPT_OWNFLAGS); ------------------ | | 51| 25|#define H_OPT_OWNFLAGS (1 << 4) // astr is independently allocated? ------------------ 394| 25| arena_free(hp); 395| 25| delete desc_copy; 396| 25| delete word_copy; 397| 25| return 0; 398| 25| } else { 399| 0| dp->next_homonym = hp; 400| 0| } 401| 25| } else { 402| 2| upcasehomonym = true; 403| 2| } 404| 27| } 405| 313| dp = dp->next; 406| 313| } 407| 1.44k| if (strcmp(hp->word, dp->word) == 0) { ------------------ | Branch (407:7): [True: 194, False: 1.24k] ------------------ 408| | // remove hidden onlyupcase homonym 409| 194| if (!onlyupcase) { ------------------ | Branch (409:9): [True: 141, False: 53] ------------------ 410| 141| if ((dp->astr) && TESTAFF(dp->astr, ONLYUPCASEFLAG, dp->alen)) { ------------------ | | 99| 27|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 27, False: 0] | | ------------------ ------------------ | Branch (410:11): [True: 27, False: 114] | Branch (410:11): [True: 27, False: 114] ------------------ 411| 27| release_flags(dp->astr, dp->var & H_OPT_OWNFLAGS); ------------------ | | 51| 27|#define H_OPT_OWNFLAGS (1 << 4) // astr is independently allocated? ------------------ 412| 27| dp->astr = hp->astr; 413| 27| dp->alen = hp->alen; 414| 27| dp->var &= ~H_OPT_OWNFLAGS; ------------------ | | 51| 27|#define H_OPT_OWNFLAGS (1 << 4) // astr is independently allocated? ------------------ 415| 27| dp->var |= (hp->var & H_OPT_OWNFLAGS); ------------------ | | 51| 27|#define H_OPT_OWNFLAGS (1 << 4) // astr is independently allocated? ------------------ 416| 27| arena_free(hp); 417| 27| delete desc_copy; 418| 27| delete word_copy; 419| 27| return 0; 420| 114| } else if (!dp->astr && dp->alen == 0 && ------------------ | Branch (420:18): [True: 114, False: 0] | Branch (420:31): [True: 114, False: 0] ------------------ 421| 114| !hp->astr && hp->alen == 0) { ------------------ | Branch (421:18): [True: 106, False: 8] | Branch (421:31): [True: 106, False: 0] ------------------ 422| | // word already exists with no flags, skip duplicate 423| 106| release_flags(hp->astr, hp->var & H_OPT_OWNFLAGS); ------------------ | | 51| 106|#define H_OPT_OWNFLAGS (1 << 4) // astr is independently allocated? ------------------ 424| 106| arena_free(hp); 425| 106| delete desc_copy; 426| 106| delete word_copy; 427| 106| return 0; 428| 106| } else { 429| 8| dp->next_homonym = hp; 430| 8| } 431| 141| } else { 432| 53| upcasehomonym = true; 433| 53| } 434| 194| } 435| 1.30k| if (!upcasehomonym) { ------------------ | Branch (435:7): [True: 1.25k, False: 55] ------------------ 436| 1.25k| dp->next = hp; 437| 1.25k| } else { 438| | // remove hidden onlyupcase homonym 439| 55| release_flags(hp->astr, hp->var & H_OPT_OWNFLAGS); ------------------ | | 51| 55|#define H_OPT_OWNFLAGS (1 << 4) // astr is independently allocated? ------------------ 440| 55| arena_free(hp); 441| 55| } 442| | 443| 1.30k| delete desc_copy; 444| 1.30k| delete word_copy; 445| 1.30k| return 0; 446| 1.44k|} _ZN7HashMgr27add_hidden_capitalized_wordERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEiPtiPS7_i: 453| 5.20k| int captype) { 454| 5.20k| if (flags == nullptr) ------------------ | Branch (454:7): [True: 5.11k, False: 98] ------------------ 455| 5.11k| flagslen = 0; 456| | 457| | // add inner capitalized forms to handle the following allcap forms: 458| | // Mixed caps: OpenOffice.org -> OPENOFFICE.ORG 459| | // Allcaps with suffixes: CIA's -> CIA'S 460| 5.20k| if (((captype == HUHCAP) || (captype == HUHINITCAP) || ------------------ | | 95| 5.20k|#define HUHCAP 3 ------------------ if (((captype == HUHCAP) || (captype == HUHINITCAP) || ------------------ | | 96| 4.99k|#define HUHINITCAP 4 ------------------ | Branch (460:8): [True: 213, False: 4.99k] | Branch (460:31): [True: 1.92k, False: 3.07k] ------------------ 461| 3.07k| ((captype == ALLCAP) && (flagslen != 0))) && ------------------ | | 94| 3.07k|#define ALLCAP 2 ------------------ | Branch (461:9): [True: 168, False: 2.90k] | Branch (461:32): [True: 16, False: 152] ------------------ 462| 2.15k| !((flagslen != 0) && TESTAFF(flags, forbiddenword, flagslen))) { ------------------ | | 99| 16|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 16] | | ------------------ ------------------ | Branch (462:9): [True: 16, False: 2.13k] ------------------ 463| 2.15k| auto flags2 = (unsigned short*)arena_alloc((flagslen + 1) * sizeof(unsigned short), 464| 2.15k| alignof(unsigned short)); 465| 2.15k| flags2[flagslen] = ONLYUPCASEFLAG; ------------------ | | 128| 2.15k|#define ONLYUPCASEFLAG 65511 ------------------ 466| 2.15k| if (flagslen) { ------------------ | Branch (466:9): [True: 16, False: 2.13k] ------------------ 467| 16| memcpy(flags2, flags, flagslen * sizeof(unsigned short)); 468| 16| std::sort(flags2, flags2 + flagslen + 1); 469| 16| } 470| 2.15k| if (utf8) { ------------------ | Branch (470:9): [True: 315, False: 1.83k] ------------------ 471| 315| std::string st; 472| 315| std::vector w; 473| 315| u8_u16(w, word); 474| 315| mkallsmall_utf(w, langnum); 475| 315| mkinitcap_utf(w, langnum); 476| 315| u16_u8(st, w); 477| 315| return add_word(st, wcl, flags2, flagslen + 1, dp, true, INITCAP, false); ------------------ | | 93| 315|#define INITCAP 1 ------------------ 478| 1.83k| } else { 479| 1.83k| std::string new_word(word); 480| 1.83k| mkallsmall(new_word, csconv); 481| 1.83k| mkinitcap(new_word, csconv); 482| 1.83k| int ret = add_word(new_word, wcl, flags2, flagslen + 1, dp, true, INITCAP, false); ------------------ | | 93| 1.83k|#define INITCAP 1 ------------------ 483| 1.83k| return ret; 484| 1.83k| } 485| 2.15k| } 486| 3.05k| return 0; 487| 5.20k|} _ZN7HashMgr20get_clen_and_captypeERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEPiRNS0_6vectorI6w_charNS4_ISB_EEEE: 490| 5.20k|int HashMgr::get_clen_and_captype(const std::string& word, int* captype, std::vector &workbuf) { 491| 5.20k| int len; 492| 5.20k| if (utf8) { ------------------ | Branch (492:7): [True: 2.48k, False: 2.72k] ------------------ 493| 2.48k| len = u8_u16(workbuf, word); 494| 2.48k| *captype = get_captype_utf8(workbuf, langnum); 495| 2.72k| } else { 496| 2.72k| len = word.size(); 497| 2.72k| *captype = get_captype(word, csconv); 498| 2.72k| } 499| 5.20k| return len; 500| 5.20k|} _ZN7HashMgr20get_clen_and_captypeERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEPi: 502| 5.04k|int HashMgr::get_clen_and_captype(const std::string& word, int* captype) { 503| 5.04k| std::vector workbuf; 504| 5.04k| return get_clen_and_captype(word, captype, workbuf); 505| 5.04k|} _ZN7HashMgr21remove_forbidden_flagERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEE: 528| 7.08k|void HashMgr::remove_forbidden_flag(const std::string& word) { 529| 7.08k| struct hentry* dp = lookup(word.c_str(), word.size()); 530| 7.08k| if (!dp) ------------------ | Branch (530:7): [True: 6.87k, False: 204] ------------------ 531| 6.87k| return; 532| 416| while (dp) { ------------------ | Branch (532:10): [True: 212, False: 204] ------------------ 533| 212| if (dp->astr && TESTAFF(dp->astr, forbiddenword, dp->alen)) { ------------------ | | 99| 34|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 34] | | ------------------ ------------------ | Branch (533:9): [True: 34, False: 178] ------------------ 534| 0| if (dp->alen == 1) { ------------------ | Branch (534:11): [True: 0, False: 0] ------------------ 535| 0| release_flags(dp->astr, dp->var & H_OPT_OWNFLAGS); ------------------ | | 51| 0|#define H_OPT_OWNFLAGS (1 << 4) // astr is independently allocated? ------------------ 536| 0| dp->astr = nullptr; 537| 0| dp->alen = 0; 538| 0| dp->var &= ~H_OPT_OWNFLAGS; ------------------ | | 51| 0|#define H_OPT_OWNFLAGS (1 << 4) // astr is independently allocated? ------------------ 539| 0| } else { 540| 0| auto newflags = new unsigned short[dp->alen - 1]; 541| 0| int j = 0; 542| 0| for (int i = 0; i < dp->alen; i++) { ------------------ | Branch (542:25): [True: 0, False: 0] ------------------ 543| 0| if (dp->astr[i] != forbiddenword) ------------------ | Branch (543:15): [True: 0, False: 0] ------------------ 544| 0| newflags[j++] = dp->astr[i]; 545| 0| } 546| 0| release_flags(dp->astr, dp->var & H_OPT_OWNFLAGS); ------------------ | | 51| 0|#define H_OPT_OWNFLAGS (1 << 4) // astr is independently allocated? ------------------ 547| 0| dp->astr = newflags; 548| 0| dp->alen = (short)j; 549| 0| dp->var |= H_OPT_OWNFLAGS; ------------------ | | 51| 0|#define H_OPT_OWNFLAGS (1 << 4) // astr is independently allocated? ------------------ 550| 0| } 551| 0| } 552| 212| dp = dp->next_homonym; 553| 212| } 554| 204|} _ZN7HashMgr3addERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEE: 557| 5.03k|int HashMgr::add(const std::string& word) { 558| 5.03k| remove_forbidden_flag(word); 559| 5.03k| int captype, al = 0; 560| 5.03k| unsigned short* flags = nullptr; 561| 5.03k| int wcl = get_clen_and_captype(word, &captype); 562| 5.03k| if (add_word(word, wcl, flags, al, nullptr, false, captype, true)) ------------------ | Branch (562:7): [True: 0, False: 5.03k] ------------------ 563| 0| return 1; 564| 5.03k| return add_hidden_capitalized_word(word, wcl, flags, al, nullptr, captype); 565| 5.03k|} _ZN7HashMgr14add_with_affixERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEES8_: 578| 2.04k|int HashMgr::add_with_affix(const std::string& word, const std::string& example) { 579| | // detect captype and modify word length for UTF-8 encoding 580| 2.04k| struct hentry* dp = lookup(example.c_str(), example.size()); 581| 2.04k| remove_forbidden_flag(word); 582| 2.04k| if (dp && dp->astr) { ------------------ | Branch (582:7): [True: 87, False: 1.96k] | Branch (582:13): [True: 16, False: 71] ------------------ 583| 16| int captype; 584| 16| int wcl = get_clen_and_captype(word, &captype); 585| 16| auto flags = new unsigned short[dp->alen]; 586| 16| memcpy(flags, dp->astr, dp->alen * sizeof(unsigned short)); 587| 16| if (add_word(word, wcl, flags, dp->alen, nullptr, false, captype, true)) ------------------ | Branch (587:9): [True: 0, False: 16] ------------------ 588| 0| return 1; 589| 16| return add_hidden_capitalized_word(word, wcl, flags, dp->alen, nullptr, captype); 590| 16| } 591| 2.03k| return 1; 592| 2.04k|} _ZNK7HashMgr14walk_hashtableERiP6hentry: 596| 2.08M|struct hentry* HashMgr::walk_hashtable(int& col, struct hentry* hp) const { 597| 2.08M| if (hp && hp->next != nullptr) ------------------ | Branch (597:7): [True: 2.02M, False: 62.3k] | Branch (597:13): [True: 175k, False: 1.84M] ------------------ 598| 175k| return hp->next; 599| 62.9M| for (col++; col < (int)tableptr.size(); ++col) { ------------------ | Branch (599:15): [True: 62.9M, False: 62.3k] ------------------ 600| 62.9M| if (tableptr[col]) ------------------ | Branch (600:9): [True: 1.84M, False: 61.0M] ------------------ 601| 1.84M| return tableptr[col]; 602| 62.9M| } 603| | // null at end and reset to start 604| 62.3k| col = -1; 605| 62.3k| return nullptr; 606| 1.91M|} _ZN7HashMgr11load_tablesEPKcS1_: 609| 38|int HashMgr::load_tables(const char* tpath, const char* key) { 610| | // open dictionary file 611| 38| FileMgr* dict = new FileMgr(tpath, key); 612| | 613| | // first read the first line of file to get hash table size 614| 38| std::string ts; 615| 38| if (!dict->getline(ts)) { ------------------ | Branch (615:7): [True: 0, False: 38] ------------------ 616| 0| fprintf(stderr, "error: empty dic file %s\n", tpath); 617| 0| delete dict; 618| 0| return 2; 619| 0| } 620| 38| mychomp(ts); 621| | 622| | /* remove byte order mark */ 623| 38| if (ts.compare(0, 3, "\xEF\xBB\xBF", 3) == 0) { ------------------ | Branch (623:7): [True: 0, False: 38] ------------------ 624| 0| ts.erase(0, 3); 625| 0| } 626| | 627| 38| int tablesize = atoi(ts.c_str()); 628| | 629| 38| const int nExtra = 5 + USERWORD; ------------------ | | 57| 38|#define USERWORD 1000 ------------------ 630| |#if !defined(FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION) 631| | // covers full-form uncompressed dictionaries (~7M for uk_UA expanded) 632| | const int max_allowed = 10000000; 633| |#else 634| 38| const int max_allowed = (10000 - 1 - nExtra) / int(sizeof(struct hentry*)); 635| 38|#endif 636| | 637| 38| if (tablesize <= 0 || tablesize >= max_allowed) { ------------------ | Branch (637:7): [True: 0, False: 38] | Branch (637:25): [True: 0, False: 38] ------------------ 638| 0| fprintf(stderr, 639| 0| "error: %s: line 1: missing or bad word count in the dic file\n", 640| 0| tpath); 641| 0| delete dict; 642| 0| return 4; 643| 0| } 644| 38| tablesize += nExtra; 645| 38| if ((tablesize & 1) == 0) ------------------ | Branch (645:7): [True: 18, False: 20] ------------------ 646| 18| tablesize++; 647| | 648| | // allocate the hash table 649| 38| tableptr.resize(tablesize, nullptr); 650| | 651| | // loop through all words on much list and add to hash 652| | // table and create word and affix strings 653| | 654| 38| std::vector workbuf; 655| | 656| 38| int nLineCount(0); 657| 198| while (dict->getline(ts)) { ------------------ | Branch (657:10): [True: 160, False: 38] ------------------ 658| 160| ++nLineCount; 659| 160|#ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION 660| | // limit words loaded to avoid O(n^2) hash chain walk timeout 661| 160| if (nLineCount >= tablesize) ------------------ | Branch (661:9): [True: 0, False: 160] ------------------ 662| 0| break; 663| 160|#endif 664| 160| mychomp(ts); 665| | // split each line into word and morphological description 666| 160| size_t dp_pos = 0; 667| 160| while ((dp_pos = ts.find(':', dp_pos)) != std::string::npos) { ------------------ | Branch (667:12): [True: 4, False: 156] ------------------ 668| 4| if ((dp_pos > 3) && (ts[dp_pos - 3] == ' ' || ts[dp_pos - 3] == '\t')) { ------------------ | Branch (668:11): [True: 4, False: 0] | Branch (668:28): [True: 0, False: 4] | Branch (668:53): [True: 4, False: 0] ------------------ 669| 4| for (dp_pos -= 3; dp_pos > 0 && (ts[dp_pos-1] == ' ' || ts[dp_pos-1] == '\t'); --dp_pos) ------------------ | Branch (669:27): [True: 4, False: 0] | Branch (669:42): [True: 0, False: 4] | Branch (669:65): [True: 0, False: 4] ------------------ 670| 0| ; 671| 4| if (dp_pos == 0) { // missing word ------------------ | Branch (671:13): [True: 0, False: 4] ------------------ 672| 0| dp_pos = std::string::npos; 673| 4| } else { 674| 4| ++dp_pos; 675| 4| } 676| 4| break; 677| 4| } 678| 0| ++dp_pos; 679| 0| } 680| | 681| | // tabulator is the old morphological field separator 682| 160| size_t dp2_pos = ts.find('\t'); 683| 160| if (dp2_pos != std::string::npos && (dp_pos == std::string::npos || dp2_pos < dp_pos)) { ------------------ | Branch (683:9): [True: 4, False: 156] | Branch (683:42): [True: 0, False: 4] | Branch (683:73): [True: 4, False: 0] ------------------ 684| 4| dp_pos = dp2_pos + 1; 685| 4| } 686| | 687| 160| std::string dp; 688| 160| if (dp_pos != std::string::npos) { ------------------ | Branch (688:9): [True: 4, False: 156] ------------------ 689| 4| dp.assign(ts.substr(dp_pos)); 690| 4| ts.resize(dp_pos - 1); 691| 4| } 692| | 693| | // split each line into word and affix char strings 694| | // "\/" signs slash in words (not affix separator) 695| | // "/" at beginning of the line is word character (not affix separator) 696| 160| size_t ap_pos = ts.find('/'); 697| 160| while (ap_pos != std::string::npos) { ------------------ | Branch (697:12): [True: 82, False: 78] ------------------ 698| 82| if (ap_pos == 0) { ------------------ | Branch (698:11): [True: 0, False: 82] ------------------ 699| 0| ++ap_pos; 700| 0| continue; 701| 82| } else if (ts[ap_pos - 1] != '\\') ------------------ | Branch (701:18): [True: 82, False: 0] ------------------ 702| 82| break; 703| | // replace "\/" with "/" 704| 0| ts.erase(ap_pos - 1, 1); 705| 0| ap_pos = ts.find('/', ap_pos); 706| 0| } 707| | 708| 160| unsigned short* flags; 709| 160| int al; 710| 160| if (ap_pos != std::string::npos && ap_pos != ts.size()) { ------------------ | Branch (710:9): [True: 82, False: 78] | Branch (710:40): [True: 82, False: 0] ------------------ 711| 82| std::string ap(ts.substr(ap_pos + 1)); 712| 82| ts.resize(ap_pos); 713| 82| if (!aliasf.empty()) { ------------------ | Branch (713:11): [True: 2, False: 80] ------------------ 714| 2| int index = atoi(ap.c_str()); 715| 2| al = get_aliasf(index, &flags, dict); 716| 2| if (!al) { ------------------ | Branch (716:13): [True: 0, False: 2] ------------------ 717| 0| HUNSPELL_WARNING(stderr, "error: line %d: bad flag vector alias\n", 718| 0| dict->getlinenum()); 719| 0| } 720| 80| } else { 721| 80| al = decode_flags(&flags, ap, dict, /* arena = */ true); 722| 80| if (al == -1) { ------------------ | Branch (722:13): [True: 0, False: 80] ------------------ 723| 0| HUNSPELL_WARNING(stderr, "Can't allocate memory.\n"); 724| 0| delete dict; 725| 0| return 6; 726| 0| } 727| 80| std::sort(flags, flags + al); 728| 80| } 729| 82| } else { 730| 78| al = 0; 731| 78| flags = nullptr; 732| 78| } 733| | 734| 160| int captype; 735| 160| int wcl = get_clen_and_captype(ts, &captype, workbuf); 736| 160| const std::string* dp_str = dp.empty() ? nullptr : &dp; ------------------ | Branch (736:33): [True: 156, False: 4] ------------------ 737| | // add the word and its index plus its capitalized form optionally 738| | // flags are arena-allocated, so own_aff must be false 739| 160| bool own = false; 740| 160| if (add_word(ts, wcl, flags, al, dp_str, false, captype, own) || ------------------ | Branch (740:9): [True: 0, False: 160] ------------------ 741| 160| add_hidden_capitalized_word(ts, wcl, flags, al, dp_str, captype)) { ------------------ | Branch (741:9): [True: 0, False: 160] ------------------ 742| 0| delete dict; 743| 0| return 5; 744| 0| } 745| 160| } 746| | 747| 38| int ret(0); 748| | 749| | // reject ludicrous tablesizes 750| 38| if (tablesize > 8192 + nExtra && tablesize > nLineCount * 10 + nExtra) { ------------------ | Branch (750:7): [True: 0, False: 38] | Branch (750:36): [True: 0, False: 0] ------------------ 751| 0| HUNSPELL_WARNING(stderr, ".dic initial approximate word count line value of %d is too large for %d lines\n", tablesize, nLineCount); 752| 0| ret = 3; 753| 0| } 754| | 755| 38| delete dict; 756| 38| return ret; 757| 38|} _ZNK7HashMgr4hashEPKcm: 761| 526M|int HashMgr::hash(const char* word, size_t len) const { 762| 526M| unsigned long hv = 0; 763| 526M| size_t i = 0; 764| 2.60G| while (i < 4 && i < len) ------------------ | Branch (764:10): [True: 2.09G, False: 503M] | Branch (764:19): [True: 2.07G, False: 22.7M] ------------------ 765| 2.07G| hv = (hv << 8) | word[i++]; 766| 11.2G| while (i < len) { ------------------ | Branch (766:10): [True: 10.7G, False: 526M] ------------------ 767| 10.7G| ROTATE(hv, ROTATE_LEN); ------------------ | | 44| 10.7G| (v) = ((v) << (q)) | (((v) >> (32 - q)) & ((1 << (q)) - 1)); ------------------ 768| 10.7G| hv ^= word[i++]; 769| 10.7G| } 770| 526M| return (unsigned long)hv % tableptr.size(); 771| 526M|} _ZNK7HashMgr12decode_flagsEPPtRKNSt3__112basic_stringIcNS2_11char_traitsIcEENS2_9allocatorIcEEEEP7FileMgr: 773| 16|int HashMgr::decode_flags(unsigned short** result, const std::string& flags, FileMgr* af) const { 774| 16| return decode_flags(result, flags, af, /* arena = */ false); 775| 16|} _ZNK7HashMgr12decode_flagsEPPtRKNSt3__112basic_stringIcNS2_11char_traitsIcEENS2_9allocatorIcEEEEP7FileMgrb: 777| 100|int HashMgr::decode_flags(unsigned short** result, const std::string& flags, FileMgr* af, bool use_arena) const { 778| 100| auto alloc = [&](int n) -> unsigned short* { 779| 100| return use_arena ? (unsigned short*)this->arena_alloc(n * sizeof(unsigned short), 780| 100| alignof(unsigned short)) 781| 100| : new unsigned short[n]; 782| 100| }; 783| 100| int len; 784| 100| if (flags.empty()) { ------------------ | Branch (784:7): [True: 0, False: 100] ------------------ 785| 0| *result = nullptr; 786| 0| return 0; 787| 0| } 788| 100| switch (flag_mode) { 789| 0| case FLAG_LONG: { // two-character flags (1x2yZz -> 1x 2y Zz) ------------------ | Branch (789:5): [True: 0, False: 100] ------------------ 790| 0| len = flags.size(); 791| 0| if ((len & 1) == 1 && af != nullptr) ------------------ | Branch (791:11): [True: 0, False: 0] | Branch (791:29): [True: 0, False: 0] ------------------ 792| 0| HUNSPELL_WARNING(stderr, "error: line %d: bad flagvector\n", 793| 0| af->getlinenum()); 794| 0| len >>= 1; 795| 0| *result = alloc(len); 796| 0| for (int i = 0; i < len; i++) { ------------------ | Branch (796:23): [True: 0, False: 0] ------------------ 797| 0| unsigned short flag = ((unsigned short)((unsigned char)flags[i << 1]) << 8) | 798| 0| ((unsigned short)((unsigned char)flags[(i << 1) | 1])); 799| | 800| 0| (*result)[i] = flag; 801| 0| } 802| 0| break; 803| 0| } 804| 0| case FLAG_NUM: { // decimal numbers separated by comma (4521,23,233 -> 4521 ------------------ | Branch (804:5): [True: 0, False: 100] ------------------ 805| | // 23 233) 806| 0| len = int(1 + std::count_if(flags.begin(), flags.end(), [](char c) { return c == ','; })); 807| 0| *result = alloc(len); 808| 0| unsigned short* dest = *result; 809| 0| const char* src = flags.c_str(); 810| 0| for (size_t p = 0; p < flags.size(); ++p) { ------------------ | Branch (810:26): [True: 0, False: 0] ------------------ 811| 0| if (flags[p] == ',') { ------------------ | Branch (811:13): [True: 0, False: 0] ------------------ 812| 0| int i = atoi(src); 813| 0| if ((i > std::numeric_limits::max() || i < 0) && af != nullptr) { ------------------ | Branch (813:16): [True: 0, False: 0] | Branch (813:66): [True: 0, False: 0] | Branch (813:76): [True: 0, False: 0] ------------------ 814| 0| HUNSPELL_WARNING( 815| 0| stderr, "error: line %d: flag id %d is out of range\n", 816| 0| af->getlinenum(), i); 817| 0| i = 0; 818| 0| } 819| 0| *dest = (unsigned short)i; 820| 0| if (*dest == 0 && af != nullptr) ------------------ | Branch (820:15): [True: 0, False: 0] | Branch (820:29): [True: 0, False: 0] ------------------ 821| 0| HUNSPELL_WARNING(stderr, "error: line %d: 0 is wrong flag id\n", 822| 0| af->getlinenum()); 823| 0| src = flags.c_str() + p + 1; 824| 0| dest++; 825| 0| } 826| 0| } 827| 0| int i = atoi(src); 828| 0| if ((i > std::numeric_limits::max() || i < 0) && af) { ------------------ | Branch (828:12): [True: 0, False: 0] | Branch (828:62): [True: 0, False: 0] | Branch (828:72): [True: 0, False: 0] ------------------ 829| 0| HUNSPELL_WARNING(stderr, 830| 0| "error: line %d: flag id %d is out of range\n", 831| 0| af->getlinenum(), i); 832| 0| i = 0; 833| 0| } 834| 0| *dest = (unsigned short)i; 835| 0| if (*dest == 0 && af) ------------------ | Branch (835:11): [True: 0, False: 0] | Branch (835:25): [True: 0, False: 0] ------------------ 836| 0| HUNSPELL_WARNING(stderr, "error: line %d: 0 is wrong flag id\n", 837| 0| af->getlinenum()); 838| 0| break; 839| 0| } 840| 0| case FLAG_UNI: { // UTF-8 characters ------------------ | Branch (840:5): [True: 0, False: 100] ------------------ 841| 0| std::vector w; 842| 0| u8_u16(w, flags); 843| 0| len = w.size(); 844| 0| *result = alloc(len); 845| 0|#if defined(_WIN32) || (defined(__BYTE_ORDER__) && (__BYTE_ORDER__==__ORDER_LITTLE_ENDIAN__)) || defined(__LITTLE_ENDIAN__) 846| 0| memcpy(*result, w.data(), len * sizeof(unsigned short)); 847| |#else 848| | unsigned short* dest = *result; 849| | for (const w_char wc : w) { 850| | *dest = (unsigned short)wc; 851| | dest++; 852| | } 853| |#endif 854| 0| break; 855| 0| } 856| 100| default: { // Ispell's one-character flags (erfg -> e r f g) ------------------ | Branch (856:5): [True: 100, False: 0] ------------------ 857| 100| len = flags.size(); 858| 100| *result = alloc(len); 859| 100| unsigned short* dest = *result; 860| 118| for (const char flag : flags) { ------------------ | Branch (860:28): [True: 118, False: 100] ------------------ 861| 118| *dest = (unsigned char)flag; 862| 118| dest++; 863| 118| } 864| 100| } 865| 100| } 866| 100| return len; 867| 100|} _ZNK7HashMgr11decode_flagERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEE: 943| 92|unsigned short HashMgr::decode_flag(const std::string& f) const { 944| 92| unsigned short s = 0; 945| 92| int i; 946| 92| switch (flag_mode) { 947| 0| case FLAG_LONG: ------------------ | Branch (947:5): [True: 0, False: 92] ------------------ 948| 0| if (f.size() >= 2) ------------------ | Branch (948:11): [True: 0, False: 0] ------------------ 949| 0| s = ((unsigned short)((unsigned char)f[0]) << 8) | ((unsigned short)((unsigned char)f[1])); 950| 0| break; 951| 0| case FLAG_NUM: ------------------ | Branch (951:5): [True: 0, False: 92] ------------------ 952| 0| i = atoi(f.c_str()); 953| 0| if (i > std::numeric_limits::max() || i < 0) { ------------------ | Branch (953:11): [True: 0, False: 0] | Branch (953:61): [True: 0, False: 0] ------------------ 954| 0| HUNSPELL_WARNING(stderr, "error: flag id %d is out of range\n", i); 955| 0| i = 0; 956| 0| } 957| 0| s = (unsigned short)i; 958| 0| break; 959| 0| case FLAG_UNI: { ------------------ | Branch (959:5): [True: 0, False: 92] ------------------ 960| 0| std::vector w; 961| 0| u8_u16(w, f); 962| 0| if (!w.empty()) ------------------ | Branch (962:11): [True: 0, False: 0] ------------------ 963| 0| s = (unsigned short)w[0]; 964| 0| break; 965| 0| } 966| 92| default: ------------------ | Branch (966:5): [True: 92, False: 0] ------------------ 967| 92| if (!f.empty()) ------------------ | Branch (967:11): [True: 92, False: 0] ------------------ 968| 92| s = (unsigned char)f[0]; 969| 92| } 970| 92| if (s == 0) ------------------ | Branch (970:7): [True: 0, False: 92] ------------------ 971| 0| HUNSPELL_WARNING(stderr, "error: 0 is wrong flag id\n"); 972| 92| return s; 973| 92|} _ZN7HashMgr11load_configEPKcS1_: 1009| 38|int HashMgr::load_config(const char* affpath, const char* key) { 1010| 38| int firstline = 1; 1011| | 1012| | // open the affix file 1013| 38| FileMgr* afflst = new FileMgr(affpath, key); 1014| | 1015| | // read in each line ignoring any that do not 1016| | // start with a known line type indicator 1017| | 1018| 38| std::string line; 1019| 922| while (afflst->getline(line)) { ------------------ | Branch (1019:10): [True: 886, False: 36] ------------------ 1020| 886| mychomp(line); 1021| | 1022| | /* remove byte order mark */ 1023| 886| if (firstline) { ------------------ | Branch (1023:9): [True: 38, False: 848] ------------------ 1024| 38| firstline = 0; 1025| 38| if (line.compare(0, 3, "\xEF\xBB\xBF", 3) == 0) { ------------------ | Branch (1025:11): [True: 0, False: 38] ------------------ 1026| 0| line.erase(0, 3); 1027| 0| } 1028| 38| } 1029| | 1030| | /* parse in the try string */ 1031| 886| if ((line.compare(0, 4, "FLAG", 4) == 0) && line.size() > 4 && isspace(line[4])) { ------------------ | Branch (1031:9): [True: 0, False: 886] | Branch (1031:49): [True: 0, False: 0] | Branch (1031:68): [True: 0, False: 0] ------------------ 1032| 0| if (flag_mode != FLAG_CHAR) { ------------------ | Branch (1032:11): [True: 0, False: 0] ------------------ 1033| 0| HUNSPELL_WARNING(stderr, 1034| 0| "error: line %d: multiple definitions of the FLAG " 1035| 0| "affix file parameter\n", 1036| 0| afflst->getlinenum()); 1037| 0| } 1038| 0| if (line.find("long") != std::string::npos) ------------------ | Branch (1038:11): [True: 0, False: 0] ------------------ 1039| 0| flag_mode = FLAG_LONG; 1040| 0| if (line.find("num") != std::string::npos) ------------------ | Branch (1040:11): [True: 0, False: 0] ------------------ 1041| 0| flag_mode = FLAG_NUM; 1042| 0| if (line.find("UTF-8") != std::string::npos) ------------------ | Branch (1042:11): [True: 0, False: 0] ------------------ 1043| 0| flag_mode = FLAG_UNI; 1044| 0| if (flag_mode == FLAG_CHAR) { ------------------ | Branch (1044:11): [True: 0, False: 0] ------------------ 1045| 0| HUNSPELL_WARNING( 1046| 0| stderr, 1047| 0| "error: line %d: FLAG needs `num', `long' or `UTF-8' parameter\n", 1048| 0| afflst->getlinenum()); 1049| 0| } 1050| 0| } 1051| | 1052| 886| if (line.compare(0, 13, "FORBIDDENWORD", 13) == 0) { ------------------ | Branch (1052:9): [True: 2, False: 884] ------------------ 1053| 2| std::string st; 1054| 2| if (!parse_string(line, st, afflst->getlinenum())) { ------------------ | Branch (1054:11): [True: 0, False: 2] ------------------ 1055| 0| delete afflst; 1056| 0| return 1; 1057| 0| } 1058| 2| forbiddenword = decode_flag(st); 1059| 2| } 1060| | 1061| 886| if (line.compare(0, 3, "SET", 3) == 0) { ------------------ | Branch (1061:9): [True: 18, False: 868] ------------------ 1062| 18| if (!parse_string(line, enc, afflst->getlinenum())) { ------------------ | Branch (1062:11): [True: 0, False: 18] ------------------ 1063| 0| delete afflst; 1064| 0| return 1; 1065| 0| } 1066| 18| if (enc == "UTF-8") { ------------------ | Branch (1066:11): [True: 18, False: 0] ------------------ 1067| 18| utf8 = 1; 1068| 18| } else 1069| 0| csconv = get_current_cs(enc); 1070| 18| } 1071| | 1072| 886| if (line.compare(0, 4, "LANG", 4) == 0) { ------------------ | Branch (1072:9): [True: 2, False: 884] ------------------ 1073| 2| if (!parse_string(line, lang, afflst->getlinenum())) { ------------------ | Branch (1073:11): [True: 0, False: 2] ------------------ 1074| 0| delete afflst; 1075| 0| return 1; 1076| 0| } 1077| 2| langnum = get_lang_num(lang); 1078| 2| } 1079| | 1080| | /* parse in the ignored characters (for example, Arabic optional diacritics 1081| | * characters */ 1082| 886| if (line.compare(0, 6, "IGNORE", 6) == 0) { ------------------ | Branch (1082:9): [True: 4, False: 882] ------------------ 1083| 4| if (!parse_array(line, ignorechars, ignorechars_utf16, ------------------ | Branch (1083:11): [True: 0, False: 4] ------------------ 1084| 4| utf8, afflst->getlinenum())) { 1085| 0| delete afflst; 1086| 0| return 1; 1087| 0| } 1088| 4| } 1089| | 1090| 886| if ((line.compare(0, 2, "AF", 2) == 0) && line.size() > 2 && isspace(line[2])) { ------------------ | Branch (1090:9): [True: 2, False: 884] | Branch (1090:47): [True: 2, False: 0] | Branch (1090:66): [True: 2, False: 0] ------------------ 1091| 2| if (!parse_aliasf(line, afflst)) { ------------------ | Branch (1091:11): [True: 0, False: 2] ------------------ 1092| 0| delete afflst; 1093| 0| return 1; 1094| 0| } 1095| 2| } 1096| | 1097| 886| if ((line.compare(0, 2, "AM", 2) == 0) && line.size() > 2 && isspace(line[2])) { ------------------ | Branch (1097:9): [True: 0, False: 886] | Branch (1097:47): [True: 0, False: 0] | Branch (1097:66): [True: 0, False: 0] ------------------ 1098| 0| if (!parse_aliasm(line, afflst)) { ------------------ | Branch (1098:11): [True: 0, False: 0] ------------------ 1099| 0| delete afflst; 1100| 0| return 1; 1101| 0| } 1102| 0| } 1103| | 1104| 886| if (line.compare(0, 15, "COMPLEXPREFIXES", 15) == 0) ------------------ | Branch (1104:9): [True: 2, False: 884] ------------------ 1105| 2| complexprefixes = 1; 1106| | 1107| | /* parse in the typical fault correcting table */ 1108| 886| if (line.compare(0, 3, "REP", 3) == 0) { ------------------ | Branch (1108:9): [True: 6, False: 880] ------------------ 1109| 6| if (!parse_reptable(line, afflst)) { ------------------ | Branch (1109:11): [True: 0, False: 6] ------------------ 1110| 0| delete afflst; 1111| 0| return 1; 1112| 0| } 1113| 6| } 1114| | 1115| | // don't check the full affix file, yet 1116| 886| if (((line.compare(0, 3, "SFX", 3) == 0) || ------------------ | Branch (1116:10): [True: 42, False: 844] ------------------ 1117| 844| (line.compare(0, 3, "PFX", 3) == 0)) && ------------------ | Branch (1117:10): [True: 24, False: 820] ------------------ 1118| 66| line.size() > 3 && isspace(line[3]) && ------------------ | Branch (1118:13): [True: 66, False: 0] | Branch (1118:32): [True: 66, False: 0] ------------------ 1119| 66| !reptable.empty()) // (REP table is in the end of Afrikaans aff file) ------------------ | Branch (1119:13): [True: 2, False: 64] ------------------ 1120| 2| break; 1121| 886| } 1122| | 1123| 38| delete afflst; 1124| 38| return 0; 1125| 38|} _ZN7HashMgr12parse_aliasfERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEP7FileMgr: 1128| 2|bool HashMgr::parse_aliasf(const std::string& line, FileMgr* af) { 1129| 2| if (!aliasf.empty()) { ------------------ | Branch (1129:7): [True: 0, False: 2] ------------------ 1130| 0| HUNSPELL_WARNING(stderr, "error: line %d: multiple table definitions\n", 1131| 0| af->getlinenum()); 1132| 0| return false; 1133| 0| } 1134| 2| int i = 0, np = 0, numaliasf = 0; 1135| 2| auto iter = line.begin(), start_piece = mystrsep(line, iter); 1136| 6| while (start_piece != line.end()) { ------------------ | Branch (1136:10): [True: 4, False: 2] ------------------ 1137| 4| switch (i) { 1138| 2| case 0: { ------------------ | Branch (1138:7): [True: 2, False: 2] ------------------ 1139| 2| np++; 1140| 2| break; 1141| 0| } 1142| 2| case 1: { ------------------ | Branch (1142:7): [True: 2, False: 2] ------------------ 1143| 2| numaliasf = atoi(std::string(start_piece, iter).c_str()); 1144| 2| if (numaliasf < 1) { ------------------ | Branch (1144:13): [True: 0, False: 2] ------------------ 1145| 0| aliasf.clear(); 1146| 0| aliasflen.clear(); 1147| 0| HUNSPELL_WARNING(stderr, "error: line %d: bad entry number\n", 1148| 0| af->getlinenum()); 1149| 0| return false; 1150| 0| } 1151| 2| aliasf.reserve(std::min(numaliasf, 16384)); 1152| 2| aliasflen.reserve(std::min(numaliasf, 16384)); 1153| 2| np++; 1154| 2| break; 1155| 2| } 1156| 0| default: ------------------ | Branch (1156:7): [True: 0, False: 4] ------------------ 1157| 0| break; 1158| 4| } 1159| 4| ++i; 1160| 4| start_piece = mystrsep(line, iter); 1161| 4| } 1162| 2| if (np != 2) { ------------------ | Branch (1162:7): [True: 0, False: 2] ------------------ 1163| 0| aliasf.clear(); 1164| 0| aliasflen.clear(); 1165| 0| HUNSPELL_WARNING(stderr, "error: line %d: missing data\n", 1166| 0| af->getlinenum()); 1167| 0| return false; 1168| 0| } 1169| | 1170| | /* now parse the numaliasf lines to read in the remainder of the table */ 1171| 6| for (int j = 0; j < numaliasf; ++j) { ------------------ | Branch (1171:19): [True: 4, False: 2] ------------------ 1172| 4| std::string nl; 1173| 4| unsigned short* alias = nullptr; 1174| 4| unsigned aliaslen = 0; 1175| 4| i = 0; 1176| 4| if (af->getline(nl)) { ------------------ | Branch (1176:9): [True: 4, False: 0] ------------------ 1177| 4| mychomp(nl); 1178| 4| iter = nl.begin(); 1179| 4| start_piece = mystrsep(nl, iter); 1180| 4| bool errored = false; 1181| 12| while (!errored && start_piece != nl.end()) { ------------------ | Branch (1181:14): [True: 12, False: 0] | Branch (1181:14): [True: 8, False: 4] | Branch (1181:26): [True: 8, False: 4] ------------------ 1182| 8| switch (i) { 1183| 4| case 0: { ------------------ | Branch (1183:11): [True: 4, False: 4] ------------------ 1184| 4| if (nl.compare(start_piece - nl.begin(), 2, "AF", 2) != 0) { ------------------ | Branch (1184:17): [True: 0, False: 4] ------------------ 1185| 0| errored = true; 1186| 0| break; 1187| 0| } 1188| 4| break; 1189| 4| } 1190| 4| case 1: { ------------------ | Branch (1190:11): [True: 4, False: 4] ------------------ 1191| 4| std::string piece(start_piece, iter); 1192| 4| aliaslen = 1193| 4| (unsigned short)decode_flags(&alias, piece, af, /* arena = */ true); 1194| 4| std::sort(alias, alias + aliaslen); 1195| 4| break; 1196| 4| } 1197| 0| default: ------------------ | Branch (1197:11): [True: 0, False: 8] ------------------ 1198| 0| break; 1199| 8| } 1200| 8| ++i; 1201| 8| start_piece = mystrsep(nl, iter); 1202| 8| } 1203| 4| } 1204| 4| if (!alias) { ------------------ | Branch (1204:9): [True: 0, False: 4] ------------------ 1205| 0| for (int k = 0; k < j; ++k) { ------------------ | Branch (1205:23): [True: 0, False: 0] ------------------ 1206| 0| arena_free(aliasf[k]); 1207| 0| } 1208| 0| aliasf.clear(); 1209| 0| aliasflen.clear(); 1210| 0| HUNSPELL_WARNING(stderr, "error: line %d: table is corrupt\n", 1211| 0| af->getlinenum()); 1212| 0| return false; 1213| 0| } 1214| | 1215| 4| aliasf.push_back(alias); 1216| 4| aliasflen.push_back(aliaslen); 1217| 4| } 1218| 2| return true; 1219| 2|} _ZNK7HashMgr9is_aliasfEv: 1221| 44|int HashMgr::is_aliasf() const { 1222| 44| return !aliasf.empty(); 1223| 44|} _ZNK7HashMgr10get_aliasfEiPPtP7FileMgr: 1225| 4|int HashMgr::get_aliasf(int index, unsigned short** fvec, FileMgr* af) const { 1226| 4| if (index > 0 && static_cast(index) <= aliasflen.size()) { ------------------ | Branch (1226:7): [True: 4, False: 0] | Branch (1226:20): [True: 4, False: 0] ------------------ 1227| 4| *fvec = aliasf[index - 1]; 1228| 4| return aliasflen[index - 1]; 1229| 4| } 1230| 0| HUNSPELL_WARNING(stderr, "error: line %d: bad flag alias index: %d\n", 1231| 0| af->getlinenum(), index); 1232| 0| *fvec = nullptr; 1233| 0| return 0; 1234| 4|} _ZNK7HashMgr9is_aliasmEv: 1332| 32|int HashMgr::is_aliasm() const { 1333| 32| return !aliasm.empty(); 1334| 32|} _ZN7HashMgr14parse_reptableERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEP7FileMgr: 1344| 6|bool HashMgr::parse_reptable(const std::string& line, FileMgr* af) { 1345| 6| if (!reptable.empty()) { ------------------ | Branch (1345:7): [True: 0, False: 6] ------------------ 1346| 0| HUNSPELL_WARNING(stderr, "error: line %d: multiple table definitions\n", 1347| 0| af->getlinenum()); 1348| 0| return false; 1349| 0| } 1350| 6| int numrep = -1, i = 0, np = 0; 1351| 6| auto iter = line.begin(), start_piece = mystrsep(line, iter); 1352| 18| while (start_piece != line.end()) { ------------------ | Branch (1352:10): [True: 12, False: 6] ------------------ 1353| 12| switch (i) { 1354| 6| case 0: { ------------------ | Branch (1354:7): [True: 6, False: 6] ------------------ 1355| 6| np++; 1356| 6| break; 1357| 0| } 1358| 6| case 1: { ------------------ | Branch (1358:7): [True: 6, False: 6] ------------------ 1359| 6| numrep = atoi(std::string(start_piece, iter).c_str()); 1360| 6| if (numrep < 1) { ------------------ | Branch (1360:13): [True: 0, False: 6] ------------------ 1361| 0| HUNSPELL_WARNING(stderr, "error: line %d: incorrect entry number\n", 1362| 0| af->getlinenum()); 1363| 0| return false; 1364| 0| } 1365| 6| reptable.reserve(std::min(numrep, 16384)); 1366| 6| np++; 1367| 6| break; 1368| 6| } 1369| 0| default: ------------------ | Branch (1369:7): [True: 0, False: 12] ------------------ 1370| 0| break; 1371| 12| } 1372| 12| ++i; 1373| 12| start_piece = mystrsep(line, iter); 1374| 12| } 1375| 6| if (np != 2) { ------------------ | Branch (1375:7): [True: 0, False: 6] ------------------ 1376| 0| HUNSPELL_WARNING(stderr, "error: line %d: missing data\n", 1377| 0| af->getlinenum()); 1378| 0| return false; 1379| 0| } 1380| | 1381| | /* now parse the numrep lines to read in the remainder of the table */ 1382| 12| for (int j = 0; j < numrep; ++j) { ------------------ | Branch (1382:19): [True: 6, False: 6] ------------------ 1383| 6| std::string nl; 1384| 6| reptable.emplace_back(); 1385| 6| int type = 0; 1386| 6| if (af->getline(nl)) { ------------------ | Branch (1386:9): [True: 6, False: 0] ------------------ 1387| 6| mychomp(nl); 1388| 6| iter = nl.begin(); 1389| 6| i = 0; 1390| 6| start_piece = mystrsep(nl, iter); 1391| 6| bool errored = false; 1392| 24| while (!errored && start_piece != nl.end()) { ------------------ | Branch (1392:14): [True: 24, False: 0] | Branch (1392:14): [True: 18, False: 6] | Branch (1392:26): [True: 18, False: 6] ------------------ 1393| 18| switch (i) { 1394| 6| case 0: { ------------------ | Branch (1394:11): [True: 6, False: 12] ------------------ 1395| 6| if (nl.compare(start_piece - nl.begin(), 3, "REP", 3) != 0) { ------------------ | Branch (1395:17): [True: 0, False: 6] ------------------ 1396| 0| errored = true; 1397| 0| break; 1398| 0| } 1399| 6| break; 1400| 6| } 1401| 6| case 1: { ------------------ | Branch (1401:11): [True: 6, False: 12] ------------------ 1402| 6| if (*start_piece == '^') ------------------ | Branch (1402:17): [True: 0, False: 6] ------------------ 1403| 0| type = 1; 1404| 6| reptable.back().pattern.assign(start_piece + type, iter); 1405| 6| mystrrep(reptable.back().pattern, "_", " "); 1406| 6| if (!reptable.back().pattern.empty() && reptable.back().pattern[reptable.back().pattern.size() - 1] == '$') { ------------------ | Branch (1406:17): [True: 6, False: 0] | Branch (1406:53): [True: 0, False: 6] ------------------ 1407| 0| type += 2; 1408| 0| reptable.back().pattern.resize(reptable.back().pattern.size() - 1); 1409| 0| } 1410| 6| break; 1411| 6| } 1412| 6| case 2: { ------------------ | Branch (1412:11): [True: 6, False: 12] ------------------ 1413| 6| reptable.back().outstrings[type].assign(start_piece, iter); 1414| 6| mystrrep(reptable.back().outstrings[type], "_", " "); 1415| 6| break; 1416| 6| } 1417| 0| default: ------------------ | Branch (1417:11): [True: 0, False: 18] ------------------ 1418| 0| break; 1419| 18| } 1420| 18| ++i; 1421| 18| start_piece = mystrsep(nl, iter); 1422| 18| } 1423| 6| } 1424| 6| if (reptable.back().pattern.empty() || reptable.back().outstrings[type].empty()) { ------------------ | Branch (1424:9): [True: 0, False: 6] | Branch (1424:44): [True: 0, False: 6] ------------------ 1425| 0| HUNSPELL_WARNING(stderr, "error: line %d: table is corrupt\n", 1426| 0| af->getlinenum()); 1427| 0| reptable.clear(); 1428| 0| return false; 1429| 0| } 1430| 6| } 1431| 6| return true; 1432| 6|} _ZNK7HashMgr12get_reptableEv: 1435| 376k|const std::vector& HashMgr::get_reptable() const { 1436| 376k| return reptable; 1437| 376k|} _ZNK7HashMgr11arena_allocEmm: 1439| 9.59k|void* HashMgr::arena_alloc(size_t num_bytes, size_t alignment) const { 1440| | // Fixed-size 64KB chunks: small enough to avoid significant waste on small 1441| | // dictionaries, large enough to amortize per-chunk malloc overhead on large 1442| | // ones. make_unique throws std::bad_alloc on OOM. 1443| 9.59k| static const size_t MIN_CHUNK_SIZE = 65536; 1444| 9.59k| static const size_t MAX_ALIGNMENT = alignof(max_align_t); 1445| | // Chunk sizes are rounded up to MAX_ALIGNMENT below; with this invariant, 1446| | // any alignment that divides MAX_ALIGNMENT keeps aligned_offset within bounds. 1447| 9.59k| assert(alignment > 0 && alignment <= MAX_ALIGNMENT); ------------------ | Branch (1447:3): [True: 9.59k, False: 0] | Branch (1447:3): [True: 9.59k, False: 0] | Branch (1447:3): [True: 9.59k, False: 0] ------------------ 1448| | // Pad the offset up to the requested alignment before placing this allocation. 1449| | // make_unique returns memory aligned for any scalar, so chunk-start is fine. 1450| 9.59k| size_t aligned_offset = (current_chunk_offset + alignment - 1) & ~(alignment - 1); 1451| 9.59k| if (arena.empty() || current_chunk_size - aligned_offset < num_bytes) { ------------------ | Branch (1451:7): [True: 38, False: 9.55k] | Branch (1451:24): [True: 47, False: 9.50k] ------------------ 1452| | // Round the new chunk's size up to a multiple of MAX_ALIGNMENT so that an 1453| | // oversized num_bytes (>= MIN_CHUNK_SIZE) cannot leave a non-aligned 1454| | // current_chunk_size that would later cause aligned_offset to overshoot. 1455| | // Allocate before mutating current_chunk_size so a throwing make_unique 1456| | // leaves the HashMgr in a consistent state. 1457| 85| size_t new_size = std::max(MIN_CHUNK_SIZE, num_bytes); 1458| 85| new_size = (new_size + MAX_ALIGNMENT - 1) & ~(MAX_ALIGNMENT - 1); 1459| 85| arena.push_back(std::make_unique(new_size)); 1460| 85| current_chunk_size = new_size; 1461| 85| aligned_offset = 0; 1462| 85| } 1463| | 1464| 9.59k| uint8_t* ptr = &arena.back()[aligned_offset]; 1465| 9.59k| current_chunk_offset = aligned_offset + num_bytes; 1466| 9.59k| ++outstanding_arena_allocations; 1467| 9.59k| return ptr; 1468| 9.59k|} _ZNK7HashMgr10arena_freeEPv: 1470| 213|void HashMgr::arena_free(void*) const { 1471| | // The arena vector owns all allocations and frees them in bulk at HashMgr 1472| | // destruction, so this is a no-op for the memory itself. The counter is a 1473| | // memory-safety check: more arena_free calls than arena_alloc calls would 1474| | // indicate a double-free or use-after-free in Hunspell. Abort hard rather 1475| | // than silently desynchronize tracking, even in release builds. 1476| 213| if (outstanding_arena_allocations == 0) { ------------------ | Branch (1476:7): [True: 0, False: 213] ------------------ 1477| 0| std::abort(); 1478| 0| } 1479| 213| --outstanding_arena_allocations; 1480| 213|} hashmgr.cxx:_ZZNK7HashMgr12decode_flagsEPPtRKNSt3__112basic_stringIcNS2_11char_traitsIcEENS2_9allocatorIcEEEEP7FileMgrbENK3$_0clEi: 778| 100| auto alloc = [&](int n) -> unsigned short* { 779| 100| return use_arena ? (unsigned short*)this->arena_alloc(n * sizeof(unsigned short), ------------------ | Branch (779:12): [True: 84, False: 16] ------------------ 780| 84| alignof(unsigned short)) 781| 100| : new unsigned short[n]; 782| 100| }; _ZN12HunspellImplC2EPKcS1_S1_: 191| 38| : affixpath(affpath) { 192| 38| csconv = nullptr; 193| 38| utf8 = 0; 194| 38| complexprefixes = 0; 195| | 196| | /* first set up the hash manager */ 197| 38| m_HMgrs.push_back(std::make_unique(dpath, affpath, key)); 198| | 199| | /* next set up the affix manager */ 200| | /* it needs access to the hash manager lookup methods */ 201| 38| pAMgr = std::make_unique(affpath, m_HMgrs, key); 202| | 203| | /* get the preferred try string and the dictionary */ 204| | /* encoding from the Affix Manager for that dictionary */ 205| 38| std::string try_string = pAMgr->get_try_string(); 206| 38| encoding = pAMgr->get_encoding(); 207| 38| langnum = pAMgr->get_langnum(); 208| 38| utf8 = pAMgr->get_utf8(); 209| 38| if (!utf8) ------------------ | Branch (209:7): [True: 20, False: 18] ------------------ 210| 20| csconv = get_current_cs(encoding); 211| 38| complexprefixes = pAMgr->get_complexprefixes(); 212| 38| wordbreak = pAMgr->get_breaktable(); 213| | 214| | /* and finally set up the suggestion manager */ 215| 38| pSMgr = std::make_unique(try_string, MAXSUGGESTION, pAMgr.get()); ------------------ | | 82| 38|#define MAXSUGGESTION 15 ------------------ 216| 38|} _ZN12HunspellImpl12clean_ignoreERNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEERKS6_: 235| 21.8M| const std::string& src) { 236| 21.8M| dest.clear(); 237| 21.8M| dest.assign(src); 238| 21.8M| const char* ignoredchars = pAMgr ? pAMgr->get_ignore() : nullptr; ------------------ | Branch (238:30): [True: 21.8M, False: 0] ------------------ 239| 21.8M| if (ignoredchars != nullptr) { ------------------ | Branch (239:7): [True: 2.17M, False: 19.6M] ------------------ 240| 2.17M| if (utf8) { ------------------ | Branch (240:9): [True: 512k, False: 1.66M] ------------------ 241| 512k| const std::vector& ignoredchars_utf16 = 242| 512k| pAMgr->get_ignore_utf16(); 243| 512k| remove_ignored_chars_utf(dest, ignoredchars_utf16); 244| 1.66M| } else { 245| 1.66M| remove_ignored_chars(dest, ignoredchars); 246| 1.66M| } 247| 2.17M| } 248| 21.8M|} _ZN12HunspellImpl10cleanword2ERNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEERNS0_6vectorI6w_charNS4_IS9_EEEERKS6_PiPm: 262| 10.3M| size_t* pabbrev) { 263| 10.3M| dest.clear(); 264| 10.3M| dest_utf.clear(); 265| | 266| | // remove IGNORE characters from the string 267| 10.3M| std::string w2; 268| 10.3M| clean_ignore(w2, src); 269| | 270| 10.3M| const char* q = w2.c_str(); 271| 10.3M| int nl = (int)w2.size(); 272| | 273| | // first skip over any leading blanks 274| 10.4M| while (*q == ' ') { ------------------ | Branch (274:10): [True: 131k, False: 10.3M] ------------------ 275| 131k| ++q; 276| 131k| nl--; 277| 131k| } 278| | 279| | // now strip off any trailing periods (recording their presence) 280| 10.3M| *pabbrev = 0; 281| | 282| 12.2M| while ((nl > 0) && (*(q + nl - 1) == '.')) { ------------------ | Branch (282:10): [True: 11.7M, False: 504k] | Branch (282:22): [True: 1.93M, False: 9.85M] ------------------ 283| 1.93M| nl--; 284| 1.93M| (*pabbrev)++; 285| 1.93M| } 286| | 287| | // if no characters are left it can't be capitalized 288| 10.3M| if (nl <= 0) { ------------------ | Branch (288:7): [True: 504k, False: 9.85M] ------------------ 289| 504k| *pcaptype = NOCAP; ------------------ | | 92| 504k|#define NOCAP 0 ------------------ 290| 504k| return 0; 291| 504k| } 292| | 293| 9.85M| dest.append(q, nl); 294| 9.85M| nl = dest.size(); 295| 9.85M| if (utf8) { ------------------ | Branch (295:7): [True: 2.27M, False: 7.57M] ------------------ 296| 2.27M| u8_u16(dest_utf, dest); 297| 2.27M| *pcaptype = get_captype_utf8(dest_utf, langnum); 298| 7.57M| } else { 299| 7.57M| *pcaptype = get_captype(dest, csconv); 300| 7.57M| } 301| 9.85M| return nl; 302| 10.3M|} _ZN12HunspellImpl9cleanwordERNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEERKS6_PiSA_: 307| 5.53k| int* pabbrev) { 308| 5.53k| dest.clear(); 309| 5.53k| const unsigned char* q = (const unsigned char*)src.c_str(); 310| 5.53k| int firstcap = 0, nl = (int)src.size(); 311| | 312| | // first skip over any leading blanks 313| 10.7k| while (*q == ' ') { ------------------ | Branch (313:10): [True: 5.18k, False: 5.53k] ------------------ 314| 5.18k| ++q; 315| 5.18k| nl--; 316| 5.18k| } 317| | 318| | // now strip off any trailing periods (recording their presence) 319| 5.53k| *pabbrev = 0; 320| | 321| 13.8k| while ((nl > 0) && (*(q + nl - 1) == '.')) { ------------------ | Branch (321:10): [True: 13.6k, False: 171] | Branch (321:22): [True: 8.30k, False: 5.36k] ------------------ 322| 8.30k| nl--; 323| 8.30k| (*pabbrev)++; 324| 8.30k| } 325| | 326| | // if no characters are left it can't be capitalized 327| 5.53k| if (nl <= 0) { ------------------ | Branch (327:7): [True: 171, False: 5.36k] ------------------ 328| 171| *pcaptype = NOCAP; ------------------ | | 92| 171|#define NOCAP 0 ------------------ 329| 171| return; 330| 171| } 331| | 332| | // now determine the capitalization type of the first nl letters 333| 5.36k| int ncap = 0; 334| 5.36k| int nneutral = 0; 335| 5.36k| int nc = 0; 336| | 337| 5.36k| if (!utf8) { ------------------ | Branch (337:7): [True: 2.82k, False: 2.53k] ------------------ 338| 445k| while (nl > 0) { ------------------ | Branch (338:12): [True: 442k, False: 2.82k] ------------------ 339| 442k| nc++; 340| 442k| if (csconv[(*q)].ccase) ------------------ | Branch (340:11): [True: 390k, False: 51.4k] ------------------ 341| 390k| ncap++; 342| 442k| if (csconv[(*q)].cupper == csconv[(*q)].clower) ------------------ | Branch (342:11): [True: 43.0k, False: 399k] ------------------ 343| 43.0k| nneutral++; 344| 442k| dest.push_back(*q++); 345| 442k| nl--; 346| 442k| } 347| | // remember to terminate the destination string 348| 2.82k| firstcap = csconv[static_cast(dest[0])].ccase; 349| 2.82k| } else { 350| 2.53k| std::vector t; 351| 2.53k| u8_u16(t, src); 352| 148k| for (auto& wc : t) { ------------------ | Branch (352:19): [True: 148k, False: 2.53k] ------------------ 353| 148k| const auto idx = (unsigned short)wc; 354| 148k| const auto low = unicodetolower(idx, langnum); 355| 148k| if (idx != low) ------------------ | Branch (355:11): [True: 104k, False: 44.5k] ------------------ 356| 104k| ncap++; 357| 148k| if (unicodetoupper(idx, langnum) == low) ------------------ | Branch (357:11): [True: 43.0k, False: 105k] ------------------ 358| 43.0k| nneutral++; 359| 148k| } 360| 2.53k| u16_u8(dest, t); 361| 2.53k| if (ncap) { ------------------ | Branch (361:9): [True: 1.37k, False: 1.15k] ------------------ 362| 1.37k| const auto idx = (unsigned short)t[0]; 363| 1.37k| firstcap = (idx != unicodetolower(idx, langnum)); 364| 1.37k| } 365| 2.53k| } 366| | 367| | // now finally set the captype 368| 5.36k| if (ncap == 0) { ------------------ | Branch (368:7): [True: 2.33k, False: 3.02k] ------------------ 369| 2.33k| *pcaptype = NOCAP; ------------------ | | 92| 2.33k|#define NOCAP 0 ------------------ 370| 3.02k| } else if ((ncap == 1) && firstcap) { ------------------ | Branch (370:14): [True: 1.69k, False: 1.33k] | Branch (370:29): [True: 1.29k, False: 404] ------------------ 371| 1.29k| *pcaptype = INITCAP; ------------------ | | 93| 1.29k|#define INITCAP 1 ------------------ 372| 1.73k| } else if ((ncap == nc) || ((ncap + nneutral) == nc)) { ------------------ | Branch (372:14): [True: 50, False: 1.68k] | Branch (372:30): [True: 339, False: 1.34k] ------------------ 373| 389| *pcaptype = ALLCAP; ------------------ | | 94| 389|#define ALLCAP 2 ------------------ 374| 1.34k| } else if ((ncap > 1) && firstcap) { ------------------ | Branch (374:14): [True: 973, False: 374] | Branch (374:28): [True: 692, False: 281] ------------------ 375| 692| *pcaptype = HUHINITCAP; ------------------ | | 96| 692|#define HUHINITCAP 4 ------------------ 376| 692| } else { 377| 655| *pcaptype = HUHCAP; ------------------ | | 95| 655|#define HUHCAP 3 ------------------ 378| 655| } 379| 5.36k|} _ZN12HunspellImpl8mkallcapERNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEE: 381| 3.37k|void HunspellImpl::mkallcap(std::string& u8) { 382| 3.37k| if (utf8) { ------------------ | Branch (382:7): [True: 1.67k, False: 1.70k] ------------------ 383| 1.67k| std::vector u16; 384| 1.67k| u8_u16(u16, u8); 385| 1.67k| ::mkallcap_utf(u16, langnum); 386| 1.67k| u16_u8(u8, u16); 387| 1.70k| } else { 388| 1.70k| ::mkallcap(u8, csconv); 389| 1.70k| } 390| 3.37k|} _ZN12HunspellImpl11mkallsmall2ERNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEERNS0_6vectorI6w_charNS4_IS9_EEEE: 392| 971k|int HunspellImpl::mkallsmall2(std::string& u8, std::vector& u16) { 393| 971k| if (utf8) { ------------------ | Branch (393:7): [True: 272k, False: 699k] ------------------ 394| 272k| ::mkallsmall_utf(u16, langnum); 395| 272k| u16_u8(u8, u16); 396| 699k| } else { 397| 699k| ::mkallsmall(u8, csconv); 398| 699k| } 399| 971k| return u8.size(); 400| 971k|} _ZN12HunspellImpl11is_keepcaseEPK6hentry: 438| 16.7k|int HunspellImpl::is_keepcase(const hentry* rv) { 439| 16.7k| return pAMgr && rv->astr && pAMgr->get_keepcase() && ------------------ | Branch (439:10): [True: 16.7k, False: 0] | Branch (439:19): [True: 1.60k, False: 15.1k] | Branch (439:31): [True: 149, False: 1.45k] ------------------ 440| 149| TESTAFF(rv->astr, pAMgr->get_keepcase(), rv->alen); ------------------ | | 99| 16.9k|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 32, False: 117] | | ------------------ ------------------ 441| 16.7k|} _ZN12HunspellImpl10insert_sugERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKS7_: 444| 13.2k|void HunspellImpl::insert_sug(std::vector& slst, const std::string& word) { 445| 13.2k| slst.insert(slst.begin(), word); 446| 13.2k|} _ZN12HunspellImpl5spellERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEERNS0_6vectorIS6_NS4_IS6_EEEEPiPS6_NS0_6chrono10time_pointINSF_12steady_clockENSF_8durationIxNS0_5ratioILl1ELl1000000000EEEEEEE: 450| 10.3M| std::chrono::steady_clock::time_point suggest_start) { 451| | // something very broken if spell ends up calling itself with the same word 452| 10.3M| if (std::find(candidate_stack.begin(), candidate_stack.end(), word) != candidate_stack.end()) ------------------ | Branch (452:7): [True: 0, False: 10.3M] ------------------ 453| 0| return false; 454| | 455| 10.3M| if (candidate_stack.size() >= MAXBREAKDEPTH) ------------------ | | 86| 10.3M|#define MAXBREAKDEPTH 10 ------------------ | Branch (455:7): [True: 0, False: 10.3M] ------------------ 456| 0| return false; 457| | 458| 10.3M| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 10.3M|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (458:7): [True: 0, False: 10.3M] ------------------ 459| 0| return false; 460| | 461| 10.3M| candidate_stack.push_back(word); 462| 10.3M| bool r = spell_internal(word, candidate_stack, info, root, suggest_start); 463| 10.3M| candidate_stack.pop_back(); 464| | 465| 10.3M| if (r && root) { ------------------ | Branch (465:7): [True: 2.14M, False: 8.21M] | Branch (465:12): [True: 0, False: 2.14M] ------------------ 466| | // output conversion 467| 0| RepList* rl = (pAMgr) ? pAMgr->get_oconvtable() : nullptr; ------------------ | Branch (467:19): [True: 0, False: 0] ------------------ 468| 0| if (rl) { ------------------ | Branch (468:9): [True: 0, False: 0] ------------------ 469| 0| std::string wspace; 470| 0| if (rl->conv(*root, wspace)) { ------------------ | Branch (470:11): [True: 0, False: 0] ------------------ 471| 0| *root = std::move(wspace); 472| 0| } 473| 0| } 474| 0| } 475| 10.3M| return r; 476| 10.3M|} _ZN12HunspellImpl14spell_internalERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEERNS0_6vectorIS6_NS4_IS6_EEEEPiPS6_NS0_6chrono10time_pointINSF_12steady_clockENSF_8durationIxNS0_5ratioILl1ELl1000000000EEEEEEE: 480| 10.3M| std::chrono::steady_clock::time_point suggest_start) { 481| 10.3M| struct hentry* rv = nullptr; 482| | 483| 10.3M| int info2 = 0; 484| 10.3M| if (!info) ------------------ | Branch (484:7): [True: 10.3M, False: 35] ------------------ 485| 10.3M| info = &info2; 486| 35| else 487| 35| *info = 0; 488| | 489| | // Hunspell supports XML input of the simplified API (see manual) 490| 10.3M| if (word == SPELL_XML) ------------------ | | 79| 10.3M|#define SPELL_XML "" ------------------ | Branch (490:7): [True: 69.2k, False: 10.2M] ------------------ 491| 69.2k| return true; 492| 10.2M| if (utf8) { ------------------ | Branch (492:7): [True: 2.33M, False: 7.94M] ------------------ 493| 2.33M| if (word.size() >= MAXWORDUTF8LEN) ------------------ | | 86| 2.33M|#define MAXWORDUTF8LEN (MAXWORDLEN * 3) | | ------------------ | | | | 89| 2.33M|#define MAXWORDLEN 100 | | ------------------ ------------------ | Branch (493:9): [True: 3.73k, False: 2.33M] ------------------ 494| 3.73k| return false; 495| 7.94M| } else { 496| 7.94M| if (word.size() >= MAXWORDLEN) ------------------ | | 89| 7.94M|#define MAXWORDLEN 100 ------------------ | Branch (496:9): [True: 6.18k, False: 7.94M] ------------------ 497| 6.18k| return false; 498| 7.94M| } 499| 10.2M| int captype = NOCAP; ------------------ | | 92| 10.2M|#define NOCAP 0 ------------------ 500| 10.2M| size_t abbv = 0; 501| 10.2M| size_t wl = 0; 502| | 503| 10.2M| std::string scw; 504| 10.2M| std::vector sunicw; 505| | 506| | // input conversion 507| 10.2M| RepList* rl = pAMgr ? pAMgr->get_iconvtable() : nullptr; ------------------ | Branch (507:17): [True: 10.2M, False: 0] ------------------ 508| 10.2M| { 509| 10.2M| std::string wspace; 510| | 511| 10.2M| bool convstatus = rl ? rl->conv(word, wspace) : false; ------------------ | Branch (511:23): [True: 340k, False: 9.93M] ------------------ 512| 10.2M| if (convstatus) ------------------ | Branch (512:9): [True: 1, False: 10.2M] ------------------ 513| 1| wl = cleanword2(scw, sunicw, wspace, &captype, &abbv); 514| 10.2M| else 515| 10.2M| wl = cleanword2(scw, sunicw, word, &captype, &abbv); 516| 10.2M| } 517| | 518| 10.2M|#if defined(FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION) 519| 10.2M| if (wl > 32768) ------------------ | Branch (519:9): [True: 0, False: 10.2M] ------------------ 520| 0| return false; 521| 10.2M|#endif 522| | 523| |#ifdef MOZILLA_CLIENT 524| | // accept the abbreviated words without dots 525| | // workaround for the incomplete tokenization of Mozilla 526| | abbv = 1; 527| |#endif 528| | 529| 10.2M| if (wl == 0 || m_HMgrs.empty()) ------------------ | Branch (529:7): [True: 504k, False: 9.77M] | Branch (529:18): [True: 0, False: 9.77M] ------------------ 530| 504k| return true; 531| 9.77M| if (root) ------------------ | Branch (531:7): [True: 0, False: 9.77M] ------------------ 532| 0| root->clear(); 533| | 534| | // allow numbers with dots, dashes and commas (but forbid double separators: 535| | // "..", "--" etc.) 536| 9.77M| enum { NBEGIN, NNUM, NSEP }; 537| 9.77M| int nstate = NBEGIN; 538| 9.77M| size_t i; 539| | 540| 15.0M| for (i = 0; (i < wl); i++) { ------------------ | Branch (540:15): [True: 14.2M, False: 737k] ------------------ 541| 14.2M| if ((scw[i] <= '9') && (scw[i] >= '0')) { ------------------ | Branch (541:9): [True: 11.1M, False: 3.08M] | Branch (541:28): [True: 4.76M, False: 6.41M] ------------------ 542| 4.76M| nstate = NNUM; 543| 9.49M| } else if ((scw[i] == ',') || (scw[i] == '.') || (scw[i] == '-')) { ------------------ | Branch (543:16): [True: 99.6k, False: 9.39M] | Branch (543:35): [True: 946k, False: 8.45M] | Branch (543:54): [True: 1.97M, False: 6.48M] ------------------ 544| 3.01M| if ((nstate == NSEP) || (i == 0)) ------------------ | Branch (544:11): [True: 104k, False: 2.91M] | Branch (544:31): [True: 2.45M, False: 462k] ------------------ 545| 2.55M| break; 546| 462k| nstate = NSEP; 547| 462k| } else 548| 6.48M| break; 549| 14.2M| } 550| 9.77M| if ((i == wl) && (nstate == NNUM)) ------------------ | Branch (550:7): [True: 737k, False: 9.03M] | Branch (550:20): [True: 579k, False: 158k] ------------------ 551| 579k| return true; 552| | 553| 9.19M| switch (captype) { ------------------ | Branch (553:11): [True: 9.19M, False: 0] ------------------ 554| 693k| case HUHCAP: ------------------ | | 95| 693k|#define HUHCAP 3 ------------------ | Branch (554:5): [True: 693k, False: 8.49M] ------------------ 555| | /* FALLTHROUGH */ 556| 1.05M| case HUHINITCAP: ------------------ | | 96| 1.05M|#define HUHINITCAP 4 ------------------ | Branch (556:5): [True: 359k, False: 8.83M] ------------------ 557| 1.05M| *info |= SPELL_ORIGCAP; ------------------ | | 83| 1.05M|#define SPELL_ORIGCAP (1 << 5) ------------------ 558| | /* FALLTHROUGH */ 559| 8.58M| case NOCAP: ------------------ | | 92| 8.58M|#define NOCAP 0 ------------------ | Branch (559:5): [True: 7.52M, False: 1.66M] ------------------ 560| 8.58M| rv = checkword(scw, info, root, suggest_start); 561| 8.58M| if ((abbv) && !(rv)) { ------------------ | Branch (561:11): [True: 1.17M, False: 7.41M] | Branch (561:21): [True: 1.16M, False: 2.78k] ------------------ 562| 1.16M| std::string u8buffer(scw); 563| 1.16M| u8buffer.push_back('.'); 564| 1.16M| rv = checkword(u8buffer, info, root, suggest_start); 565| 1.16M| } 566| 8.58M| break; 567| 196k| case ALLCAP: { ------------------ | | 94| 196k|#define ALLCAP 2 ------------------ | Branch (567:5): [True: 196k, False: 8.99M] ------------------ 568| 196k| *info |= SPELL_ORIGCAP; ------------------ | | 83| 196k|#define SPELL_ORIGCAP (1 << 5) ------------------ 569| 196k| rv = checkword(scw, info, root, suggest_start); 570| 196k| if (rv) ------------------ | Branch (570:11): [True: 22, False: 196k] ------------------ 571| 22| break; 572| 196k| if (abbv) { ------------------ | Branch (572:11): [True: 15.5k, False: 180k] ------------------ 573| 15.5k| std::string u8buffer(scw); 574| 15.5k| u8buffer.push_back('.'); 575| 15.5k| rv = checkword(u8buffer, info, root, suggest_start); 576| 15.5k| if (rv) ------------------ | Branch (576:13): [True: 0, False: 15.5k] ------------------ 577| 0| break; 578| 15.5k| } 579| | // Spec. prefix handling for Catalan, French, Italian: 580| | // prefixes separated by apostrophe (SANT'ELIA -> Sant'+Elia). 581| 196k| size_t apos = pAMgr ? scw.find('\'') : std::string::npos; ------------------ | Branch (581:21): [True: 196k, False: 0] ------------------ 582| 196k| if (apos != std::string::npos) { ------------------ | Branch (582:11): [True: 76.3k, False: 119k] ------------------ 583| 76.3k| mkallsmall2(scw, sunicw); 584| | //conversion may result in string with different len to pre-mkallsmall2 585| | //so re-scan 586| 76.3k| if (apos != std::string::npos && apos < scw.size() - 1) { ------------------ | Branch (586:13): [True: 76.3k, False: 0] | Branch (586:42): [True: 69.8k, False: 6.51k] ------------------ 587| 69.8k| std::string part1 = scw.substr(0, apos + 1), part2 = scw.substr(apos + 1); 588| 69.8k| if (utf8) { ------------------ | Branch (588:15): [True: 15.0k, False: 54.7k] ------------------ 589| 15.0k| std::vector part1u, part2u; 590| 15.0k| u8_u16(part1u, part1); 591| 15.0k| u8_u16(part2u, part2); 592| 15.0k| mkinitcap2(part2, part2u); 593| 15.0k| scw = part1 + part2; 594| 15.0k| sunicw = std::move(part1u); 595| 15.0k| sunicw.insert(sunicw.end(), part2u.begin(), part2u.end()); 596| 15.0k| rv = checkword(scw, info, root, suggest_start); 597| 15.0k| if (rv) ------------------ | Branch (597:17): [True: 0, False: 15.0k] ------------------ 598| 0| break; 599| 54.7k| } else { 600| 54.7k| mkinitcap2(part2, sunicw); 601| 54.7k| scw = part1 + part2; 602| 54.7k| rv = checkword(scw, info, root, suggest_start); 603| 54.7k| if (rv) ------------------ | Branch (603:17): [True: 1, False: 54.7k] ------------------ 604| 1| break; 605| 54.7k| } 606| 69.8k| mkinitcap2(scw, sunicw); 607| 69.8k| rv = checkword(scw, info, root, suggest_start); 608| 69.8k| if (rv) ------------------ | Branch (608:15): [True: 0, False: 69.8k] ------------------ 609| 0| break; 610| 69.8k| } 611| 76.3k| } 612| 196k| if (pAMgr && pAMgr->get_checksharps() && scw.find("SS") != std::string::npos) { ------------------ | Branch (612:11): [True: 196k, False: 0] | Branch (612:20): [True: 0, False: 196k] | Branch (612:48): [True: 0, False: 0] ------------------ 613| | 614| 0| mkallsmall2(scw, sunicw); 615| 0| std::string u8buffer(scw); 616| 0| rv = spellsharps(u8buffer, 0, 0, 0, info, root, suggest_start); 617| 0| if (!rv) { ------------------ | Branch (617:13): [True: 0, False: 0] ------------------ 618| 0| mkinitcap2(scw, sunicw); 619| 0| rv = spellsharps(scw, 0, 0, 0, info, root, suggest_start); 620| 0| } 621| 0| if ((abbv) && !(rv)) { ------------------ | Branch (621:13): [True: 0, False: 0] | Branch (621:23): [True: 0, False: 0] ------------------ 622| 0| u8buffer.push_back('.'); 623| 0| rv = spellsharps(u8buffer, 0, 0, 0, info, root, suggest_start); 624| 0| if (!rv) { ------------------ | Branch (624:15): [True: 0, False: 0] ------------------ 625| 0| u8buffer = std::string(scw); 626| 0| u8buffer.push_back('.'); 627| 0| rv = spellsharps(u8buffer, 0, 0, 0, info, root, suggest_start); 628| 0| } 629| 0| } 630| 0| if (rv) ------------------ | Branch (630:13): [True: 0, False: 0] ------------------ 631| 0| break; 632| 0| } 633| 196k| } 634| | /* FALLTHROUGH */ 635| 611k| case INITCAP: { ------------------ | | 93| 611k|#define INITCAP 1 ------------------ | Branch (635:5): [True: 414k, False: 8.77M] ------------------ 636| | // handle special capitalization of dotted I 637| 611k| bool Idot = (utf8 && (unsigned char) scw[0] == 0xc4 && (unsigned char) scw[1] == 0xb0); ------------------ | Branch (637:20): [True: 164k, False: 446k] | Branch (637:28): [True: 0, False: 164k] | Branch (637:62): [True: 0, False: 0] ------------------ 638| 611k| *info |= SPELL_ORIGCAP; ------------------ | | 83| 611k|#define SPELL_ORIGCAP (1 << 5) ------------------ 639| 611k| if (captype == ALLCAP) { ------------------ | | 94| 611k|#define ALLCAP 2 ------------------ | Branch (639:11): [True: 196k, False: 414k] ------------------ 640| 196k| mkallsmall2(scw, sunicw); 641| 196k| mkinitcap2(scw, sunicw); 642| 196k| if (Idot) ------------------ | Branch (642:15): [True: 0, False: 196k] ------------------ 643| 0| scw.replace(0, 1, "\xc4\xb0"); 644| 196k| } 645| 611k| if (captype == INITCAP) ------------------ | | 93| 611k|#define INITCAP 1 ------------------ | Branch (645:11): [True: 414k, False: 196k] ------------------ 646| 414k| *info |= SPELL_INITCAP; ------------------ | | 82| 414k|#define SPELL_INITCAP (1 << 4) ------------------ 647| 611k| rv = checkword(scw, info, root, suggest_start); 648| 611k| if (captype == INITCAP) ------------------ | | 93| 611k|#define INITCAP 1 ------------------ | Branch (648:11): [True: 414k, False: 196k] ------------------ 649| 414k| *info &= ~SPELL_INITCAP; ------------------ | | 82| 414k|#define SPELL_INITCAP (1 << 4) ------------------ 650| | // forbid bad capitalization 651| | // (for example, ijs -> Ijs instead of IJs in Dutch) 652| | // use explicit forms in dic: Ijs/F (F = FORBIDDENWORD flag) 653| 611k| if (*info & SPELL_FORBIDDEN) { ------------------ | | 79| 611k|#define SPELL_FORBIDDEN (1 << 1) ------------------ | Branch (653:11): [True: 0, False: 611k] ------------------ 654| 0| rv = nullptr; 655| 0| break; 656| 0| } 657| 611k| if (rv && is_keepcase(rv) && (captype == ALLCAP)) ------------------ | | 94| 13|#define ALLCAP 2 ------------------ | Branch (657:11): [True: 13.0k, False: 598k] | Branch (657:17): [True: 13, False: 13.0k] | Branch (657:36): [True: 1, False: 12] ------------------ 658| 1| rv = nullptr; 659| 611k| if (rv || (Idot && langnum != LANG_az && langnum != LANG_tr && langnum != LANG_crh)) ------------------ | Branch (659:11): [True: 13.0k, False: 598k] | Branch (659:18): [True: 0, False: 598k] | Branch (659:26): [True: 0, False: 0] | Branch (659:48): [True: 0, False: 0] | Branch (659:70): [True: 0, False: 0] ------------------ 660| 13.0k| break; 661| | 662| 598k| mkallsmall2(scw, sunicw); 663| 598k| std::string u8buffer(scw); 664| 598k| mkinitcap2(scw, sunicw); 665| | 666| 598k| rv = checkword(u8buffer, info, root, suggest_start); 667| 598k| if (abbv && !rv) { ------------------ | Branch (667:11): [True: 92.6k, False: 505k] | Branch (667:19): [True: 92.1k, False: 499] ------------------ 668| 92.1k| u8buffer.push_back('.'); 669| 92.1k| rv = checkword(u8buffer, info, root, suggest_start); 670| 92.1k| if (!rv) { ------------------ | Branch (670:13): [True: 92.1k, False: 1] ------------------ 671| 92.1k| u8buffer = scw; 672| 92.1k| u8buffer.push_back('.'); 673| 92.1k| if (captype == INITCAP) ------------------ | | 93| 92.1k|#define INITCAP 1 ------------------ | Branch (673:15): [True: 76.5k, False: 15.5k] ------------------ 674| 76.5k| *info |= SPELL_INITCAP; ------------------ | | 82| 76.5k|#define SPELL_INITCAP (1 << 4) ------------------ 675| 92.1k| rv = checkword(u8buffer, info, root, suggest_start); 676| 92.1k| if (captype == INITCAP) ------------------ | | 93| 92.1k|#define INITCAP 1 ------------------ | Branch (676:15): [True: 76.5k, False: 15.5k] ------------------ 677| 76.5k| *info &= ~SPELL_INITCAP; ------------------ | | 82| 76.5k|#define SPELL_INITCAP (1 << 4) ------------------ 678| 92.1k| if (rv && is_keepcase(rv) && (captype == ALLCAP)) ------------------ | | 94| 14|#define ALLCAP 2 ------------------ | Branch (678:15): [True: 14, False: 92.1k] | Branch (678:21): [True: 14, False: 0] | Branch (678:40): [True: 5, False: 9] ------------------ 679| 5| rv = nullptr; 680| 92.1k| break; 681| 92.1k| } 682| 92.1k| } 683| 505k| if (rv && is_keepcase(rv) && ------------------ | Branch (683:11): [True: 3.69k, False: 502k] | Branch (683:17): [True: 5, False: 3.68k] ------------------ 684| 5| ((captype == ALLCAP) || ------------------ | | 94| 5|#define ALLCAP 2 ------------------ | Branch (684:12): [True: 2, False: 3] ------------------ 685| | // if CHECKSHARPS: KEEPCASE words with \xDF are allowed 686| | // in INITCAP form, too. 687| 3| !(pAMgr->get_checksharps() && ------------------ | Branch (687:14): [True: 0, False: 3] ------------------ 688| 0| ((utf8 && u8buffer.find("\xC3\x9F") != std::string::npos) || ------------------ | Branch (688:16): [True: 0, False: 0] | Branch (688:24): [True: 0, False: 0] ------------------ 689| 0| (!utf8 && u8buffer.find('\xDF') != std::string::npos))))) ------------------ | Branch (689:16): [True: 0, False: 0] | Branch (689:25): [True: 0, False: 0] ------------------ 690| 5| rv = nullptr; 691| 505k| break; 692| 598k| } 693| 9.19M| } 694| | 695| 9.19M| if (rv) { ------------------ | Branch (695:7): [True: 27.6k, False: 9.16M] ------------------ 696| 27.6k| if (pAMgr && pAMgr->get_warn() && rv->astr && ------------------ | Branch (696:9): [True: 27.6k, False: 0] | Branch (696:9): [True: 4, False: 27.6k] | Branch (696:18): [True: 2.19k, False: 25.4k] | Branch (696:39): [True: 162, False: 2.02k] ------------------ 697| 162| TESTAFF(rv->astr, pAMgr->get_warn(), rv->alen)) { ------------------ | | 99| 162|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 4, False: 158] | | ------------------ ------------------ 698| 4| *info |= SPELL_WARN; ------------------ | | 84| 4|#define SPELL_WARN (1 << 6) ------------------ 699| 4| if (pAMgr->get_forbidwarn()) ------------------ | Branch (699:11): [True: 0, False: 4] ------------------ 700| 0| return false; 701| 4| return true; 702| 4| } 703| 27.6k| return true; 704| 27.6k| } 705| | 706| | // recursive breaking at break points 707| 9.16M| if (!wordbreak.empty() && !(*info & SPELL_FORBIDDEN)) { ------------------ | | 79| 9.16M|#define SPELL_FORBIDDEN (1 << 1) ------------------ | Branch (707:7): [True: 9.16M, False: 0] | Branch (707:29): [True: 9.16M, False: 0] ------------------ 708| | 709| 9.16M| int nbr = 0; 710| 9.16M| wl = scw.size(); 711| | 712| | // calculate break points for recursion limit 713| 27.4M| for (auto& j : wordbreak) { ------------------ | Branch (713:18): [True: 27.4M, False: 9.16M] ------------------ 714| 27.4M| size_t pos = 0; 715| 37.5M| while ((pos = scw.find(j, pos)) != std::string::npos) { ------------------ | Branch (715:14): [True: 10.1M, False: 27.4M] ------------------ 716| 10.1M| ++nbr; 717| 10.1M| pos += j.size(); 718| 10.1M| } 719| 27.4M| } 720| 9.16M| if (nbr >= MAXBREAKDEPTH) ------------------ | | 86| 9.16M|#define MAXBREAKDEPTH 10 ------------------ | Branch (720:9): [True: 2.63k, False: 9.16M] ------------------ 721| 2.63k| return false; 722| | 723| | // check boundary patterns (^begin and end$) 724| 27.3M| for (auto& j : wordbreak) { ------------------ | Branch (724:18): [True: 27.3M, False: 8.24M] ------------------ 725| 27.3M| size_t plen = j.size(); 726| 27.3M| if (plen == 1 || plen > wl) ------------------ | Branch (726:11): [True: 9.16M, False: 18.1M] | Branch (726:24): [True: 2.78M, False: 15.3M] ------------------ 727| 11.9M| continue; 728| | 729| 15.3M| if (j[0] == '^' && scw.compare(0, plen - 1, j, 1, plen - 1) == 0 && ------------------ | Branch (729:11): [True: 7.72M, False: 7.63M] | Branch (729:11): [True: 111k, False: 15.2M] | Branch (729:26): [True: 880k, False: 6.84M] ------------------ 730| 880k| spell(scw.substr(plen - 1), candidate_stack, nullptr, nullptr, suggest_start)) { ------------------ | Branch (730:11): [True: 111k, False: 768k] ------------------ 731| 111k| *info |= SPELL_COMPOUND; ------------------ | | 78| 111k|#define SPELL_COMPOUND (1 << 0) // the result is a compound word ------------------ 732| 111k| return true; 733| 111k| } 734| | 735| 15.2M| if (j[plen - 1] == '$' && ------------------ | Branch (735:11): [True: 7.61M, False: 7.63M] ------------------ 736| 7.61M| scw.compare(wl - plen + 1, plen - 1, j, 0, plen - 1) == 0) { ------------------ | Branch (736:11): [True: 3.15M, False: 4.46M] ------------------ 737| 3.15M| std::string suffix(scw.substr(wl - plen + 1)); 738| 3.15M| scw.resize(wl - plen + 1); 739| 3.15M| if (spell(scw, candidate_stack, nullptr, nullptr, suggest_start)) { ------------------ | Branch (739:13): [True: 808k, False: 2.34M] ------------------ 740| 808k| *info |= SPELL_COMPOUND; ------------------ | | 78| 808k|#define SPELL_COMPOUND (1 << 0) // the result is a compound word ------------------ 741| 808k| return true; 742| 808k| } 743| 2.34M| scw.append(suffix); 744| 2.34M| } 745| 15.2M| } 746| | 747| | // other patterns 748| 24.6M| for (auto& j : wordbreak) { ------------------ | Branch (748:18): [True: 24.6M, False: 8.20M] ------------------ 749| 24.6M| size_t plen = j.size(); 750| 24.6M| size_t found = scw.find(j); 751| 24.6M| if ((found > 0) && (found < wl - plen)) { ------------------ | Branch (751:11): [True: 23.1M, False: 1.44M] | Branch (751:26): [True: 2.52M, False: 20.6M] ------------------ 752| 2.52M| size_t found2 = scw.find(j, found + 1); 753| | // try to break at the second occurance 754| | // to recognize dictionary words with wordbreak 755| 2.52M| if (found2 > 0 && (found2 < wl - plen)) ------------------ | Branch (755:13): [True: 2.52M, False: 0] | Branch (755:27): [True: 1.01M, False: 1.51M] ------------------ 756| 1.01M| found = found2; 757| 2.52M| std::string substring(scw.substr(found + plen)); 758| 2.52M| if (!spell(substring, candidate_stack, nullptr, nullptr, suggest_start)) ------------------ | Branch (758:13): [True: 1.87M, False: 653k] ------------------ 759| 1.87M| continue; 760| 653k| std::string suffix(scw.substr(found)); 761| 653k| scw.resize(found); 762| | // examine 2 sides of the break point 763| 653k| if (spell(scw, candidate_stack, nullptr, nullptr, suggest_start)) { ------------------ | Branch (763:13): [True: 39.5k, False: 613k] ------------------ 764| 39.5k| *info |= SPELL_COMPOUND; ------------------ | | 78| 39.5k|#define SPELL_COMPOUND (1 << 0) // the result is a compound word ------------------ 765| 39.5k| return true; 766| 39.5k| } 767| 613k| scw.append(suffix); 768| | 769| | // LANG_hu: spec. dash rule 770| 613k| if (langnum == LANG_hu && j == "-") { ------------------ | Branch (770:13): [True: 285, False: 613k] | Branch (770:35): [True: 285, False: 0] ------------------ 771| 285| suffix = scw.substr(found + 1); 772| 285| scw.resize(found + 1); 773| 285| if (spell(scw, candidate_stack, nullptr, nullptr, suggest_start)) { ------------------ | Branch (773:15): [True: 0, False: 285] ------------------ 774| 0| *info |= SPELL_COMPOUND; ------------------ | | 78| 0|#define SPELL_COMPOUND (1 << 0) // the result is a compound word ------------------ 775| 0| return true; // check the first part with dash 776| 0| } 777| 285| scw.append(suffix); 778| 285| } 779| | // end of LANG specific region 780| 613k| } 781| 24.6M| } 782| | 783| | // other patterns (break at first break point) 784| 24.5M| for (auto& j : wordbreak) { ------------------ | Branch (784:18): [True: 24.5M, False: 8.20M] ------------------ 785| 24.5M| size_t plen = j.size(), found = scw.find(j); 786| 24.5M| if ((found > 0) && (found < wl - plen)) { ------------------ | Branch (786:11): [True: 23.0M, False: 1.44M] | Branch (786:26): [True: 2.47M, False: 20.6M] ------------------ 787| 2.47M| if (!spell(scw.substr(found + plen), candidate_stack, nullptr, nullptr, suggest_start)) ------------------ | Branch (787:13): [True: 1.99M, False: 478k] ------------------ 788| 1.99M| continue; 789| 478k| std::string suffix(scw.substr(found)); 790| 478k| scw.resize(found); 791| | // examine 2 sides of the break point 792| 478k| if (spell(scw, candidate_stack, nullptr, nullptr, suggest_start)) { ------------------ | Branch (792:13): [True: 3.26k, False: 475k] ------------------ 793| 3.26k| *info |= SPELL_COMPOUND; ------------------ | | 78| 3.26k|#define SPELL_COMPOUND (1 << 0) // the result is a compound word ------------------ 794| 3.26k| return true; 795| 3.26k| } 796| 475k| scw.append(suffix); 797| | 798| | // LANG_hu: spec. dash rule 799| 475k| if (langnum == LANG_hu && j == "-") { ------------------ | Branch (799:13): [True: 130, False: 475k] | Branch (799:35): [True: 130, False: 0] ------------------ 800| 130| suffix = scw.substr(found + 1); 801| 130| scw.resize(found + 1); 802| 130| if (spell(scw, candidate_stack, nullptr, nullptr, suggest_start)) { ------------------ | Branch (802:15): [True: 0, False: 130] ------------------ 803| 0| *info |= SPELL_COMPOUND; ------------------ | | 78| 0|#define SPELL_COMPOUND (1 << 0) // the result is a compound word ------------------ 804| 0| return true; // check the first part with dash 805| 0| } 806| 130| scw.append(suffix); 807| 130| } 808| | // end of LANG specific region 809| 475k| } 810| 24.5M| } 811| 8.20M| } 812| | 813| 8.20M| return false; 814| 9.16M|} _ZN12HunspellImpl9checkwordERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEPiPS6_NS0_6chrono10time_pointINSB_12steady_clockENSB_8durationIxNS0_5ratioILl1ELl1000000000EEEEEEE: 817| 11.5M| std::chrono::steady_clock::time_point suggest_start) { 818| | // check overall suggest time limit 819| 11.5M| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 11.5M|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (819:7): [True: 0, False: 11.5M] ------------------ 820| 0| return nullptr; 821| | 822| 11.5M| std::string word; 823| | 824| | // remove IGNORE characters from the string 825| 11.5M| clean_ignore(word, w); 826| | 827| 11.5M| if (word.empty()) ------------------ | Branch (827:7): [True: 347, False: 11.5M] ------------------ 828| 347| return nullptr; 829| | 830| | // word reversing wrapper for complex prefixes 831| 11.5M| if (complexprefixes) { ------------------ | Branch (831:7): [True: 903k, False: 10.6M] ------------------ 832| 903k| if (utf8) ------------------ | Branch (832:9): [True: 0, False: 903k] ------------------ 833| 0| reverseword_utf(word); 834| 903k| else 835| 903k| reverseword(word); 836| 903k| } 837| | 838| 11.5M| int len = word.size(); 839| | 840| | // look word in hash table 841| 11.5M| struct hentry* he = nullptr; 842| 23.0M| for (size_t i = 0; (i < m_HMgrs.size()) && !he; ++i) { ------------------ | Branch (842:22): [True: 11.5M, False: 11.5M] | Branch (842:46): [True: 11.5M, False: 0] ------------------ 843| 11.5M| he = m_HMgrs[i]->lookup(word.c_str(), word.size()); 844| | 845| | // check forbidden and onlyincompound words 846| 11.5M| if ((he) && (he->astr) && (pAMgr) && ------------------ | Branch (846:9): [True: 27.6k, False: 11.4M] | Branch (846:9): [True: 0, False: 11.5M] | Branch (846:17): [True: 2.04k, False: 25.6k] | Branch (846:31): [True: 2.04k, False: 0] ------------------ 847| 2.04k| TESTAFF(he->astr, pAMgr->get_forbiddenword(), he->alen)) { ------------------ | | 99| 2.04k|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 2.04k] | | ------------------ ------------------ 848| 0| if (info) ------------------ | Branch (848:11): [True: 0, False: 0] ------------------ 849| 0| *info |= SPELL_FORBIDDEN; ------------------ | | 79| 0|#define SPELL_FORBIDDEN (1 << 1) ------------------ 850| | // LANG_hu section: set dash information for suggestions 851| 0| if (langnum == LANG_hu) { ------------------ | Branch (851:11): [True: 0, False: 0] ------------------ 852| 0| if (pAMgr->get_compoundflag() && ------------------ | Branch (852:13): [True: 0, False: 0] | Branch (852:13): [True: 0, False: 0] ------------------ 853| 0| TESTAFF(he->astr, pAMgr->get_compoundflag(), he->alen)) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 854| 0| if (info) ------------------ | Branch (854:15): [True: 0, False: 0] ------------------ 855| 0| *info |= SPELL_COMPOUND; ------------------ | | 78| 0|#define SPELL_COMPOUND (1 << 0) // the result is a compound word ------------------ 856| 0| } 857| 0| } 858| 0| return nullptr; 859| 0| } 860| | 861| | // he = next not needaffix, onlyincompound homonym or onlyupcase word 862| 11.5M| while (he && (he->astr) && pAMgr && ------------------ | Branch (862:12): [True: 27.6k, False: 11.4M] | Branch (862:12): [True: 28, False: 11.5M] | Branch (862:18): [True: 2.04k, False: 25.6k] | Branch (862:32): [True: 2.04k, False: 0] ------------------ 863| 2.04k| ((pAMgr->get_needaffix() && ------------------ | Branch (863:14): [True: 172, False: 1.87k] ------------------ 864| 172| TESTAFF(he->astr, pAMgr->get_needaffix(), he->alen)) || ------------------ | | 99| 172|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 172] | | ------------------ ------------------ 865| 2.04k| (pAMgr->get_onlyincompound() && ------------------ | Branch (865:14): [True: 0, False: 2.04k] ------------------ 866| 0| TESTAFF(he->astr, pAMgr->get_onlyincompound(), he->alen)) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 867| 2.04k| (info && (*info & SPELL_INITCAP) && ------------------ | | 82| 2.04k|#define SPELL_INITCAP (1 << 4) ------------------ | Branch (867:14): [True: 2.04k, False: 0] | Branch (867:22): [True: 49, False: 1.99k] ------------------ 868| 49| TESTAFF(he->astr, ONLYUPCASEFLAG, he->alen)))) ------------------ | | 99| 49|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 28, False: 21] | | ------------------ ------------------ 869| 28| he = he->next_homonym; 870| 11.5M| } 871| | 872| | // check with affixes 873| 11.5M| if (!he && pAMgr) { ------------------ | Branch (873:7): [True: 11.4M, False: 27.6k] | Branch (873:14): [True: 11.4M, False: 0] ------------------ 874| | // try stripping off affixes 875| 11.4M| AffixScratch scratch; 876| 11.4M| he = pAMgr->affix_check(word, 0, len, scratch, 0); 877| | 878| | // check compound restriction and onlyupcase 879| 11.4M| if (he && he->astr && ------------------ | Branch (879:9): [True: 0, False: 11.4M] | Branch (879:9): [True: 0, False: 11.4M] | Branch (879:15): [True: 0, False: 0] ------------------ 880| 0| ((pAMgr->get_onlyincompound() && ------------------ | Branch (880:11): [True: 0, False: 0] ------------------ 881| 0| TESTAFF(he->astr, pAMgr->get_onlyincompound(), he->alen)) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 882| 0| (info && (*info & SPELL_INITCAP) && ------------------ | | 82| 0|#define SPELL_INITCAP (1 << 4) ------------------ | Branch (882:11): [True: 0, False: 0] | Branch (882:19): [True: 0, False: 0] ------------------ 883| 0| TESTAFF(he->astr, ONLYUPCASEFLAG, he->alen)))) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 884| 0| he = nullptr; 885| 0| } 886| | 887| 11.4M| if (he) { ------------------ | Branch (887:9): [True: 0, False: 11.4M] ------------------ 888| 0| if ((he->astr) && (pAMgr) && ------------------ | Branch (888:11): [True: 0, False: 0] | Branch (888:11): [True: 0, False: 0] | Branch (888:25): [True: 0, False: 0] ------------------ 889| 0| TESTAFF(he->astr, pAMgr->get_forbiddenword(), he->alen)) { ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 890| 0| if (info) ------------------ | Branch (890:13): [True: 0, False: 0] ------------------ 891| 0| *info |= SPELL_FORBIDDEN; ------------------ | | 79| 0|#define SPELL_FORBIDDEN (1 << 1) ------------------ 892| 0| return nullptr; 893| 0| } 894| 0| if (root) { ------------------ | Branch (894:11): [True: 0, False: 0] ------------------ 895| 0| root->assign(he->word); 896| 0| if (complexprefixes) { ------------------ | Branch (896:13): [True: 0, False: 0] ------------------ 897| 0| if (utf8) ------------------ | Branch (897:15): [True: 0, False: 0] ------------------ 898| 0| reverseword_utf(*root); 899| 0| else 900| 0| reverseword(*root); 901| 0| } 902| 0| } 903| | // try check compound word 904| 11.4M| } else if (pAMgr->get_compound()) { ------------------ | Branch (904:16): [True: 2.71M, False: 8.76M] ------------------ 905| 2.71M| struct hentry* rwords[100] = {}; // buffer for COMPOUND pattern checking 906| | 907| | // first allow only 2 words in the compound 908| 2.71M| int setinfo = SPELL_COMPOUND_2; ------------------ | | 85| 2.71M|#define SPELL_COMPOUND_2 (1 << 7) // permit only 2 dictionary words in the compound ------------------ 909| 2.71M| if (info) ------------------ | Branch (909:11): [True: 2.71M, False: 0] ------------------ 910| 2.71M| setinfo |= *info; 911| 2.71M| he = pAMgr->compound_check(word, 0, 0, 100, 0, nullptr, (hentry**)&rwords, 0, 0, &setinfo, scratch); 912| 2.71M| if (info) ------------------ | Branch (912:11): [True: 2.71M, False: 0] ------------------ 913| 2.71M| *info = setinfo & ~SPELL_COMPOUND_2; ------------------ | | 85| 2.71M|#define SPELL_COMPOUND_2 (1 << 7) // permit only 2 dictionary words in the compound ------------------ 914| | // if not 2-word compoud word, try with 3 or more words 915| | // (only if original info didn't forbid it) 916| 2.71M| if (!he && info && !(*info & SPELL_COMPOUND_2)) { ------------------ | | 85| 2.71M|#define SPELL_COMPOUND_2 (1 << 7) // permit only 2 dictionary words in the compound ------------------ | Branch (916:11): [True: 2.71M, False: 0] | Branch (916:18): [True: 2.71M, False: 0] | Branch (916:26): [True: 2.71M, False: 0] ------------------ 917| 2.71M| *info &= ~SPELL_COMPOUND_2; ------------------ | | 85| 2.71M|#define SPELL_COMPOUND_2 (1 << 7) // permit only 2 dictionary words in the compound ------------------ 918| 2.71M| he = pAMgr->compound_check(word, 0, 0, 100, 0, nullptr, (hentry**)&rwords, 0, 0, info, scratch); 919| | // accept the compound with 3 or more words only if it is 920| | // - not a dictionary word with a typo and 921| | // - not two words written separately, 922| | // - or if it's an arbitrary number accepted by compound rules (e.g. 999%) 923| 2.71M| if (he && !isdigit(word[0])) ------------------ | Branch (923:13): [True: 0, False: 2.71M] | Branch (923:19): [True: 0, False: 0] ------------------ 924| 0| { 925| 0| std::vector slst; 926| 0| if (pSMgr->suggest(slst, word, nullptr, /*test_simplesug=*/true)) ------------------ | Branch (926:15): [True: 0, False: 0] ------------------ 927| 0| he = nullptr; 928| 0| } 929| 2.71M| } 930| | 931| | // LANG_hu section: `moving rule' with last dash 932| 2.71M| if ((!he) && (langnum == LANG_hu) && (word[len - 1] == '-')) { ------------------ | Branch (932:11): [True: 2.71M, False: 0] | Branch (932:20): [True: 28.1k, False: 2.68M] | Branch (932:44): [True: 10.7k, False: 17.4k] ------------------ 933| 10.7k| std::string dup(word, 0, len - 1); 934| 10.7k| he = pAMgr->compound_check(dup, -5, 0, 100, 0, nullptr, (hentry**)&rwords, 1, 0, info, scratch); 935| 10.7k| } 936| | // end of LANG specific region 937| 2.71M| if (he) { ------------------ | Branch (937:11): [True: 0, False: 2.71M] ------------------ 938| 0| if (root) { ------------------ | Branch (938:13): [True: 0, False: 0] ------------------ 939| 0| root->assign(he->word); 940| 0| if (complexprefixes) { ------------------ | Branch (940:15): [True: 0, False: 0] ------------------ 941| 0| if (utf8) ------------------ | Branch (941:17): [True: 0, False: 0] ------------------ 942| 0| reverseword_utf(*root); 943| 0| else 944| 0| reverseword(*root); 945| 0| } 946| 0| } 947| 0| if (info) ------------------ | Branch (947:13): [True: 0, False: 0] ------------------ 948| 0| *info |= SPELL_COMPOUND; ------------------ | | 78| 0|#define SPELL_COMPOUND (1 << 0) // the result is a compound word ------------------ 949| 0| } 950| 2.71M| } 951| 11.4M| } 952| | 953| 11.5M| return he; 954| 11.5M|} _ZN12HunspellImpl7suggestERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEERNS0_6vectorIS6_NS4_IS6_EEEENS0_6chrono10time_pointINSD_12steady_clockENSD_8durationIxNS0_5ratioILl1ELl1000000000EEEEEEE: 962| 90.8k|std::vector HunspellImpl::suggest(const std::string& word, std::vector& suggest_candidate_stack, std::chrono::steady_clock::time_point suggest_start) { 963| | 964| 90.8k| if (suggest_candidate_stack.size() > MAX_CANDIDATE_STACK_DEPTH || // apply a fairly arbitrary depth limit ------------------ | | 957| 181k|#define MAX_CANDIDATE_STACK_DEPTH 512 ------------------ | Branch (964:7): [True: 0, False: 90.8k] | Branch (964:7): [True: 0, False: 90.8k] ------------------ 965| | // something very broken if suggest ends up calling itself with the same word 966| 90.8k| std::find(suggest_candidate_stack.begin(), suggest_candidate_stack.end(), word) != suggest_candidate_stack.end()) { ------------------ | Branch (966:7): [True: 0, False: 90.8k] ------------------ 967| 0| return { }; 968| 0| } 969| | 970| 90.8k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 90.8k|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (970:7): [True: 0, False: 90.8k] ------------------ 971| 0| return { }; 972| | 973| 90.8k| bool capwords; 974| 90.8k| size_t abbv; 975| 90.8k| int captype; 976| 90.8k| std::vector spell_candidate_stack; 977| 90.8k| suggest_candidate_stack.push_back(word); 978| 90.8k| std::vector slst = suggest_internal(word, spell_candidate_stack, suggest_candidate_stack, 979| 90.8k| capwords, abbv, captype, suggest_start); 980| 90.8k| suggest_candidate_stack.pop_back(); 981| | // word reversing wrapper for complex prefixes 982| 90.8k| if (complexprefixes) { ------------------ | Branch (982:7): [True: 4.72k, False: 86.1k] ------------------ 983| 4.72k| for (auto& j : slst) { ------------------ | Branch (983:18): [True: 2.20k, False: 4.72k] ------------------ 984| 2.20k| if (utf8) ------------------ | Branch (984:11): [True: 0, False: 2.20k] ------------------ 985| 0| reverseword_utf(j); 986| 2.20k| else 987| 2.20k| reverseword(j); 988| 2.20k| } 989| 4.72k| } 990| | 991| | // capitalize 992| 90.8k| if (capwords) { ------------------ | Branch (992:7): [True: 25.5k, False: 65.2k] ------------------ 993| 25.5k| for (auto& j : slst) { ------------------ | Branch (993:18): [True: 14.2k, False: 25.5k] ------------------ 994| 14.2k| std::string capitalized(j); 995| 14.2k| mkinitcap(capitalized); 996| 14.2k| if (capitalized == word) ------------------ | Branch (996:11): [True: 35, False: 14.2k] ------------------ 997| 35| continue; // capitalizing would just reproduce the misspelled word 998| 14.2k| j = std::move(capitalized); 999| 14.2k| } 1000| 25.5k| } 1001| | 1002| | // expand suggestions with dot(s) 1003| 90.8k| if (abbv && pAMgr && pAMgr->get_sugswithdots() && word.size() >= abbv) { ------------------ | Branch (1003:7): [True: 3.70k, False: 87.1k] | Branch (1003:15): [True: 3.70k, False: 0] | Branch (1003:24): [True: 0, False: 3.70k] | Branch (1003:53): [True: 0, False: 0] ------------------ 1004| 0| for (auto& j : slst) { ------------------ | Branch (1004:18): [True: 0, False: 0] ------------------ 1005| 0| j.append(word.substr(word.size() - abbv)); 1006| 0| } 1007| 0| } 1008| | 1009| | // remove bad capitalized and forbidden forms 1010| 90.8k| if (pAMgr && (pAMgr->get_keepcase() || pAMgr->get_forbiddenword())) { ------------------ | Branch (1010:7): [True: 90.8k, False: 0] | Branch (1010:17): [True: 4.75k, False: 86.0k] | Branch (1010:42): [True: 86.0k, False: 0] ------------------ 1011| 90.8k| switch (captype) { ------------------ | Branch (1011:13): [True: 23.1k, False: 67.6k] ------------------ 1012| 12.9k| case INITCAP: ------------------ | | 93| 12.9k|#define INITCAP 1 ------------------ | Branch (1012:7): [True: 12.9k, False: 77.8k] ------------------ 1013| 23.1k| case ALLCAP: { ------------------ | | 94| 23.1k|#define ALLCAP 2 ------------------ | Branch (1013:7): [True: 10.1k, False: 80.6k] ------------------ 1014| 23.1k| size_t l = 0; 1015| 33.1k| for (size_t j = 0; j < slst.size(); ++j) { ------------------ | Branch (1015:28): [True: 9.97k, False: 23.1k] ------------------ 1016| 9.97k| if (slst[j].find(' ') == std::string::npos && !spell(slst[j], spell_candidate_stack, nullptr, nullptr, suggest_start)) { ------------------ | Branch (1016:15): [True: 9.48k, False: 491] | Branch (1016:57): [True: 4.26k, False: 5.22k] ------------------ 1017| 4.26k| std::string s; 1018| 4.26k| std::vector w; 1019| 4.26k| if (utf8) { ------------------ | Branch (1019:17): [True: 3.46k, False: 798] ------------------ 1020| 3.46k| u8_u16(w, slst[j]); 1021| 3.46k| } else { 1022| 798| s = slst[j]; 1023| 798| } 1024| 4.26k| mkallsmall2(s, w); 1025| 4.26k| if (spell(s, spell_candidate_stack, nullptr, nullptr, suggest_start)) { ------------------ | Branch (1025:17): [True: 5, False: 4.26k] ------------------ 1026| 5| slst[l] = std::move(s); 1027| 5| ++l; 1028| 4.26k| } else { 1029| 4.26k| mkinitcap2(s, w); 1030| 4.26k| if (spell(s, spell_candidate_stack, nullptr, nullptr, suggest_start)) { ------------------ | Branch (1030:19): [True: 6, False: 4.25k] ------------------ 1031| 6| slst[l] = std::move(s); 1032| 6| ++l; 1033| 6| } 1034| 4.26k| } 1035| 5.71k| } else { 1036| 5.71k| slst[l] = slst[j]; 1037| 5.71k| ++l; 1038| 5.71k| } 1039| 9.97k| } 1040| 23.1k| slst.resize(l); 1041| 23.1k| } 1042| 90.8k| } 1043| 90.8k| } 1044| | 1045| | // remove duplications 1046| 90.8k| size_t l = 0; 1047| 117k| for (size_t j = 0; j < slst.size(); ++j) { ------------------ | Branch (1047:22): [True: 26.4k, False: 90.8k] ------------------ 1048| 26.4k| slst[l] = slst[j]; 1049| 41.6k| for (size_t k = 0; k < l; ++k) { ------------------ | Branch (1049:24): [True: 16.2k, False: 25.3k] ------------------ 1050| 16.2k| if (slst[k] == slst[j]) { ------------------ | Branch (1050:11): [True: 1.04k, False: 15.2k] ------------------ 1051| 1.04k| --l; 1052| 1.04k| break; 1053| 1.04k| } 1054| 16.2k| } 1055| 26.4k| ++l; 1056| 26.4k| } 1057| 90.8k| slst.resize(l); 1058| | 1059| | // output conversion 1060| 90.8k| RepList* rl = (pAMgr) ? pAMgr->get_oconvtable() : nullptr; ------------------ | Branch (1060:17): [True: 90.8k, False: 0] ------------------ 1061| 90.8k| if (rl) { ------------------ | Branch (1061:7): [True: 4.85k, False: 85.9k] ------------------ 1062| 4.85k| size_t l = 0; 1063| 5.79k| for (size_t i = 0; i < slst.size(); ++i) { ------------------ | Branch (1063:24): [True: 940, False: 4.85k] ------------------ 1064| 940| std::string wspace; 1065| 940| if (rl->conv(slst[i], wspace)) { ------------------ | Branch (1065:11): [True: 302, False: 638] ------------------ 1066| 302| slst[i] = std::move(wspace); 1067| 302| } 1068| | // gh#1002: OCONV can map a generated form back to the input word 1069| | // (e.g. "románórum" -> "romanórum" when the user typed "romanórum"), 1070| | // leaving the misspelled word as its own suggestion. 1071| 940| if (slst[i] == word) ------------------ | Branch (1071:11): [True: 0, False: 940] ------------------ 1072| 0| continue; 1073| 940| if (l != i) ------------------ | Branch (1073:11): [True: 0, False: 940] ------------------ 1074| 0| slst[l] = std::move(slst[i]); 1075| 940| ++l; 1076| 940| } 1077| 4.85k| slst.resize(l); 1078| 4.85k| } 1079| 90.8k| return slst; 1080| 90.8k|} _ZN12HunspellImpl7suggestERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEE: 1082| 59.9k|std::vector HunspellImpl::suggest(const std::string& word) { 1083| 59.9k| std::vector suggest_candidate_stack; 1084| 59.9k| auto suggest_start = std::chrono::steady_clock::now(); 1085| 59.9k| return suggest(word, suggest_candidate_stack, suggest_start); 1086| 59.9k|} _ZN12HunspellImpl16suggest_internalERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEERNS0_6vectorIS6_NS4_IS6_EEEESC_RbRmRiNS0_6chrono10time_pointINSG_12steady_clockENSG_8durationIxNS0_5ratioILl1ELl1000000000EEEEEEE: 1092| 90.8k| std::chrono::steady_clock::time_point suggest_start) { 1093| 90.8k| captype = NOCAP; ------------------ | | 92| 90.8k|#define NOCAP 0 ------------------ 1094| 90.8k| abbv = 0; 1095| 90.8k| capwords = false; 1096| | 1097| 90.8k| std::vector slst; 1098| | 1099| 90.8k| int onlycmpdsug = 0; 1100| 90.8k| if (!pSMgr || m_HMgrs.empty()) ------------------ | Branch (1100:7): [True: 0, False: 90.8k] | Branch (1100:17): [True: 0, False: 90.8k] ------------------ 1101| 0| return slst; 1102| | 1103| | // process XML input of the simplified API (see manual) 1104| 90.8k| if (word.compare(0, sizeof(SPELL_XML) - 3, SPELL_XML, sizeof(SPELL_XML) - 3) == 0) { ------------------ | | 79| 90.8k|#define SPELL_XML "" ------------------ if (word.compare(0, sizeof(SPELL_XML) - 3, SPELL_XML, sizeof(SPELL_XML) - 3) == 0) { ------------------ | | 79| 90.8k|#define SPELL_XML "" ------------------ if (word.compare(0, sizeof(SPELL_XML) - 3, SPELL_XML, sizeof(SPELL_XML) - 3) == 0) { ------------------ | | 79| 90.8k|#define SPELL_XML "" ------------------ | Branch (1104:7): [True: 20.5k, False: 70.2k] ------------------ 1105| 20.5k| if (word.size() > MAXSPELLMLLEN) ------------------ | | 87| 20.5k|#define MAXSPELLMLLEN 8192 ------------------ | Branch (1105:9): [True: 456, False: 20.1k] ------------------ 1106| 456| return slst; 1107| 20.1k| return spellml(word); 1108| 20.5k| } 1109| 70.2k| if (utf8) { ------------------ | Branch (1109:7): [True: 33.6k, False: 36.5k] ------------------ 1110| 33.6k| if (word.size() >= MAXWORDUTF8LEN) ------------------ | | 86| 33.6k|#define MAXWORDUTF8LEN (MAXWORDLEN * 3) | | ------------------ | | | | 89| 33.6k|#define MAXWORDLEN 100 | | ------------------ ------------------ | Branch (1110:9): [True: 126, False: 33.5k] ------------------ 1111| 126| return slst; 1112| 36.5k| } else { 1113| 36.5k| if (word.size() >= MAXWORDLEN) ------------------ | | 89| 36.5k|#define MAXWORDLEN 100 ------------------ | Branch (1113:9): [True: 160, False: 36.4k] ------------------ 1114| 160| return slst; 1115| 36.5k| } 1116| 69.9k| size_t wl = 0; 1117| | 1118| 69.9k| std::string scw; 1119| 69.9k| std::vector sunicw; 1120| | 1121| | // input conversion 1122| 69.9k| RepList* rl = (pAMgr) ? pAMgr->get_iconvtable() : nullptr; ------------------ | Branch (1122:17): [True: 69.9k, False: 0] ------------------ 1123| 69.9k| { 1124| 69.9k| std::string wspace; 1125| | 1126| 69.9k| bool convstatus = rl ? rl->conv(word, wspace) : false; ------------------ | Branch (1126:23): [True: 3.75k, False: 66.2k] ------------------ 1127| 69.9k| if (convstatus) ------------------ | Branch (1127:9): [True: 0, False: 69.9k] ------------------ 1128| 0| wl = cleanword2(scw, sunicw, wspace, &captype, &abbv); 1129| 69.9k| else 1130| 69.9k| wl = cleanword2(scw, sunicw, word, &captype, &abbv); 1131| | 1132| 69.9k| if (wl == 0) ------------------ | Branch (1132:9): [True: 0, False: 69.9k] ------------------ 1133| 0| return slst; 1134| | 1135| 69.9k|#if defined(FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION) 1136| 69.9k| if (wl > 32768) ------------------ | Branch (1136:9): [True: 0, False: 69.9k] ------------------ 1137| 0| return slst; 1138| 69.9k|#endif 1139| 69.9k| } 1140| | 1141| 69.9k| bool good = false; 1142| | 1143| | // check capitalized form for FORCEUCASE 1144| 69.9k| if (pAMgr && captype == NOCAP && pAMgr->get_forceucase()) { ------------------ | | 92| 139k|#define NOCAP 0 ------------------ | Branch (1144:7): [True: 69.9k, False: 0] | Branch (1144:16): [True: 20.5k, False: 49.4k] | Branch (1144:36): [True: 0, False: 20.5k] ------------------ 1145| 0| int info = SPELL_ORIGCAP; ------------------ | | 83| 0|#define SPELL_ORIGCAP (1 << 5) ------------------ 1146| 0| if (checkword(scw, &info, nullptr, suggest_start)) { ------------------ | Branch (1146:9): [True: 0, False: 0] ------------------ 1147| 0| std::string form(std::move(scw)); 1148| 0| mkinitcap(form); 1149| 0| slst.push_back(std::move(form)); 1150| 0| return slst; 1151| 0| } 1152| 0| } 1153| | 1154| 69.9k| switch (captype) { ------------------ | Branch (1154:11): [True: 69.9k, False: 0] ------------------ 1155| 20.5k| case NOCAP: { ------------------ | | 92| 20.5k|#define NOCAP 0 ------------------ | Branch (1155:5): [True: 20.5k, False: 49.4k] ------------------ 1156| 20.5k| good |= pSMgr->suggest(slst, scw, &onlycmpdsug); 1157| 20.5k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 20.5k|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (1157:11): [True: 0, False: 20.5k] ------------------ 1158| 0| return slst; 1159| 20.5k| if (abbv) { ------------------ | Branch (1159:11): [True: 2.20k, False: 18.3k] ------------------ 1160| 2.20k| std::string wspace(scw); 1161| 2.20k| wspace.push_back('.'); 1162| 2.20k| good |= pSMgr->suggest(slst, wspace, &onlycmpdsug); 1163| 2.20k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 2.20k|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (1163:13): [True: 0, False: 2.20k] ------------------ 1164| 0| return slst; 1165| 2.20k| } 1166| 20.5k| break; 1167| 20.5k| } 1168| | 1169| 20.5k| case INITCAP: { ------------------ | | 93| 12.9k|#define INITCAP 1 ------------------ | Branch (1169:5): [True: 12.9k, False: 57.0k] ------------------ 1170| 12.9k| capwords = true; 1171| 12.9k| good |= pSMgr->suggest(slst, scw, &onlycmpdsug); 1172| 12.9k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 12.9k|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (1172:11): [True: 0, False: 12.9k] ------------------ 1173| 0| return slst; 1174| 12.9k| std::string wspace(scw); 1175| 12.9k| mkallsmall2(wspace, sunicw); 1176| 12.9k| good |= pSMgr->suggest(slst, wspace, &onlycmpdsug); 1177| 12.9k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 12.9k|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (1177:11): [True: 0, False: 12.9k] ------------------ 1178| 0| return slst; 1179| 12.9k| break; 1180| 12.9k| } 1181| 12.9k| case HUHINITCAP: ------------------ | | 96| 12.6k|#define HUHINITCAP 4 ------------------ | Branch (1181:5): [True: 12.6k, False: 57.3k] ------------------ 1182| 12.6k| capwords = true; 1183| | /* FALLTHROUGH */ 1184| 26.2k| case HUHCAP: { ------------------ | | 95| 26.2k|#define HUHCAP 3 ------------------ | Branch (1184:5): [True: 13.6k, False: 56.3k] ------------------ 1185| 26.2k| good |= pSMgr->suggest(slst, scw, &onlycmpdsug); 1186| 26.2k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 26.2k|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (1186:11): [True: 0, False: 26.2k] ------------------ 1187| 0| return slst; 1188| | // something.The -> something. The 1189| 26.2k| size_t dot_pos = scw.find('.'); 1190| 26.2k| if (dot_pos != std::string::npos) { ------------------ | Branch (1190:11): [True: 8.44k, False: 17.8k] ------------------ 1191| 8.44k| std::string postdot = scw.substr(dot_pos + 1); 1192| 8.44k| int captype_; 1193| 8.44k| if (utf8) { ------------------ | Branch (1193:13): [True: 2.17k, False: 6.26k] ------------------ 1194| 2.17k| std::vector postdotu; 1195| 2.17k| u8_u16(postdotu, postdot); 1196| 2.17k| captype_ = get_captype_utf8(postdotu, langnum); 1197| 6.26k| } else { 1198| 6.26k| captype_ = get_captype(postdot, csconv); 1199| 6.26k| } 1200| 8.44k| if (captype_ == INITCAP) { ------------------ | | 93| 8.44k|#define INITCAP 1 ------------------ | Branch (1200:13): [True: 2.01k, False: 6.42k] ------------------ 1201| 2.01k| std::string str(scw); 1202| 2.01k| str.insert(dot_pos + 1, 1, ' '); 1203| 2.01k| insert_sug(slst, str); 1204| 2.01k| } 1205| 8.44k| } 1206| | 1207| 26.2k| std::string wspace; 1208| | 1209| 26.2k| if (captype == HUHINITCAP) { ------------------ | | 96| 26.2k|#define HUHINITCAP 4 ------------------ | Branch (1209:11): [True: 12.6k, False: 13.6k] ------------------ 1210| | // TheOpenOffice.org -> The OpenOffice.org 1211| 12.6k| wspace = scw; 1212| 12.6k| mkinitsmall2(wspace, sunicw); 1213| 12.6k| good |= pSMgr->suggest(slst, wspace, &onlycmpdsug); 1214| 12.6k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 12.6k|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (1214:13): [True: 1, False: 12.6k] ------------------ 1215| 1| return slst; 1216| 12.6k| } 1217| 26.2k| wspace = scw; 1218| 26.2k| mkallsmall2(wspace, sunicw); 1219| 26.2k| if (spell(wspace, spell_candidate_stack, nullptr, nullptr, suggest_start)) ------------------ | Branch (1219:11): [True: 895, False: 25.3k] ------------------ 1220| 895| insert_sug(slst, wspace); 1221| 26.2k| size_t prevns = slst.size(); 1222| 26.2k| good |= pSMgr->suggest(slst, wspace, &onlycmpdsug); 1223| 26.2k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 26.2k|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (1223:11): [True: 0, False: 26.2k] ------------------ 1224| 0| return slst; 1225| 26.2k| if (captype == HUHINITCAP) { ------------------ | | 96| 26.2k|#define HUHINITCAP 4 ------------------ | Branch (1225:11): [True: 12.6k, False: 13.6k] ------------------ 1226| 12.6k| mkinitcap2(wspace, sunicw); 1227| 12.6k| if (spell(wspace, spell_candidate_stack, nullptr, nullptr, suggest_start)) ------------------ | Branch (1227:13): [True: 278, False: 12.3k] ------------------ 1228| 278| insert_sug(slst, wspace); 1229| 12.6k| good |= pSMgr->suggest(slst, wspace, &onlycmpdsug); 1230| 12.6k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 12.6k|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (1230:13): [True: 5, False: 12.6k] ------------------ 1231| 5| return slst; 1232| 12.6k| } 1233| | // aNew -> "a New" (instead of "a new") 1234| 26.6k| for (size_t j = prevns; j < slst.size(); ++j) { ------------------ | Branch (1234:31): [True: 418, False: 26.2k] ------------------ 1235| 418| const char* space = strchr(slst[j].c_str(), ' '); 1236| 418| if (space) { ------------------ | Branch (1236:13): [True: 293, False: 125] ------------------ 1237| 293| size_t slen = strlen(space + 1); 1238| | // different case after space (need capitalisation) 1239| 293| if ((slen < wl) && strcmp(scw.c_str() + wl - slen, space + 1) != 0) { ------------------ | Branch (1239:15): [True: 293, False: 0] | Branch (1239:30): [True: 293, False: 0] ------------------ 1240| 293| std::string first(slst[j].c_str(), space + 1); 1241| 293| std::string second(space + 1); 1242| 293| std::vector w; 1243| 293| if (utf8) ------------------ | Branch (1243:17): [True: 87, False: 206] ------------------ 1244| 87| u8_u16(w, second); 1245| 293| mkinitcap2(second, w); 1246| | // set as first suggestion 1247| 293| slst.erase(slst.begin() + j); 1248| 293| slst.insert(slst.begin(), first + second); 1249| 293| } 1250| 293| } 1251| 418| } 1252| 26.2k| break; 1253| 26.2k| } 1254| | 1255| 10.1k| case ALLCAP: { ------------------ | | 94| 10.1k|#define ALLCAP 2 ------------------ | Branch (1255:5): [True: 10.1k, False: 59.7k] ------------------ 1256| 10.1k| std::string wspace(scw); 1257| 10.1k| mkallsmall2(wspace, sunicw); 1258| 10.1k| good |= pSMgr->suggest(slst, wspace, &onlycmpdsug); 1259| 10.1k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 10.1k|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (1259:11): [True: 0, False: 10.1k] ------------------ 1260| 0| return slst; 1261| 10.1k| if (pAMgr && pAMgr->get_keepcase() && spell(wspace, spell_candidate_stack, nullptr, nullptr, suggest_start)) ------------------ | Branch (1261:11): [True: 10.1k, False: 0] | Branch (1261:20): [True: 457, False: 9.73k] | Branch (1261:45): [True: 0, False: 457] ------------------ 1262| 0| insert_sug(slst, wspace); 1263| 10.1k| mkinitcap2(wspace, sunicw); 1264| 10.1k| good |= pSMgr->suggest(slst, wspace, &onlycmpdsug); 1265| 10.1k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 10.1k|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (1265:11): [True: 0, False: 10.1k] ------------------ 1266| 0| return slst; 1267| 10.1k| for (auto& j : slst) { ------------------ | Branch (1267:20): [True: 484, False: 10.1k] ------------------ 1268| 484| mkallcap(j); 1269| 484| if (pAMgr && pAMgr->get_checksharps()) { ------------------ | Branch (1269:13): [True: 484, False: 0] | Branch (1269:22): [True: 0, False: 484] ------------------ 1270| 0| if (utf8) { ------------------ | Branch (1270:15): [True: 0, False: 0] ------------------ 1271| 0| mystrrep(j, "\xC3\x9F", "SS"); 1272| 0| } else { 1273| 0| mystrrep(j, "\xDF", "SS"); 1274| 0| } 1275| 0| } 1276| 484| } 1277| 10.1k| break; 1278| 10.1k| } 1279| 69.9k| } 1280| | 1281| | // LANG_hu section: replace '-' with ' ' in Hungarian 1282| 69.9k| if (langnum == LANG_hu) { ------------------ | Branch (1282:7): [True: 3.81k, False: 66.1k] ------------------ 1283| 3.81k| for (auto& j : slst) { ------------------ | Branch (1283:18): [True: 173, False: 3.81k] ------------------ 1284| 173| size_t pos = j.find('-'); 1285| 173| if (pos != std::string::npos) { ------------------ | Branch (1285:11): [True: 35, False: 138] ------------------ 1286| 35| int info = 0; 1287| 35| std::string w(j.substr(0, pos)); 1288| 35| w.append(j.substr(pos + 1)); 1289| 35| (void)spell(w, spell_candidate_stack, &info, nullptr, suggest_start); 1290| 35| if ((info & SPELL_COMPOUND) && (info & SPELL_FORBIDDEN)) { ------------------ | | 78| 35|#define SPELL_COMPOUND (1 << 0) // the result is a compound word ------------------ if ((info & SPELL_COMPOUND) && (info & SPELL_FORBIDDEN)) { ------------------ | | 79| 0|#define SPELL_FORBIDDEN (1 << 1) ------------------ | Branch (1290:13): [True: 0, False: 35] | Branch (1290:40): [True: 0, False: 0] ------------------ 1291| 0| j[pos] = ' '; 1292| 0| } else 1293| 35| j[pos] = '-'; 1294| 35| } 1295| 173| } 1296| 3.81k| } 1297| | // END OF LANG_hu section 1298| | // try ngram approach since found nothing good suggestion 1299| 69.9k| if (!good && pAMgr && (slst.empty() || onlycmpdsug) && (pAMgr->get_maxngramsugs() != 0)) { ------------------ | Branch (1299:7): [True: 69.9k, False: 75] | Branch (1299:16): [True: 69.9k, False: 0] | Branch (1299:26): [True: 65.2k, False: 4.67k] | Branch (1299:42): [True: 4.65k, False: 25] | Branch (1299:58): [True: 62.3k, False: 7.53k] ------------------ 1300| 62.3k| switch (captype) { ------------------ | Branch (1300:13): [True: 62.3k, False: 0] ------------------ 1301| 17.7k| case NOCAP: { ------------------ | | 92| 17.7k|#define NOCAP 0 ------------------ | Branch (1301:7): [True: 17.7k, False: 44.6k] ------------------ 1302| 17.7k| pSMgr->ngsuggest(slst, scw.c_str(), m_HMgrs, NOCAP); ------------------ | | 92| 17.7k|#define NOCAP 0 ------------------ 1303| 17.7k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 17.7k|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (1303:13): [True: 0, False: 17.7k] ------------------ 1304| 0| return slst; 1305| 17.7k| break; 1306| 17.7k| } 1307| | /* FALLTHROUGH */ 1308| 17.7k| case HUHINITCAP: ------------------ | | 96| 11.9k|#define HUHINITCAP 4 ------------------ | Branch (1308:7): [True: 11.9k, False: 50.3k] ------------------ 1309| 11.9k| capwords = true; 1310| | /* FALLTHROUGH */ 1311| 24.5k| case HUHCAP: { ------------------ | | 95| 24.5k|#define HUHCAP 3 ------------------ | Branch (1311:7): [True: 12.6k, False: 49.7k] ------------------ 1312| 24.5k| std::string wspace(scw); 1313| 24.5k| mkallsmall2(wspace, sunicw); 1314| 24.5k| pSMgr->ngsuggest(slst, wspace.c_str(), m_HMgrs, HUHCAP); ------------------ | | 95| 24.5k|#define HUHCAP 3 ------------------ 1315| 24.5k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 24.5k|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (1315:13): [True: 0, False: 24.5k] ------------------ 1316| 0| return slst; 1317| 24.5k| break; 1318| 24.5k| } 1319| 24.5k| case INITCAP: { ------------------ | | 93| 11.1k|#define INITCAP 1 ------------------ | Branch (1319:7): [True: 11.1k, False: 51.2k] ------------------ 1320| 11.1k| capwords = true; 1321| 11.1k| std::string wspace(scw); 1322| 11.1k| mkallsmall2(wspace, sunicw); 1323| 11.1k| pSMgr->ngsuggest(slst, wspace.c_str(), m_HMgrs, INITCAP); ------------------ | | 93| 11.1k|#define INITCAP 1 ------------------ 1324| 11.1k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 11.1k|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (1324:13): [True: 0, False: 11.1k] ------------------ 1325| 0| return slst; 1326| 11.1k| break; 1327| 11.1k| } 1328| 11.1k| case ALLCAP: { ------------------ | | 94| 8.91k|#define ALLCAP 2 ------------------ | Branch (1328:7): [True: 8.91k, False: 53.4k] ------------------ 1329| 8.91k| std::string wspace(scw); 1330| 8.91k| mkallsmall2(wspace, sunicw); 1331| 8.91k| size_t oldns = slst.size(); 1332| 8.91k| pSMgr->ngsuggest(slst, wspace.c_str(), m_HMgrs, ALLCAP); ------------------ | | 94| 8.91k|#define ALLCAP 2 ------------------ 1333| 8.91k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 8.91k|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (1333:13): [True: 0, False: 8.91k] ------------------ 1334| 0| return slst; 1335| 11.8k| for (size_t j = oldns; j < slst.size(); ++j) { ------------------ | Branch (1335:32): [True: 2.89k, False: 8.91k] ------------------ 1336| 2.89k| mkallcap(slst[j]); 1337| 2.89k| } 1338| 8.91k| break; 1339| 8.91k| } 1340| 62.3k| } 1341| 62.3k| } 1342| | 1343| | // try dash suggestion (Afo-American -> Afro-American) 1344| | // Note: LibreOffice was modified to treat dashes as word 1345| | // characters to check "scot-free" etc. word forms, but 1346| | // we need to handle suggestions for "Afo-American", etc., 1347| | // while "Afro-American" is missing from the dictionary. 1348| | // TODO avoid possible overgeneration 1349| 69.9k| size_t dash_pos = scw.find('-'); 1350| 69.9k| if (dash_pos != std::string::npos) { ------------------ | Branch (1350:7): [True: 34.3k, False: 35.6k] ------------------ 1351| 34.3k| int nodashsug = 1; 1352| 41.2k| for (size_t j = 0; j < slst.size() && nodashsug == 1; ++j) { ------------------ | Branch (1352:24): [True: 7.33k, False: 33.8k] | Branch (1352:43): [True: 6.88k, False: 459] ------------------ 1353| 6.88k| if (slst[j].find('-') != std::string::npos) ------------------ | Branch (1353:11): [True: 2.76k, False: 4.11k] ------------------ 1354| 2.76k| nodashsug = 0; 1355| 6.88k| } 1356| | 1357| 34.3k| size_t prev_pos = 0; 1358| 34.3k| bool last = false; 1359| | 1360| 107k| while (!good && nodashsug && !last) { ------------------ | Branch (1360:12): [True: 107k, False: 42] | Branch (1360:21): [True: 73.7k, False: 33.6k] | Branch (1360:34): [True: 73.0k, False: 667] ------------------ 1361| 73.0k| if (dash_pos == scw.size()) ------------------ | Branch (1361:11): [True: 5.70k, False: 67.3k] ------------------ 1362| 5.70k| last = true; 1363| 73.0k| std::string chunk = scw.substr(prev_pos, dash_pos - prev_pos); 1364| 73.0k| if (chunk != word && !spell(chunk, spell_candidate_stack, nullptr, nullptr, suggest_start)) { ------------------ | Branch (1364:11): [True: 73.0k, False: 0] | Branch (1364:28): [True: 30.8k, False: 42.1k] ------------------ 1365| 30.8k| std::vector nlst = suggest(chunk, suggest_candidate_stack, suggest_start); 1366| 30.8k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 30.8k|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (1366:13): [True: 3, False: 30.8k] ------------------ 1367| 3| return slst; 1368| 40.9k| for (auto j = nlst.rbegin(); j != nlst.rend(); ++j) { ------------------ | Branch (1368:38): [True: 10.0k, False: 30.8k] ------------------ 1369| 10.0k| std::string wspace = scw.substr(0, prev_pos); 1370| 10.0k| wspace.append(*j); 1371| 10.0k| if (!last) { ------------------ | Branch (1371:15): [True: 8.02k, False: 2.01k] ------------------ 1372| 8.02k| wspace.append("-"); 1373| 8.02k| wspace.append(scw.substr(dash_pos + 1)); 1374| 8.02k| } 1375| 10.0k| int info = 0; 1376| 10.0k| if (pAMgr && pAMgr->get_forbiddenword()) ------------------ | Branch (1376:15): [True: 10.0k, False: 0] | Branch (1376:24): [True: 10.0k, False: 0] ------------------ 1377| 10.0k| checkword(wspace, &info, nullptr, suggest_start); 1378| 10.0k| if (!(info & SPELL_FORBIDDEN)) ------------------ | | 79| 10.0k|#define SPELL_FORBIDDEN (1 << 1) ------------------ | Branch (1378:15): [True: 10.0k, False: 0] ------------------ 1379| 10.0k| insert_sug(slst, wspace); 1380| 10.0k| } 1381| 30.8k| nodashsug = 0; 1382| 30.8k| } 1383| 73.0k| if (!last) { ------------------ | Branch (1383:11): [True: 67.3k, False: 5.70k] ------------------ 1384| 67.3k| prev_pos = dash_pos + 1; 1385| 67.3k| dash_pos = scw.find('-', prev_pos); 1386| 67.3k| } 1387| 73.0k| if (dash_pos == std::string::npos) ------------------ | Branch (1387:11): [True: 12.8k, False: 60.1k] ------------------ 1388| 12.8k| dash_pos = scw.size(); 1389| 73.0k| } 1390| 34.3k| } 1391| 69.9k| return slst; 1392| 69.9k|} _ZN12HunspellImpl4stemERKNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEE: 1398| 1.10k|std::vector HunspellImpl::stem(const std::vector& desc) { 1399| 1.10k| std::vector slst; 1400| | 1401| 1.10k| std::string result2; 1402| 1.10k| if (desc.empty()) ------------------ | Branch (1402:7): [True: 974, False: 128] ------------------ 1403| 974| return slst; 1404| 128| for (const auto& i : desc) { ------------------ | Branch (1404:22): [True: 128, False: 128] ------------------ 1405| 128| std::string result; 1406| | 1407| | // add compound word parts (except the last one) 1408| 128| const char* s = i.c_str(); 1409| 128| const char* part = strstr(s, MORPH_PART); ------------------ | | 116| 128|#define MORPH_PART "pa:" ------------------ 1410| 128| if (part) { ------------------ | Branch (1410:9): [True: 0, False: 128] ------------------ 1411| 0| const char* nextpart = strstr(part + 1, MORPH_PART); ------------------ | | 116| 0|#define MORPH_PART "pa:" ------------------ 1412| 0| while (nextpart) { ------------------ | Branch (1412:14): [True: 0, False: 0] ------------------ 1413| 0| std::string field; 1414| 0| copy_field(field, part, MORPH_PART); ------------------ | | 116| 0|#define MORPH_PART "pa:" ------------------ 1415| 0| result.append(field); 1416| 0| part = nextpart; 1417| 0| nextpart = strstr(part + 1, MORPH_PART); ------------------ | | 116| 0|#define MORPH_PART "pa:" ------------------ 1418| 0| } 1419| 0| s = part; 1420| 0| } 1421| | 1422| 128| std::string tok(s); 1423| 128| size_t alt = 0; 1424| 128| while ((alt = tok.find(" | ", alt)) != std::string::npos) { ------------------ | Branch (1424:12): [True: 0, False: 128] ------------------ 1425| 0| tok[alt + 1] = MSEP_ALT; ------------------ | | 123| 0|#define MSEP_ALT '\v' ------------------ 1426| 0| } 1427| 128| std::vector pl = line_tok(tok, MSEP_ALT); ------------------ | | 123| 128|#define MSEP_ALT '\v' ------------------ 1428| 128| for (auto& k : pl) { ------------------ | Branch (1428:18): [True: 128, False: 128] ------------------ 1429| | // add derivational suffixes 1430| 128| if (k.find(MORPH_DERI_SFX) != std::string::npos) { ------------------ | | 109| 128|#define MORPH_DERI_SFX "ds:" ------------------ | Branch (1430:11): [True: 0, False: 128] ------------------ 1431| | // remove inflectional suffixes 1432| 0| const size_t is = k.find(MORPH_INFL_SFX); ------------------ | | 110| 0|#define MORPH_INFL_SFX "is:" ------------------ 1433| 0| if (is != std::string::npos) ------------------ | Branch (1433:13): [True: 0, False: 0] ------------------ 1434| 0| k.resize(is); 1435| 0| std::vector singlepl; 1436| 0| singlepl.push_back(k); 1437| 0| std::string sg = pSMgr->suggest_gen(singlepl, k); 1438| 0| if (!sg.empty()) { ------------------ | Branch (1438:13): [True: 0, False: 0] ------------------ 1439| 0| std::vector gen = line_tok(sg, MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 1440| 0| for (auto& j : gen) { ------------------ | Branch (1440:24): [True: 0, False: 0] ------------------ 1441| 0| result2.push_back(MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 1442| 0| result2.append(result); 1443| 0| result2.append(j); 1444| 0| } 1445| 0| } 1446| 128| } else { 1447| 128| result2.push_back(MSEP_REC); ------------------ | | 122| 128|#define MSEP_REC '\n' ------------------ 1448| 128| result2.append(result); 1449| 128| if (k.find(MORPH_SURF_PFX) != std::string::npos) { ------------------ | | 112| 128|#define MORPH_SURF_PFX "sp:" ------------------ | Branch (1449:13): [True: 0, False: 128] ------------------ 1450| 0| std::string field; 1451| 0| copy_field(field, k, MORPH_SURF_PFX); ------------------ | | 112| 0|#define MORPH_SURF_PFX "sp:" ------------------ 1452| 0| result2.append(field); 1453| 0| } 1454| 128| std::string field; 1455| 128| copy_field(field, k, MORPH_STEM); ------------------ | | 103| 128|#define MORPH_STEM "st:" ------------------ 1456| 128| result2.append(field); 1457| 128| } 1458| 128| } 1459| 128| } 1460| 128| slst = line_tok(result2, MSEP_REC); ------------------ | | 122| 128|#define MSEP_REC '\n' ------------------ 1461| 128| uniqlist(slst); 1462| 128| return slst; 1463| 1.10k|} _ZN12HunspellImpl4stemERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEE: 1465| 1.10k|std::vector HunspellImpl::stem(const std::string& word) { 1466| 1.10k| return stem(analyze(word)); 1467| 1.10k|} _ZN12HunspellImpl9mkinitcapERNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEE: 1477| 14.2k|void HunspellImpl::mkinitcap(std::string& u8) { 1478| 14.2k| if (utf8) { ------------------ | Branch (1478:7): [True: 4.17k, False: 10.0k] ------------------ 1479| 4.17k| std::vector u16; 1480| 4.17k| u8_u16(u16, u8); 1481| 4.17k| ::mkinitcap_utf(u16, langnum); 1482| 4.17k| u16_u8(u8, u16); 1483| 10.0k| } else { 1484| 10.0k| ::mkinitcap(u8, csconv); 1485| 10.0k| } 1486| 14.2k|} _ZN12HunspellImpl10mkinitcap2ERNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEERNS0_6vectorI6w_charNS4_IS9_EEEE: 1488| 964k|int HunspellImpl::mkinitcap2(std::string& u8, std::vector& u16) { 1489| 964k| if (utf8) { ------------------ | Branch (1489:7): [True: 258k, False: 705k] ------------------ 1490| 258k| ::mkinitcap_utf(u16, langnum); 1491| 258k| u16_u8(u8, u16); 1492| 705k| } else { 1493| 705k| ::mkinitcap(u8, csconv); 1494| 705k| } 1495| 964k| return u8.size(); 1496| 964k|} _ZN12HunspellImpl12mkinitsmall2ERNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEERNS0_6vectorI6w_charNS4_IS9_EEEE: 1498| 12.6k|int HunspellImpl::mkinitsmall2(std::string& u8, std::vector& u16) { 1499| 12.6k| if (utf8) { ------------------ | Branch (1499:7): [True: 2.87k, False: 9.75k] ------------------ 1500| 2.87k| ::mkinitsmall_utf(u16, langnum); 1501| 2.87k| u16_u8(u8, u16); 1502| 9.75k| } else { 1503| 9.75k| ::mkinitsmall(u8, csconv); 1504| 9.75k| } 1505| 12.6k| return u8.size(); 1506| 12.6k|} _ZN12HunspellImpl3addERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEE: 1508| 5.03k|int HunspellImpl::add(const std::string& word) { 1509| 5.03k| if (!m_HMgrs.empty()) ------------------ | Branch (1509:7): [True: 5.03k, False: 0] ------------------ 1510| 5.03k| return m_HMgrs[0]->add(word); 1511| 0| return 0; 1512| 5.03k|} _ZN12HunspellImpl14add_with_affixERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEES8_: 1520| 2.04k|int HunspellImpl::add_with_affix(const std::string& word, const std::string& example) { 1521| 2.04k| if (!m_HMgrs.empty()) ------------------ | Branch (1521:7): [True: 2.04k, False: 0] ------------------ 1522| 2.04k| return m_HMgrs[0]->add_with_affix(word, example); 1523| 0| return 0; 1524| 2.04k|} _ZN12HunspellImpl10cat_resultERNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEERKS6_: 1543| 383k|void HunspellImpl::cat_result(std::string& result, const std::string& st) { 1544| 383k| if (!st.empty()) { ------------------ | Branch (1544:7): [True: 2.58k, False: 381k] ------------------ 1545| 2.58k| if (!result.empty()) ------------------ | Branch (1545:9): [True: 152, False: 2.43k] ------------------ 1546| 152| result.append("\n"); 1547| 2.58k| result.append(st); 1548| 2.58k| } 1549| 383k|} _ZN12HunspellImpl7analyzeERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEE: 1551| 7.69k|std::vector HunspellImpl::analyze(const std::string& word) { 1552| 7.69k| std::vector slst = analyze_internal(word); 1553| | // output conversion 1554| 7.69k| RepList* rl = (pAMgr) ? pAMgr->get_oconvtable() : nullptr; ------------------ | Branch (1554:17): [True: 7.69k, False: 0] ------------------ 1555| 7.69k| if (rl) { ------------------ | Branch (1555:7): [True: 403, False: 7.29k] ------------------ 1556| 522| for (size_t i = 0; rl && i < slst.size(); ++i) { ------------------ | Branch (1556:24): [True: 522, False: 0] | Branch (1556:30): [True: 119, False: 403] ------------------ 1557| 119| std::string wspace; 1558| 119| if (rl->conv(slst[i], wspace)) { ------------------ | Branch (1558:11): [True: 5, False: 114] ------------------ 1559| 5| slst[i] = std::move(wspace); 1560| 5| } 1561| 119| } 1562| 403| } 1563| 7.69k| return slst; 1564| 7.69k|} _ZN12HunspellImpl16analyze_internalERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEE: 1566| 7.69k|std::vector HunspellImpl::analyze_internal(const std::string& word) { 1567| 7.69k| std::vector candidate_stack, slst; 1568| 7.69k| if (!pSMgr || m_HMgrs.empty()) ------------------ | Branch (1568:7): [True: 0, False: 7.69k] | Branch (1568:17): [True: 0, False: 7.69k] ------------------ 1569| 0| return slst; 1570| 7.69k| if (utf8) { ------------------ | Branch (1570:7): [True: 3.64k, False: 4.05k] ------------------ 1571| 3.64k| if (word.size() >= MAXWORDUTF8LEN) ------------------ | | 86| 3.64k|#define MAXWORDUTF8LEN (MAXWORDLEN * 3) | | ------------------ | | | | 89| 3.64k|#define MAXWORDLEN 100 | | ------------------ ------------------ | Branch (1571:9): [True: 244, False: 3.39k] ------------------ 1572| 244| return slst; 1573| 4.05k| } else { 1574| 4.05k| if (word.size() >= MAXWORDLEN) ------------------ | | 89| 4.05k|#define MAXWORDLEN 100 ------------------ | Branch (1574:9): [True: 450, False: 3.60k] ------------------ 1575| 450| return slst; 1576| 4.05k| } 1577| 7.00k| int captype = NOCAP; ------------------ | | 92| 7.00k|#define NOCAP 0 ------------------ 1578| 7.00k| size_t abbv = 0; 1579| 7.00k| size_t wl = 0; 1580| | 1581| 7.00k| std::string scw; 1582| 7.00k| std::vector sunicw; 1583| | 1584| | // input conversion 1585| 7.00k| RepList* rl = (pAMgr) ? pAMgr->get_iconvtable() : nullptr; ------------------ | Branch (1585:17): [True: 7.00k, False: 0] ------------------ 1586| 7.00k| { 1587| 7.00k| std::string wspace; 1588| | 1589| 7.00k| bool convstatus = rl ? rl->conv(word, wspace) : false; ------------------ | Branch (1589:23): [True: 376, False: 6.62k] ------------------ 1590| 7.00k| if (convstatus) ------------------ | Branch (1590:9): [True: 0, False: 7.00k] ------------------ 1591| 0| wl = cleanword2(scw, sunicw, wspace, &captype, &abbv); 1592| 7.00k| else 1593| 7.00k| wl = cleanword2(scw, sunicw, word, &captype, &abbv); 1594| | 1595| 7.00k|#if defined(FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION) 1596| 7.00k| if (wl > 32768) ------------------ | Branch (1596:9): [True: 0, False: 7.00k] ------------------ 1597| 0| return slst; 1598| 7.00k|#endif 1599| 7.00k| } 1600| | 1601| 7.00k| if (wl == 0) { ------------------ | Branch (1601:7): [True: 421, False: 6.58k] ------------------ 1602| 421| if (abbv) { ------------------ | Branch (1602:9): [True: 266, False: 155] ------------------ 1603| 266| scw.clear(); 1604| 5.13k| for (wl = 0; wl < abbv; wl++) ------------------ | Branch (1604:20): [True: 4.86k, False: 266] ------------------ 1605| 4.86k| scw.push_back('.'); 1606| 266| abbv = 0; 1607| 266| } else 1608| 155| return slst; 1609| 421| } 1610| | 1611| 6.84k| std::string result; 1612| | 1613| 6.84k| size_t n = 0; 1614| | // test numbers 1615| | // LANG_hu section: set dash information for suggestions 1616| 6.84k| if (langnum == LANG_hu) { ------------------ | Branch (1616:7): [True: 370, False: 6.47k] ------------------ 1617| 370| size_t n2 = 0; 1618| 370| size_t n3 = 0; 1619| | 1620| 524| while ((n < wl) && (((scw[n] <= '9') && (scw[n] >= '0')) || ------------------ | Branch (1620:12): [True: 491, False: 33] | Branch (1620:26): [True: 249, False: 242] | Branch (1620:45): [True: 154, False: 95] ------------------ 1621| 337| (((scw[n] == '.') || (scw[n] == ',')) && (n > 0)))) { ------------------ | Branch (1621:27): [True: 22, False: 315] | Branch (1621:46): [True: 0, False: 315] | Branch (1621:66): [True: 0, False: 22] ------------------ 1622| 154| n++; 1623| 154| if ((scw[n] == '.') || (scw[n] == ',')) { ------------------ | Branch (1623:11): [True: 0, False: 154] | Branch (1623:30): [True: 0, False: 154] ------------------ 1624| 0| if (((n2 == 0) && (n > 3)) || ------------------ | Branch (1624:14): [True: 0, False: 0] | Branch (1624:27): [True: 0, False: 0] ------------------ 1625| 0| ((n2 > 0) && ((scw[n - 1] == '.') || (scw[n - 1] == ',')))) ------------------ | Branch (1625:14): [True: 0, False: 0] | Branch (1625:27): [True: 0, False: 0] | Branch (1625:50): [True: 0, False: 0] ------------------ 1626| 0| break; 1627| 0| n2++; 1628| 0| n3 = n; 1629| 0| } 1630| 154| } 1631| | 1632| 370| if ((n == wl) && (n3 > 0) && (n - n3 > 3)) ------------------ | Branch (1632:9): [True: 33, False: 337] | Branch (1632:22): [True: 0, False: 33] | Branch (1632:34): [True: 0, False: 0] ------------------ 1633| 0| return slst; 1634| 370| if ((n == wl) || ((n > 0) && ((scw[n] == '%') || (scw[n] == '\xB0')) && checkword(scw.substr(n), nullptr, nullptr))) { ------------------ | Branch (1634:9): [True: 33, False: 337] | Branch (1634:9): [True: 33, False: 337] | Branch (1634:23): [True: 2, False: 335] | Branch (1634:35): [True: 0, False: 2] | Branch (1634:54): [True: 0, False: 2] | Branch (1634:77): [True: 0, False: 0] ------------------ 1635| 33| result.append(scw); 1636| 33| result.resize(n - 1); 1637| 33| if (n == wl) ------------------ | Branch (1637:11): [True: 33, False: 0] ------------------ 1638| 33| cat_result(result, pSMgr->suggest_morph(scw.substr(n - 1))); 1639| 0| else { 1640| 0| std::string chunk = scw.substr(n - 1, 1); 1641| 0| cat_result(result, pSMgr->suggest_morph(chunk)); 1642| 0| result.push_back('+'); // XXX SPEC. MORPHCODE 1643| 0| cat_result(result, pSMgr->suggest_morph(scw.substr(n))); 1644| 0| } 1645| 33| return line_tok(result, MSEP_REC); ------------------ | | 122| 33|#define MSEP_REC '\n' ------------------ 1646| 33| } 1647| 370| } 1648| | // END OF LANG_hu section 1649| | 1650| 6.81k| switch (captype) { ------------------ | Branch (1650:11): [True: 6.81k, False: 0] ------------------ 1651| 592| case HUHCAP: ------------------ | | 95| 592|#define HUHCAP 3 ------------------ | Branch (1651:5): [True: 592, False: 6.22k] ------------------ 1652| 950| case HUHINITCAP: ------------------ | | 96| 950|#define HUHINITCAP 4 ------------------ | Branch (1652:5): [True: 358, False: 6.45k] ------------------ 1653| 3.95k| case NOCAP: { ------------------ | | 92| 3.95k|#define NOCAP 0 ------------------ | Branch (1653:5): [True: 3.00k, False: 3.80k] ------------------ 1654| 3.95k| cat_result(result, pSMgr->suggest_morph(scw)); 1655| 3.95k| if (abbv) { ------------------ | Branch (1655:11): [True: 389, False: 3.56k] ------------------ 1656| 389| std::string u8buffer(scw); 1657| 389| u8buffer.push_back('.'); 1658| 389| cat_result(result, pSMgr->suggest_morph(u8buffer)); 1659| 389| } 1660| 3.95k| break; 1661| 950| } 1662| 1.82k| case INITCAP: { ------------------ | | 93| 1.82k|#define INITCAP 1 ------------------ | Branch (1662:5): [True: 1.82k, False: 4.99k] ------------------ 1663| 1.82k| mkallsmall2(scw, sunicw); 1664| 1.82k| std::string u8buffer(scw); 1665| 1.82k| mkinitcap2(scw, sunicw); 1666| 1.82k| cat_result(result, pSMgr->suggest_morph(u8buffer)); 1667| 1.82k| cat_result(result, pSMgr->suggest_morph(scw)); 1668| 1.82k| if (abbv) { ------------------ | Branch (1668:11): [True: 357, False: 1.46k] ------------------ 1669| 357| u8buffer.push_back('.'); 1670| 357| cat_result(result, pSMgr->suggest_morph(u8buffer)); 1671| | 1672| 357| u8buffer = scw; 1673| 357| u8buffer.push_back('.'); 1674| | 1675| 357| cat_result(result, pSMgr->suggest_morph(u8buffer)); 1676| 357| } 1677| 1.82k| break; 1678| 950| } 1679| 1.03k| case ALLCAP: { ------------------ | | 94| 1.03k|#define ALLCAP 2 ------------------ | Branch (1679:5): [True: 1.03k, False: 5.78k] ------------------ 1680| 1.03k| cat_result(result, pSMgr->suggest_morph(scw)); 1681| 1.03k| if (abbv) { ------------------ | Branch (1681:11): [True: 298, False: 735] ------------------ 1682| 298| std::string u8buffer(scw); 1683| 298| u8buffer.push_back('.'); 1684| 298| cat_result(result, pSMgr->suggest_morph(u8buffer)); 1685| 298| } 1686| 1.03k| mkallsmall2(scw, sunicw); 1687| 1.03k| std::string u8buffer(scw); 1688| 1.03k| mkinitcap2(scw, sunicw); 1689| | 1690| 1.03k| cat_result(result, pSMgr->suggest_morph(u8buffer)); 1691| 1.03k| cat_result(result, pSMgr->suggest_morph(scw)); 1692| 1.03k| if (abbv) { ------------------ | Branch (1692:11): [True: 298, False: 735] ------------------ 1693| 298| u8buffer.push_back('.'); 1694| 298| cat_result(result, pSMgr->suggest_morph(u8buffer)); 1695| | 1696| 298| u8buffer = scw; 1697| 298| u8buffer.push_back('.'); 1698| | 1699| 298| cat_result(result, pSMgr->suggest_morph(u8buffer)); 1700| 298| } 1701| 1.03k| break; 1702| 950| } 1703| 6.81k| } 1704| | 1705| 6.81k| if (!result.empty()) { ------------------ | Branch (1705:7): [True: 2.41k, False: 4.40k] ------------------ 1706| | // word reversing wrapper for complex prefixes 1707| 2.41k| if (complexprefixes) { ------------------ | Branch (1707:9): [True: 135, False: 2.27k] ------------------ 1708| 135| if (utf8) ------------------ | Branch (1708:11): [True: 0, False: 135] ------------------ 1709| 0| reverseword_utf(result); 1710| 135| else 1711| 135| reverseword(result); 1712| 135| } 1713| 2.41k| return line_tok(result, MSEP_REC); ------------------ | | 122| 2.41k|#define MSEP_REC '\n' ------------------ 1714| 2.41k| } 1715| | 1716| | // compound word with dash (HU) I18n 1717| | // LANG_hu section: set dash information for suggestions 1718| | 1719| 4.40k| size_t dash_pos = langnum == LANG_hu ? scw.find('-') : std::string::npos; ------------------ | Branch (1719:21): [True: 237, False: 4.16k] ------------------ 1720| 4.40k| if (dash_pos != std::string::npos) { ------------------ | Branch (1720:7): [True: 38, False: 4.36k] ------------------ 1721| 38| int nresult = 0; 1722| | 1723| 38| std::string part1 = scw.substr(0, dash_pos), part2 = scw.substr(dash_pos + 1); 1724| | 1725| | // examine 2 sides of the dash 1726| 38| if (part2.empty()) { // base word ending with dash ------------------ | Branch (1726:9): [True: 7, False: 31] ------------------ 1727| 7| if (spell(part1, candidate_stack)) { ------------------ | Branch (1727:11): [True: 1, False: 6] ------------------ 1728| 1| std::string p = pSMgr->suggest_morph(part1); 1729| 1| if (!p.empty()) { ------------------ | Branch (1729:13): [True: 0, False: 1] ------------------ 1730| 0| slst = line_tok(p, MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 1731| 0| return slst; 1732| 0| } 1733| 1| } 1734| 31| } else if (part2.size() == 1 && part2[0] == 'e') { // XXX (HU) -e hat. ------------------ | Branch (1734:16): [True: 1, False: 30] | Branch (1734:37): [True: 0, False: 1] ------------------ 1735| 0| if (spell(part1, candidate_stack) && (spell("-e", candidate_stack))) { ------------------ | Branch (1735:11): [True: 0, False: 0] | Branch (1735:11): [True: 0, False: 0] | Branch (1735:44): [True: 0, False: 0] ------------------ 1736| 0| std::string st = pSMgr->suggest_morph(part1); 1737| 0| if (!st.empty()) { ------------------ | Branch (1737:13): [True: 0, False: 0] ------------------ 1738| 0| result.append(st); 1739| 0| } 1740| 0| result.push_back('+'); // XXX spec. separator in MORPHCODE 1741| 0| st = pSMgr->suggest_morph("-e"); 1742| 0| if (!st.empty()) { ------------------ | Branch (1742:13): [True: 0, False: 0] ------------------ 1743| 0| result.append(st); 1744| 0| } 1745| 0| return line_tok(result, MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 1746| 0| } 1747| 31| } else { 1748| | // first word ending with dash: word- XXX ??? 1749| 31| part1.push_back(' '); 1750| 31| nresult = spell(part1, candidate_stack); 1751| 31| part1.erase(part1.size() - 1); 1752| 31| if (nresult && spell(part2, candidate_stack) && ------------------ | Branch (1752:11): [True: 3, False: 28] | Branch (1752:22): [True: 1, False: 2] ------------------ 1753| 1| ((part2.size() > 1) || ((part2[0] > '0') && (part2[0] < '9')))) { ------------------ | Branch (1753:12): [True: 0, False: 1] | Branch (1753:35): [True: 1, False: 0] | Branch (1753:55): [True: 1, False: 0] ------------------ 1754| 1| std::string st = pSMgr->suggest_morph(part1); 1755| 1| if (!st.empty()) { ------------------ | Branch (1755:13): [True: 0, False: 1] ------------------ 1756| 0| result.append(st); 1757| 0| result.push_back('+'); // XXX spec. separator in MORPHCODE 1758| 0| } 1759| 1| st = pSMgr->suggest_morph(part2); 1760| 1| if (!st.empty()) { ------------------ | Branch (1760:13): [True: 1, False: 0] ------------------ 1761| 1| result.append(st); 1762| 1| } 1763| 1| return line_tok(result, MSEP_REC); ------------------ | | 122| 1|#define MSEP_REC '\n' ------------------ 1764| 1| } 1765| 31| } 1766| | // affixed number in correct word 1767| 37| if (nresult && (dash_pos > 0) && ------------------ | Branch (1767:9): [True: 2, False: 35] | Branch (1767:20): [True: 0, False: 2] ------------------ 1768| 0| (((scw[dash_pos - 1] <= '9') && (scw[dash_pos - 1] >= '0')) || ------------------ | Branch (1768:11): [True: 0, False: 0] | Branch (1768:41): [True: 0, False: 0] ------------------ 1769| 0| (scw[dash_pos - 1] == '.'))) { ------------------ | Branch (1769:10): [True: 0, False: 0] ------------------ 1770| 0| n = 1; 1771| 0| if (scw[dash_pos - n] == '.') ------------------ | Branch (1771:11): [True: 0, False: 0] ------------------ 1772| 0| n++; 1773| | // search first not a number character to left from dash 1774| 0| while ((dash_pos >= n) && ((scw[dash_pos - n] == '0') || (n < 3)) && ------------------ | Branch (1774:14): [True: 0, False: 0] | Branch (1774:34): [True: 0, False: 0] | Branch (1774:64): [True: 0, False: 0] ------------------ 1775| 0| (n < 6)) { ------------------ | Branch (1775:14): [True: 0, False: 0] ------------------ 1776| 0| n++; 1777| 0| } 1778| 0| if (dash_pos < n) ------------------ | Branch (1778:11): [True: 0, False: 0] ------------------ 1779| 0| n--; 1780| | // numbers: valami1000000-hoz 1781| | // examine 100000-hoz, 10000-hoz 1000-hoz, 10-hoz, 1782| | // 56-hoz, 6-hoz 1783| 0| for (; n >= 1; n--) { ------------------ | Branch (1783:14): [True: 0, False: 0] ------------------ 1784| 0| if (scw[dash_pos - n] < '0' || scw[dash_pos - n] > '9') { ------------------ | Branch (1784:13): [True: 0, False: 0] | Branch (1784:40): [True: 0, False: 0] ------------------ 1785| 0| continue; 1786| 0| } 1787| 0| std::string chunk = scw.substr(dash_pos - n); 1788| 0| if (checkword(chunk, nullptr, nullptr)) { ------------------ | Branch (1788:13): [True: 0, False: 0] ------------------ 1789| 0| result.append(chunk); 1790| 0| std::string st = pSMgr->suggest_morph(chunk); 1791| 0| if (!st.empty()) { ------------------ | Branch (1791:15): [True: 0, False: 0] ------------------ 1792| 0| result.append(st); 1793| 0| } 1794| 0| return line_tok(result, MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 1795| 0| } 1796| 0| } 1797| 0| } 1798| 37| } 1799| 4.40k| return slst; 1800| 4.40k|} _ZN12HunspellImpl8generateERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEERKNS0_6vectorIS6_NS4_IS6_EEEE: 1802| 6.08k|std::vector HunspellImpl::generate(const std::string& word, const std::vector& pl) { 1803| 6.08k| std::vector slst; 1804| 6.08k| if (!pSMgr || pl.empty()) ------------------ | Branch (1804:7): [True: 0, False: 6.08k] | Branch (1804:17): [True: 548, False: 5.53k] ------------------ 1805| 548| return slst; 1806| 5.53k| std::vector pl2 = analyze(word); 1807| 5.53k| int captype = NOCAP, abbv = 0; ------------------ | | 92| 5.53k|#define NOCAP 0 ------------------ 1808| 5.53k| std::string cw; 1809| 5.53k| cleanword(cw, word, &captype, &abbv); 1810| 5.53k| std::string result; 1811| | 1812| 5.53k| auto suggest_start = std::chrono::steady_clock::now(); 1813| 371k| for (const auto& i : pl) { ------------------ | Branch (1813:22): [True: 371k, False: 5.53k] ------------------ 1814| 371k| cat_result(result, pSMgr->suggest_gen(pl2, i, suggest_start)); 1815| 371k| } 1816| | 1817| 5.53k| if (!result.empty()) { ------------------ | Branch (1817:7): [True: 0, False: 5.53k] ------------------ 1818| | // allcap 1819| 0| if (captype == ALLCAP) ------------------ | | 94| 0|#define ALLCAP 2 ------------------ | Branch (1819:9): [True: 0, False: 0] ------------------ 1820| 0| mkallcap(result); 1821| | 1822| | // line split 1823| 0| slst = line_tok(result, MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 1824| | 1825| | // capitalize 1826| 0| if (captype == INITCAP || captype == HUHINITCAP) { ------------------ | | 93| 0|#define INITCAP 1 ------------------ if (captype == INITCAP || captype == HUHINITCAP) { ------------------ | | 96| 0|#define HUHINITCAP 4 ------------------ | Branch (1826:9): [True: 0, False: 0] | Branch (1826:31): [True: 0, False: 0] ------------------ 1827| 0| for (auto& str : slst) { ------------------ | Branch (1827:22): [True: 0, False: 0] ------------------ 1828| 0| mkinitcap(str); 1829| 0| } 1830| 0| } 1831| | 1832| | // temporary filtering of prefix related errors (eg. 1833| | // generate("undrinkable", "eats") --> "undrinkables" and "*undrinks") 1834| 0| auto it = slst.begin(); 1835| 0| while (it != slst.end()) { ------------------ | Branch (1835:12): [True: 0, False: 0] ------------------ 1836| 0| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 0|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (1836:11): [True: 0, False: 0] ------------------ 1837| 0| break; 1838| 0| std::vector candidate_stack; 1839| 0| if (!spell(*it, candidate_stack, nullptr, nullptr, suggest_start)) { ------------------ | Branch (1839:11): [True: 0, False: 0] ------------------ 1840| 0| it = slst.erase(it); 1841| 0| } else { 1842| 0| ++it; 1843| 0| } 1844| 0| } 1845| 0| } 1846| 5.53k| return slst; 1847| 6.08k|} _ZN12HunspellImpl8generateERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEES8_: 1849| 779|std::vector HunspellImpl::generate(const std::string& word, const std::string& pattern) { 1850| 779| std::vector pl = analyze(pattern); 1851| 779| std::vector slst = generate(word, pl); 1852| 779| uniqlist(slst); 1853| 779| return slst; 1854| 779|} _ZN12HunspellImpl11get_xml_parERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEm: 1857| 457k|std::string HunspellImpl::get_xml_par(const std::string& in_par, std::string::size_type pos) { 1858| 457k| std::string dest; 1859| 457k| if (pos == std::string::npos) ------------------ | Branch (1859:7): [True: 5.92k, False: 451k] ------------------ 1860| 5.92k| return dest; 1861| 451k| const char* par = in_par.c_str() + pos; 1862| 451k| char end = *par; 1863| 451k| if (end == '>') ------------------ | Branch (1863:7): [True: 448k, False: 3.02k] ------------------ 1864| 448k| end = '<'; 1865| 3.02k| else if (end != '\'' && end != '"') ------------------ | Branch (1865:12): [True: 2.56k, False: 456] | Branch (1865:27): [True: 1.44k, False: 1.12k] ------------------ 1866| 1.44k| return dest; // bad XML 1867| 11.4M| for (par++; *par != '\0' && *par != end; ++par) { ------------------ | Branch (1867:15): [True: 11.4M, False: 35.9k] | Branch (1867:31): [True: 11.0M, False: 414k] ------------------ 1868| 11.0M| dest.push_back(*par); 1869| 11.0M| } 1870| 450k| mystrrep(dest, "<", "<"); 1871| 450k| mystrrep(dest, "&", "&"); 1872| 450k| return dest; 1873| 451k|} _ZN12HunspellImpl11get_xml_posERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEmPKc: 1889| 65.4k|std::string::size_type HunspellImpl::get_xml_pos(const std::string& s, std::string::size_type pos, const char* attr) { 1890| 65.4k| if (pos == std::string::npos) ------------------ | Branch (1890:7): [True: 0, False: 65.4k] ------------------ 1891| 0| return std::string::npos; 1892| | 1893| 65.4k| std::string::size_type endpos = s.find('>', pos); 1894| 65.4k| if (attr == nullptr) ------------------ | Branch (1894:7): [True: 0, False: 65.4k] ------------------ 1895| 0| return endpos; 1896| 250k| while (true) { ------------------ | Branch (1896:10): [True: 250k, Folded] ------------------ 1897| 250k| pos = s.find(attr, pos); 1898| 250k| if (pos == std::string::npos || pos >= endpos) ------------------ | Branch (1898:9): [True: 2.81k, False: 247k] | Branch (1898:37): [True: 1.90k, False: 245k] ------------------ 1899| 4.71k| return std::string::npos; 1900| 245k| if (pos == 0 || s[pos - 1] == ' ' || s[pos - 1] == '\n') ------------------ | Branch (1900:9): [True: 0, False: 245k] | Branch (1900:21): [True: 2.01k, False: 243k] | Branch (1900:42): [True: 58.7k, False: 185k] ------------------ 1901| 60.7k| break; 1902| 185k| pos += strlen(attr); 1903| 185k| } 1904| 60.7k| return pos + strlen(attr); 1905| 65.4k|} _ZN12HunspellImpl13check_xml_parERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEmPKcSA_: 1909| 65.4k| const char* value) { 1910| 65.4k| const std::string cw = get_xml_par(q, get_xml_pos(q, pos, attr)); 1911| 65.4k| return cw == value ? 1 : 0; ------------------ | Branch (1911:10): [True: 16.3k, False: 49.0k] ------------------ 1912| 65.4k|} _ZN12HunspellImpl12get_xml_listERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEmPKc: 1914| 5.89k|std::vector HunspellImpl::get_xml_list(const std::string& list, std::string::size_type pos, const char* tag) { 1915| 5.89k| std::vector slst; 1916| 5.89k| if (pos == std::string::npos) ------------------ | Branch (1916:7): [True: 19, False: 5.87k] ------------------ 1917| 19| return slst; 1918| 376k| while (true) { ------------------ | Branch (1918:10): [True: 376k, Folded] ------------------ 1919| 376k| pos = list.find(tag, pos); 1920| 376k| if (pos == std::string::npos) ------------------ | Branch (1920:9): [True: 4.35k, False: 372k] ------------------ 1921| 4.35k| break; 1922| 372k| std::string cw = get_xml_par(list, pos + strlen(tag) - 1); 1923| 372k| if (cw.empty()) { ------------------ | Branch (1923:9): [True: 1.52k, False: 370k] ------------------ 1924| 1.52k| break; 1925| 1.52k| } 1926| 370k| slst.push_back(std::move(cw)); 1927| 370k| ++pos; 1928| 370k| } 1929| 5.87k| return slst; 1930| 5.89k|} _ZN12HunspellImpl7spellmlERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEE: 1932| 20.1k|std::vector HunspellImpl::spellml(const std::string& in_word) { 1933| 20.1k| std::vector slst; 1934| | 1935| 20.1k| std::string::size_type qpos = in_word.find("', qpos); 1940| 19.7k| if (q2pos == std::string::npos) ------------------ | Branch (1940:7): [True: 133, False: 19.6k] ------------------ 1941| 133| return slst; // bad XML input 1942| | 1943| 19.6k| q2pos = in_word.find("', q2pos)); 1949| 589| if (!cw.empty()) ------------------ | Branch (1949:9): [True: 285, False: 304] ------------------ 1950| 285| slst = analyze(cw); 1951| 589| if (slst.empty()) ------------------ | Branch (1951:9): [True: 548, False: 41] ------------------ 1952| 548| return slst; 1953| | // convert the result to ana1ana2 format 1954| 41| std::string r; 1955| 41| r.append(""); 1956| 41| for (auto entry : slst) { ------------------ | Branch (1956:21): [True: 41, False: 41] ------------------ 1957| 41| r.append(""); 1958| | 1959| 41| mystrrep(entry, "\t", " "); 1960| 41| mystrrep(entry, "&", "&"); 1961| 41| mystrrep(entry, "<", "<"); 1962| 41| r.append(entry); 1963| | 1964| 41| r.append(""); 1965| 41| } 1966| 41| r.append(""); 1967| 41| slst.clear(); 1968| 41| slst.push_back(std::move(r)); 1969| 41| return slst; 1970| 18.7k| } else if (check_xml_par(in_word, qpos, "type=", "stem")) { ------------------ | Branch (1970:14): [True: 1.48k, False: 17.2k] ------------------ 1971| 1.48k| std::string cw = get_xml_par(in_word, in_word.find('>', q2pos)); 1972| 1.48k| if (!cw.empty()) ------------------ | Branch (1972:9): [True: 1.10k, False: 380] ------------------ 1973| 1.10k| return stem(cw); 1974| 17.2k| } else if (check_xml_par(in_word, qpos, "type=", "generate")) { ------------------ | Branch (1974:14): [True: 7.08k, False: 10.1k] ------------------ 1975| 7.08k| std::string cw = get_xml_par(in_word, in_word.find('>', q2pos)); 1976| 7.08k| if (cw.empty()) ------------------ | Branch (1976:9): [True: 209, False: 6.87k] ------------------ 1977| 209| return slst; 1978| 6.87k| std::string::size_type q3pos = in_word.find("', q3pos)); 1981| 855| if (!cw2.empty()) { ------------------ | Branch (1981:11): [True: 779, False: 76] ------------------ 1982| 779| return generate(cw, cw2); 1983| 779| } 1984| 6.02k| } else { 1985| 6.02k| q2pos = in_word.find(" slst2 = get_xml_list(in_word, in_word.find('>', q2pos), ""); 1988| 5.89k| if (!slst2.empty()) { ------------------ | Branch (1988:13): [True: 5.30k, False: 589] ------------------ 1989| 5.30k| slst = generate(cw, slst2); 1990| 5.30k| uniqlist(slst); 1991| 5.30k| return slst; 1992| 5.30k| } 1993| 5.89k| } 1994| 6.02k| } 1995| 10.1k| } else if (check_xml_par(in_word, qpos, "type=", "add")) { ------------------ | Branch (1995:14): [True: 7.19k, False: 2.96k] ------------------ 1996| 7.19k| std::string cw = get_xml_par(in_word, in_word.find('>', q2pos)); 1997| 7.19k| if (cw.empty()) ------------------ | Branch (1997:9): [True: 115, False: 7.08k] ------------------ 1998| 115| return slst; 1999| 7.08k| std::string::size_type q3pos = in_word.find("', q3pos)); 2002| 2.38k| if (!cw2.empty()) { ------------------ | Branch (2002:11): [True: 2.04k, False: 341] ------------------ 2003| 2.04k| add_with_affix(cw, cw2); 2004| 2.04k| } else { 2005| 341| add(cw); 2006| 341| } 2007| 4.69k| } else { 2008| 4.69k| add(cw); 2009| 4.69k| } 2010| 7.08k| } 2011| 11.2k| return slst; 2012| 19.3k|} _ZN8HunspellC2EPKcS1_S1_: 2162| 38| : m_Impl(new HunspellImpl(affpath, dpath, key)) { 2163| 38|} _ZN8Hunspell5spellERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEPiPS6_: 2174| 60.3k|bool Hunspell::spell(const std::string& word, int* info, std::string* root) { 2175| 60.3k| std::vector candidate_stack; 2176| 60.3k| return m_Impl->spell(word, candidate_stack, info, root, 2177| 60.3k| std::chrono::steady_clock::now()); 2178| 60.3k|} _ZN8Hunspell7suggestERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEE: 2180| 59.9k|std::vector Hunspell::suggest(const std::string& word) { 2181| 59.9k| return m_Impl->suggest(word); 2182| 59.9k|} _Z16init_phonet_hashR10phonetable: 38| 2|void init_phonet_hash(phonetable& parms) { 39| 2| memset(parms.hash, 0xff, HASHSIZE * sizeof(int)); ------------------ | | 33| 2|#define HASHSIZE 256 ------------------ 40| | 41| 212| for (int i = 0; parms.rules[i][0] != '\0'; i += 2) { ------------------ | Branch (41:19): [True: 210, False: 2] ------------------ 42| | /** set hash value **/ 43| 210| int k = (unsigned char)parms.rules[i][0]; 44| | 45| 210| if (parms.hash[k] < 0) { ------------------ | Branch (45:9): [True: 54, False: 156] ------------------ 46| 54| parms.hash[k] = i; 47| 54| } 48| 210| } 49| 2|} _Z6phonetRKNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEER10phonetable: 69| 10.9k|std::string phonet(const std::string& inword, phonetable& parms) { 70| | 71| 10.9k| int i, k = 0, p, z, k0, n0, p0 = -333; 72| 10.9k| char c; 73| 10.9k| using uchar = unsigned char; 74| | 75| 10.9k| size_t len = inword.size(); 76| 10.9k| if (len > MAXPHONETUTF8LEN) ------------------ | | 35| 10.9k|#define MAXPHONETUTF8LEN (MAXPHONETLEN * 4) | | ------------------ | | | | 34| 10.9k|#define MAXPHONETLEN 256 | | ------------------ ------------------ | Branch (76:7): [True: 0, False: 10.9k] ------------------ 77| 0| return {}; 78| 10.9k| char word[MAXPHONETUTF8LEN + 1]; 79| 10.9k| strncpy(word, inword.c_str(), MAXPHONETUTF8LEN); ------------------ | | 35| 10.9k|#define MAXPHONETUTF8LEN (MAXPHONETLEN * 4) | | ------------------ | | | | 34| 10.9k|#define MAXPHONETLEN 256 | | ------------------ ------------------ 80| 10.9k| word[MAXPHONETUTF8LEN] = '\0'; ------------------ | | 35| 10.9k|#define MAXPHONETUTF8LEN (MAXPHONETLEN * 4) | | ------------------ | | | | 34| 10.9k|#define MAXPHONETLEN 256 | | ------------------ ------------------ 81| | 82| 10.9k| std::string target; 83| | /** check word **/ 84| 10.9k| i = z = 0; 85| 94.5k| while ((c = word[i]) != '\0') { ------------------ | Branch (85:10): [True: 83.6k, False: 10.9k] ------------------ 86| 83.6k| int n = parms.hash[(uchar)c]; 87| 83.6k| int z0 = 0; 88| | 89| 83.6k| if (n >= 0 && !parms.rules[n].empty()) { ------------------ | Branch (89:9): [True: 24.8k, False: 58.7k] | Branch (89:19): [True: 24.8k, False: 0] ------------------ 90| | /** check all rules for the same letter **/ 91| 60.0k| while (parms.rules[n][0] == c) { ------------------ | Branch (91:14): [True: 58.4k, False: 1.58k] ------------------ 92| | /** check whole string **/ 93| 58.4k| k = 1; /** number of found letters **/ 94| 58.4k| p = 5; /** default priority **/ 95| 58.4k| const char*s = parms.rules[n].c_str(); 96| 58.4k| s++; /** important for (see below) "*(s-1)" **/ 97| | 98| 59.9k| while (*s != '\0' && word[i + k] == *s && !isdigit((unsigned char)*s) && strchr("(-<^$", *s) == nullptr) { ------------------ | Branch (98:16): [True: 38.3k, False: 21.5k] | Branch (98:30): [True: 1.51k, False: 36.8k] | Branch (98:51): [True: 1.51k, False: 0] | Branch (98:82): [True: 1.48k, False: 24] ------------------ 99| 1.48k| k++; 100| 1.48k| s++; 101| 1.48k| } 102| 58.4k| if (*s == '(') { ------------------ | Branch (102:13): [True: 3.33k, False: 55.1k] ------------------ 103| | /** check letters in "(..)" **/ 104| 3.33k| if (myisalpha(word[i + k]) // ...could be implied? ------------------ | Branch (104:15): [True: 2.16k, False: 1.16k] ------------------ 105| 2.16k| && strchr(s + 1, word[i + k]) != nullptr) { ------------------ | Branch (105:18): [True: 282, False: 1.88k] ------------------ 106| 282| k++; 107| 1.60k| while (*s && *s != ')') ------------------ | Branch (107:20): [True: 1.60k, False: 0] | Branch (107:26): [True: 1.32k, False: 282] ------------------ 108| 1.32k| s++; 109| 282| if (*s == ')') ------------------ | Branch (109:17): [True: 282, False: 0] ------------------ 110| 282| s++; 111| 282| } 112| 3.33k| } 113| 58.4k| p0 = (int)*s; 114| 58.4k| k0 = k; 115| 59.5k| while (*s == '-' && k > 1) { ------------------ | Branch (115:16): [True: 1.08k, False: 58.4k] | Branch (115:29): [True: 1.08k, False: 0] ------------------ 116| 1.08k| k--; 117| 1.08k| s++; 118| 1.08k| } 119| 58.4k| if (*s == '<') ------------------ | Branch (119:13): [True: 143, False: 58.3k] ------------------ 120| 143| s++; 121| 58.4k| if (isdigit((unsigned char)*s)) { ------------------ | Branch (121:13): [True: 0, False: 58.4k] ------------------ 122| | /** determine priority **/ 123| 0| p = *s - '0'; 124| 0| s++; 125| 0| } 126| 58.4k| if (*s == '^' && *(s + 1) == '^') ------------------ | Branch (126:13): [True: 2.28k, False: 56.1k] | Branch (126:26): [True: 0, False: 2.28k] ------------------ 127| 0| s++; 128| | 129| 58.4k| if (*s == '\0' || (*s == '^' && (i == 0 || !myisalpha(word[i - 1])) && ------------------ | Branch (129:13): [True: 22.7k, False: 35.6k] | Branch (129:28): [True: 2.28k, False: 33.3k] | Branch (129:42): [True: 150, False: 2.13k] | Branch (129:52): [True: 596, False: 1.53k] ------------------ 130| 746| (*(s + 1) != '$' || (!myisalpha(word[i + k0])))) || ------------------ | Branch (130:29): [True: 746, False: 0] | Branch (130:48): [True: 0, False: 0] ------------------ 131| 34.9k| (*s == '$' && i > 0 && myisalpha(word[i - 1]) && ------------------ | Branch (131:14): [True: 0, False: 34.9k] | Branch (131:27): [True: 0, False: 0] | Branch (131:36): [True: 0, False: 0] ------------------ 132| 23.5k| (!myisalpha(word[i + k0])))) { ------------------ | Branch (132:14): [True: 0, False: 0] ------------------ 133| | /** search for followup rules, if: **/ 134| | /** parms.followup and k > 1 and NO '-' in searchstring **/ 135| 23.5k| char c0 = word[i + k - 1]; 136| 23.5k| n0 = parms.hash[(uchar)c0]; 137| | 138| | // if (parms.followup && k > 1 && n0 >= 0 139| 23.5k| if (k > 1 && n0 >= 0 && p0 != (int)'-' && word[i + k] != '\0' && !parms.rules[n0].empty()) { ------------------ | Branch (139:15): [True: 406, False: 23.1k] | Branch (139:24): [True: 406, False: 0] | Branch (139:35): [True: 361, False: 45] | Branch (139:53): [True: 328, False: 33] | Branch (139:76): [True: 328, False: 0] ------------------ 140| | /** test follow-up rule for "word[i+k]" **/ 141| 1.11k| while (parms.rules[n0][0] == c0) { ------------------ | Branch (141:20): [True: 1.01k, False: 98] ------------------ 142| | /** check whole string **/ 143| 1.01k| k0 = k; 144| 1.01k| p0 = 5; 145| 1.01k| s = parms.rules[n0].c_str(); 146| 1.01k| s++; 147| 1.24k| while (*s != '\0' && word[i + k0] == *s && !isdigit((unsigned char)*s) && strchr("(-<^$", *s) == nullptr) { ------------------ | Branch (147:22): [True: 966, False: 277] | Branch (147:36): [True: 230, False: 736] | Branch (147:58): [True: 230, False: 0] | Branch (147:89): [True: 230, False: 0] ------------------ 148| 230| k0++; 149| 230| s++; 150| 230| } 151| 1.01k| if (*s == '(') { ------------------ | Branch (151:19): [True: 52, False: 961] ------------------ 152| | /** check letters **/ 153| 52| if (myisalpha(word[i + k0]) && strchr(s + 1, word[i + k0]) != nullptr) { ------------------ | Branch (153:21): [True: 52, False: 0] | Branch (153:48): [True: 0, False: 52] ------------------ 154| 0| k0++; 155| 0| while (*s != ')' && *s != '\0') ------------------ | Branch (155:26): [True: 0, False: 0] | Branch (155:39): [True: 0, False: 0] ------------------ 156| 0| s++; 157| 0| if (*s == ')') ------------------ | Branch (157:23): [True: 0, False: 0] ------------------ 158| 0| s++; 159| 0| } 160| 52| } 161| 1.01k| while (*s == '-') { ------------------ | Branch (161:22): [True: 0, False: 1.01k] ------------------ 162| | /** "k0" gets NOT reduced **/ 163| | /** because "if (k0 == k)" **/ 164| 0| s++; 165| 0| } 166| 1.01k| if (*s == '<') ------------------ | Branch (166:19): [True: 51, False: 962] ------------------ 167| 51| s++; 168| 1.01k| if (isdigit((unsigned char)*s)) { ------------------ | Branch (168:19): [True: 0, False: 1.01k] ------------------ 169| 0| p0 = *s - '0'; 170| 0| s++; 171| 0| } 172| | 173| 1.01k| if (*s == '\0' ------------------ | Branch (173:19): [True: 328, False: 685] ------------------ 174| | /** *s == '^' cuts **/ 175| 685| || (*s == '$' && !myisalpha(word[i + k0]))) { ------------------ | Branch (175:23): [True: 0, False: 685] | Branch (175:36): [True: 0, False: 0] ------------------ 176| 328| if (k0 == k) { ------------------ | Branch (176:21): [True: 98, False: 230] ------------------ 177| | /** this is just a piece of the string **/ 178| 98| n0 += 2; 179| 98| continue; 180| 98| } 181| | 182| 230| if (p0 < p) { ------------------ | Branch (182:21): [True: 0, False: 230] ------------------ 183| | /** priority too low **/ 184| 0| n0 += 2; 185| 0| continue; 186| 0| } 187| | /** rule fits; stop search **/ 188| 230| break; 189| 230| } 190| 685| n0 += 2; 191| 685| } /** End of "while (parms.rules[n0][0] == c0)" **/ 192| | 193| 328| if (p0 >= p && parms.rules[n0][0] == c0) { ------------------ | Branch (193:17): [True: 328, False: 0] | Branch (193:28): [True: 230, False: 98] ------------------ 194| 230| n += 2; 195| 230| continue; 196| 230| } 197| 328| } /** end of follow-up stuff **/ 198| | 199| | /** replace string **/ 200| 23.3k| s = parms.rules[n + 1].c_str(); 201| 23.3k| p0 = (!parms.rules[n].empty() && strchr(parms.rules[n].c_str() + 1, '<') != nullptr) ? 1 : 0; ------------------ | Branch (201:17): [True: 23.3k, False: 0] | Branch (201:44): [True: 92, False: 23.2k] ------------------ 202| 23.3k| if (p0 == 1 && z == 0) { ------------------ | Branch (202:15): [True: 92, False: 23.2k] | Branch (202:26): [True: 92, False: 0] ------------------ 203| | /** rule with '<' is used **/ 204| 92| if (!target.empty() && *s != '\0' && ------------------ | Branch (204:17): [True: 67, False: 25] | Branch (204:36): [True: 67, False: 0] ------------------ 205| 67| (target[target.size()-1] == c || target[target.size()-1] == *s)) { ------------------ | Branch (205:18): [True: 0, False: 67] | Branch (205:50): [True: 0, False: 67] ------------------ 206| 0| target.erase(target.size() - 1); 207| 0| } 208| 92| z0 = 1; 209| 92| z = 1; 210| 92| k0 = 0; 211| 184| while (*s != '\0' && word[i + k0] != '\0') { ------------------ | Branch (211:20): [True: 92, False: 92] | Branch (211:34): [True: 92, False: 0] ------------------ 212| 92| word[i + k0] = *s; 213| 92| k0++; 214| 92| s++; 215| 92| } 216| 92| if (k > k0) ------------------ | Branch (216:17): [True: 6, False: 86] ------------------ 217| 6| strmove(&word[0] + i + k0, &word[0] + i + k); 218| | 219| | /** new "actual letter" **/ 220| 92| c = word[i]; 221| 23.2k| } else { /** no '<' rule used **/ 222| 23.2k| i += k - 1; 223| 23.2k| z = 0; 224| 23.5k| while (*s != '\0' && *(s + 1) != '\0' && target.size() < len) { ------------------ | Branch (224:20): [True: 22.5k, False: 989] | Branch (224:34): [True: 365, False: 22.1k] | Branch (224:54): [True: 305, False: 60] ------------------ 225| 305| if (target.empty() || target[target.size()-1] != *s) { ------------------ | Branch (225:19): [True: 8, False: 297] | Branch (225:37): [True: 93, False: 204] ------------------ 226| 101| target.push_back(*s); 227| 101| } 228| 305| s++; 229| 305| } 230| | /** new "actual letter" **/ 231| 23.2k| c = *s; 232| 23.2k| if (!parms.rules[n].empty() && strstr(parms.rules[n].c_str() + 1, "^^") != nullptr) { ------------------ | Branch (232:17): [True: 23.2k, False: 0] | Branch (232:44): [True: 0, False: 23.2k] ------------------ 233| 0| if (c != '\0') { ------------------ | Branch (233:19): [True: 0, False: 0] ------------------ 234| 0| target.push_back(c); 235| 0| } 236| 0| strmove(&word[0], &word[0] + i + 1); 237| 0| i = 0; 238| 0| z0 = 1; 239| 0| } 240| 23.2k| } 241| 23.3k| break; 242| 23.5k| } /** end of follow-up stuff **/ 243| 34.9k| n += 2; 244| 34.9k| } /** end of while (parms.rules[n][0] == c) **/ 245| 24.8k| } /** end of if (n >= 0) **/ 246| 83.6k| if (z0 == 0) { ------------------ | Branch (246:9): [True: 83.5k, False: 92] ------------------ 247| 83.5k| if (k && !p0 && target.size() < len && c != '\0') { ------------------ | Branch (247:11): [True: 24.8k, False: 58.7k] | Branch (247:16): [True: 23.2k, False: 1.58k] | Branch (247:23): [True: 23.1k, False: 75] | Branch (247:46): [True: 22.1k, False: 989] ------------------ 248| | /** condense only double letters **/ 249| 22.1k| target.push_back(c); 250| | /// printf("\n setting \n"); 251| 22.1k| } 252| | 253| 83.5k| i++; 254| 83.5k| z = 0; 255| 83.5k| k = 0; 256| 83.5k| } 257| 83.6k| } /** end of while ((c = word[i]) != '\0') **/ 258| | 259| 10.9k| return target; 260| 10.9k|} /** end of function "phonet" **/ phonet.cxx:_ZL9myisalphac: 59| 5.51k|static int myisalpha(char ch) { 60| 5.51k| if ((unsigned char)ch < 128) ------------------ | Branch (60:7): [True: 5.02k, False: 489] ------------------ 61| 5.02k| return isalpha(ch); 62| 489| return 1; 63| 5.51k|} phonet.cxx:_ZL7strmovePcS_: 53| 6|static inline void strmove(char* dest, char* src) { 54| 20| while (*src) ------------------ | Branch (54:10): [True: 14, False: 6] ------------------ 55| 14| *dest++ = *src++; 56| 6| *dest = '\0'; 57| 6|} _ZN7RepListC2Ei: 79| 4|RepList::RepList(int n) { 80| 4| dat.reserve(std::min(n, 16384)); 81| 4|} _ZN7RepList4findEPKc: 89| 3.05M|int RepList::find(const char* word) { 90| 3.05M| int p1 = 0; 91| 3.05M| int p2 = dat.size() - 1; 92| 3.05M| int ret = -1; 93| 9.54M| while (p1 <= p2) { ------------------ | Branch (93:10): [True: 6.49M, False: 3.05M] ------------------ 94| 6.49M| int m = ((unsigned)p1 + (unsigned)p2) >> 1; 95| 6.49M| int c = strncmp(word, dat[m]->pattern.c_str(), dat[m]->pattern.size()); 96| 6.49M| if (c < 0) ------------------ | Branch (96:9): [True: 5.36M, False: 1.12M] ------------------ 97| 5.36M| p2 = m - 1; 98| 1.12M| else if (c > 0) ------------------ | Branch (98:14): [True: 1.12M, False: 479] ------------------ 99| 1.12M| p1 = m + 1; 100| 479| else { // scan in the right half for a longer match 101| 479| ret = m; 102| 479| p1 = m + 1; 103| 479| } 104| 6.49M| } 105| 3.05M| return ret; 106| 3.05M|} _ZN7RepList7replaceEmib: 108| 479|std::string RepList::replace(const size_t wordlen, int ind, bool atstart) { 109| 479| int type = atstart ? 1 : 0; ------------------ | Branch (109:14): [True: 9, False: 470] ------------------ 110| 479| if (wordlen == dat[ind]->pattern.size()) ------------------ | Branch (110:7): [True: 101, False: 378] ------------------ 111| 101| type = atstart ? 3 : 2; ------------------ | Branch (111:12): [True: 0, False: 101] ------------------ 112| 589| while (type && dat[ind]->outstrings[type].empty()) ------------------ | Branch (112:10): [True: 110, False: 479] | Branch (112:18): [True: 110, False: 0] ------------------ 113| 110| type = (type == 2 && !atstart) ? 0 : type - 1; ------------------ | Branch (113:13): [True: 101, False: 9] | Branch (113:26): [True: 101, False: 0] ------------------ 114| 479| return dat[ind]->outstrings[type]; 115| 479|} _ZN7RepList3addERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEES8_: 117| 22|int RepList::add(const std::string& in_pat1, const std::string& pat2) { 118| 22| if (in_pat1.empty() || pat2.empty()) { ------------------ | Branch (118:7): [True: 0, False: 22] | Branch (118:26): [True: 0, False: 22] ------------------ 119| 0| return 1; 120| 0| } 121| | // analyse word context 122| 22| int type = 0; 123| 22| std::string pat1(in_pat1); 124| 22| if (pat1[0] == '_') { ------------------ | Branch (124:7): [True: 0, False: 22] ------------------ 125| 0| pat1.erase(0, 1); 126| 0| type = 1; 127| 0| } 128| 22| if (!pat1.empty() && pat1[pat1.size() - 1] == '_') { ------------------ | Branch (128:7): [True: 22, False: 0] | Branch (128:24): [True: 0, False: 22] ------------------ 129| 0| type = type + 2; 130| 0| pat1.erase(pat1.size() - 1); 131| 0| } 132| 22| mystrrep(pat1, "_", " "); 133| | 134| | // find existing entry 135| 22| int m = find(pat1.c_str()); 136| 22| if (m >= 0 && dat[m]->pattern == pat1) { ------------------ | Branch (136:7): [True: 0, False: 22] | Branch (136:17): [True: 0, False: 0] ------------------ 137| | // since already used 138| 0| dat[m]->outstrings[type] = pat2; 139| 0| mystrrep(dat[m]->outstrings[type], "_", " "); 140| 0| return 0; 141| 0| } 142| | 143| | // make a new entry if none exists 144| 22| replentry* r = new replentry; 145| 22| r->pattern = std::move(pat1); 146| 22| r->outstrings[type] = pat2; 147| 22| mystrrep(r->outstrings[type], "_", " "); 148| 22| dat.push_back(r); 149| | // sort to the right place in the list 150| 22| size_t i; 151| 36| for (i = dat.size() - 1; i > 0; --i) { ------------------ | Branch (151:28): [True: 30, False: 6] ------------------ 152| 30| if (strcmp(r->pattern.c_str(), dat[i - 1]->pattern.c_str()) < 0) { ------------------ | Branch (152:9): [True: 14, False: 16] ------------------ 153| 14| dat[i] = dat[i - 1]; 154| 14| } else 155| 16| break; 156| 30| } 157| 22| dat[i] = r; 158| 22| return 0; 159| 22|} _ZN7RepList4convERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEERS6_: 161| 345k|bool RepList::conv(const std::string& in_word, std::string& dest) { 162| 345k| dest.clear(); 163| | 164| 345k| const size_t wordlen = in_word.size(); 165| 345k| const char* word = in_word.c_str(); 166| | 167| 345k| bool change = false; 168| 3.40M| for (size_t i = 0; i < wordlen; ++i) { ------------------ | Branch (168:22): [True: 3.05M, False: 345k] ------------------ 169| 3.05M| int n = find(word + i); 170| | 171| 3.05M| bool empty = n < 0; 172| 3.05M| if (empty) { ------------------ | Branch (172:9): [True: 3.05M, False: 479] ------------------ 173| 3.05M| dest.push_back(word[i]); 174| 3.05M| continue; 175| 3.05M| } 176| | 177| 479| std::string l = replace(wordlen - i, n, i == 0); 178| 479| if (l.empty()) { ------------------ | Branch (178:9): [True: 0, False: 479] ------------------ 179| 0| dest.push_back(word[i]); 180| 0| continue; 181| 0| } 182| | 183| 479| dest.append(l); 184| 479| if (!dat[n]->pattern.empty()) { ------------------ | Branch (184:9): [True: 479, False: 0] ------------------ 185| 479| i += dat[n]->pattern.size() - 1; 186| 479| } 187| 479| change = true; 188| 479| } 189| | 190| 345k| return change; 191| 345k|} _ZNK7RepList24check_against_breaktableERKNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEE: 193| 2|bool RepList::check_against_breaktable(const std::vector& breaktable) const { 194| 8| for (const auto i : dat) { ------------------ | Branch (194:21): [True: 8, False: 2] ------------------ 195| 32| for (auto& outstring : i->outstrings) { ------------------ | Branch (195:26): [True: 32, False: 8] ------------------ 196| 96| for (const auto& str : breaktable) { ------------------ | Branch (196:28): [True: 96, False: 32] ------------------ 197| 96| if (outstring.find(str) != std::string::npos) { ------------------ | Branch (197:13): [True: 0, False: 96] ------------------ 198| 0| return false; 199| 0| } 200| 96| } 201| 32| } 202| 8| } 203| | 204| 2| return true; 205| 2|} _ZN10SuggestMgrC2ERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEjP8AffixMgr: 87| 38|SuggestMgr::SuggestMgr(const std::string& tryme, unsigned int maxn, AffixMgr* aptr) { 88| | // register affix manager and check in string of chars to 89| | // try when building candidate suggestions 90| 38| pAMgr = aptr; 91| | 92| 38| csconv = nullptr; 93| | 94| 38| ckeyl = 0; 95| | 96| 38| ctryl = 0; 97| | 98| 38| utf8 = 0; 99| 38| langnum = 0; 100| 38| complexprefixes = 0; 101| | 102| 38| maxSug = maxn; 103| 38| nosplitsugs = 0; 104| 38| maxngramsugs = MAXNGRAMSUGS; ------------------ | | 77| 38|#define MAXNGRAMSUGS 4 ------------------ 105| 38| maxcpdsugs = MAXCOMPOUNDSUGS; ------------------ | | 79| 38|#define MAXCOMPOUNDSUGS 3 ------------------ 106| | 107| 38| if (pAMgr) { ------------------ | Branch (107:7): [True: 38, False: 0] ------------------ 108| 38| langnum = pAMgr->get_langnum(); 109| 38| ckey = pAMgr->get_key_string(); 110| 38| nosplitsugs = pAMgr->get_nosplitsugs(); 111| 38| if (pAMgr->get_maxngramsugs() >= 0) ------------------ | Branch (111:9): [True: 4, False: 34] ------------------ 112| 4| maxngramsugs = pAMgr->get_maxngramsugs(); 113| 38| utf8 = pAMgr->get_utf8(); 114| 38| if (pAMgr->get_maxcpdsugs() >= 0) ------------------ | Branch (114:9): [True: 0, False: 38] ------------------ 115| 0| maxcpdsugs = pAMgr->get_maxcpdsugs(); 116| 38| if (!utf8) { ------------------ | Branch (116:9): [True: 20, False: 18] ------------------ 117| 20| csconv = get_current_cs(pAMgr->get_encoding()); 118| 20| } 119| 38| complexprefixes = pAMgr->get_complexprefixes(); 120| 38| } 121| | 122| 38| if (!ckey.empty()) { ------------------ | Branch (122:7): [True: 38, False: 0] ------------------ 123| 38| if (utf8) { ------------------ | Branch (123:9): [True: 18, False: 20] ------------------ 124| 18| int len = u8_u16(ckey_utf, ckey); 125| 18| if (len != -1) { ------------------ | Branch (125:11): [True: 18, False: 0] ------------------ 126| 18| ckeyl = len; 127| 18| } 128| 20| } else { 129| 20| ckeyl = ckey.size(); 130| 20| } 131| 38| } 132| | 133| 38| ctry = tryme; 134| 38| if (!ctry.empty()) { ------------------ | Branch (134:7): [True: 2, False: 36] ------------------ 135| 2| if (utf8) { ------------------ | Branch (135:9): [True: 2, False: 0] ------------------ 136| 2| int len = u8_u16(ctry_utf, ctry); 137| 2| if (len != -1) { ------------------ | Branch (137:11): [True: 2, False: 0] ------------------ 138| 2| ctryl = len; 139| 2| } 140| 2| } else { 141| 0| ctryl = ctry.size(); 142| 0| } 143| 2| } 144| | 145| | // language with possible dash usage 146| | // (latin letters or dash in TRY characters) 147| 38| lang_with_dash_usage = ctry.find('-') != std::string::npos || ------------------ | Branch (147:26): [True: 0, False: 38] ------------------ 148| 38| ctry.find('a') != std::string::npos; ------------------ | Branch (148:26): [True: 0, False: 38] ------------------ 149| 38|} _ZN10SuggestMgr7testsugERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKS7_iPiPNS0_6chrono10time_pointINSE_12steady_clockENSE_8durationIxNS0_5ratioILl1ELl1000000000EEEEEEERi: 166| 80.5M| int& info) { 167| 80.5M| if (wlst.size() == maxSug) ------------------ | Branch (167:7): [True: 0, False: 80.5M] ------------------ 168| 0| return; 169| | 170| 80.5M| const int cwrd = std::find(wlst.begin(), wlst.end(), candidate) != wlst.end() ? 0 : 1; ------------------ | Branch (170:20): [True: 21.7k, False: 80.4M] ------------------ 171| | 172| 80.5M| if (cwrd) { ------------------ | Branch (172:7): [True: 80.4M, False: 21.7k] ------------------ 173| 80.4M| if (int result = checkword(candidate, cpdsuggest, timer, timelimit)) { ------------------ | Branch (173:13): [True: 1.13k, False: 80.4M] ------------------ 174| | // compound word in the dictionary 175| 1.13k| if (cpdsuggest == 0 && result >= 2) ------------------ | Branch (175:11): [True: 1.13k, False: 0] | Branch (175:30): [True: 3, False: 1.13k] ------------------ 176| 3| info |= SPELL_COMPOUND; ------------------ | | 78| 3|#define SPELL_COMPOUND (1 << 0) // the result is a compound word ------------------ 177| 1.13k| wlst.push_back(candidate); 178| 1.13k| } 179| 80.4M| } 180| 80.5M|} _ZN10SuggestMgr7suggestERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKS7_Pib: 190| 146k| int* onlycompoundsug, bool test_simplesug) { 191| 146k| int nocompoundtwowords = 0; // no second or third loops, see below 192| 146k| std::vector word_utf; 193| 146k| size_t nsugorig = slst.size(), oldSug = 0; 194| 146k| std::string w2; 195| 146k| bool good_suggestion = false; 196| | 197| | // word reversing wrapper for complex prefixes 198| 146k| if (complexprefixes) { ------------------ | Branch (198:7): [True: 8.50k, False: 138k] ------------------ 199| 8.50k| w2.assign(w); 200| 8.50k| if (utf8) ------------------ | Branch (200:9): [True: 0, False: 8.50k] ------------------ 201| 0| reverseword_utf(w2); 202| 8.50k| else 203| 8.50k| reverseword(w2); 204| 8.50k| } 205| | 206| 146k| const std::string& word = complexprefixes ? w2 : w; ------------------ | Branch (206:29): [True: 8.50k, False: 138k] ------------------ 207| | 208| 146k| suggest_start = std::chrono::steady_clock::now(); 209| | 210| 146k| if (utf8) { ------------------ | Branch (210:7): [True: 61.6k, False: 85.2k] ------------------ 211| 61.6k| int wl = u8_u16(word_utf, word); 212| 61.6k| if (wl == -1) { ------------------ | Branch (212:9): [True: 9.90k, False: 51.7k] ------------------ 213| 9.90k| return false; 214| 9.90k| } 215| 61.6k| } 216| | 217| | // three loops: 218| | // - the first without compounding, 219| | // - the second one with 2-word compounding, 220| | // - the third one with 3-or-more-word compounding 221| | // Run second and third loops only if: 222| | // - no ~good suggestion in the first loop 223| | // - not for testing compound words with 3 or more words (test_simplesug == false) 224| 136k| int info = 0; 225| 547k| for (int cpdsuggest = 0; cpdsuggest < 3 && nocompoundtwowords == 0; cpdsuggest++) { ------------------ | Branch (225:28): [True: 410k, False: 136k] | Branch (225:46): [True: 410k, False: 35] ------------------ 226| | 227| | // limit compound suggestion 228| 410k| if (cpdsuggest > 0) ------------------ | Branch (228:9): [True: 273k, False: 136k] ------------------ 229| 273k| oldSug = slst.size(); 230| | 231| | // suggestions for an uppercase word (html -> HTML) 232| 410k| if (slst.size() < maxSug) { ------------------ | Branch (232:9): [True: 410k, False: 21] ------------------ 233| 410k| size_t i = slst.size(); 234| 410k| if (utf8) ------------------ | Branch (234:11): [True: 155k, False: 255k] ------------------ 235| 155k| capchars_utf(slst, word_utf, cpdsuggest, info); 236| 255k| else 237| 255k| capchars(slst, word, cpdsuggest, info); 238| 410k| if (slst.size() > i) ------------------ | Branch (238:11): [True: 33, False: 410k] ------------------ 239| 33| good_suggestion = true; 240| 410k| } 241| | 242| | // perhaps we made a typical fault of spelling 243| 410k| if ((slst.size() < maxSug) && (!cpdsuggest || (slst.size() < oldSug + maxcpdsugs))) { ------------------ | Branch (243:9): [True: 410k, False: 21] | Branch (243:36): [True: 136k, False: 273k] | Branch (243:51): [True: 273k, False: 0] ------------------ 244| 410k| size_t i = slst.size(); 245| 410k| replchars(slst, word, cpdsuggest, info); 246| 410k| if (slst.size() > i) { ------------------ | Branch (246:11): [True: 0, False: 410k] ------------------ 247| 0| good_suggestion = true; 248| 0| if (info & SPELL_BEST_SUG) ------------------ | | 86| 0|#define SPELL_BEST_SUG (1 << 8) // limit suggestions for the best ones, i.e. ph: ------------------ | Branch (248:13): [True: 0, False: 0] ------------------ 249| 0| return true; 250| 0| } 251| 410k| } 252| 410k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_SUGGESTION_MS) ------------------ | | 105| 410k|#define TIMELIMIT_SUGGESTION_MS std::chrono::milliseconds(100) ------------------ | Branch (252:9): [True: 0, False: 410k] ------------------ 253| 0| return good_suggestion; 254| 410k| if (test_simplesug && slst.size()) ------------------ | Branch (254:9): [True: 0, False: 410k] | Branch (254:27): [True: 0, False: 0] ------------------ 255| 0| return true; 256| | 257| | // perhaps we made chose the wrong char from a related set 258| 410k| if ((slst.size() < maxSug) && ------------------ | Branch (258:9): [True: 410k, False: 21] ------------------ 259| 410k| (!cpdsuggest || (slst.size() < oldSug + maxcpdsugs))) { ------------------ | Branch (259:10): [True: 136k, False: 273k] | Branch (259:25): [True: 273k, False: 0] ------------------ 260| 410k| mapchars(slst, word, cpdsuggest, info); 261| 410k| } 262| 410k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_SUGGESTION_MS) ------------------ | | 105| 410k|#define TIMELIMIT_SUGGESTION_MS std::chrono::milliseconds(100) ------------------ | Branch (262:9): [True: 2, False: 410k] ------------------ 263| 2| return good_suggestion; 264| 410k| if (test_simplesug && slst.size()) ------------------ | Branch (264:9): [True: 0, False: 410k] | Branch (264:27): [True: 0, False: 0] ------------------ 265| 0| return true; 266| | 267| | // only suggest compound words when no other ~good suggestion 268| 410k| if ((cpdsuggest == 0) && (slst.size() > nsugorig)) ------------------ | Branch (268:9): [True: 136k, False: 273k] | Branch (268:30): [True: 33, False: 136k] ------------------ 269| 33| nocompoundtwowords = 1; 270| | 271| | // did we swap the order of chars by mistake 272| 410k| if ((slst.size() < maxSug) && (!cpdsuggest || (slst.size() < oldSug + maxcpdsugs))) { ------------------ | Branch (272:9): [True: 410k, False: 21] | Branch (272:36): [True: 136k, False: 273k] | Branch (272:51): [True: 273k, False: 0] ------------------ 273| 410k| if (utf8) ------------------ | Branch (273:11): [True: 155k, False: 255k] ------------------ 274| 155k| swapchar_utf(slst, word_utf, cpdsuggest, info); 275| 255k| else 276| 255k| swapchar(slst, word, cpdsuggest, info); 277| 410k| } 278| 410k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_SUGGESTION_MS) ------------------ | | 105| 410k|#define TIMELIMIT_SUGGESTION_MS std::chrono::milliseconds(100) ------------------ | Branch (278:9): [True: 0, False: 410k] ------------------ 279| 0| return good_suggestion; 280| 410k| if (test_simplesug && slst.size()) ------------------ | Branch (280:9): [True: 0, False: 410k] | Branch (280:27): [True: 0, False: 0] ------------------ 281| 0| return true; 282| | 283| | // did we swap the order of non adjacent chars by mistake 284| 410k| if ((slst.size() < maxSug) && (!cpdsuggest || (slst.size() < oldSug + maxcpdsugs))) { ------------------ | Branch (284:9): [True: 410k, False: 21] | Branch (284:36): [True: 136k, False: 273k] | Branch (284:51): [True: 273k, False: 0] ------------------ 285| 410k| if (utf8) ------------------ | Branch (285:11): [True: 155k, False: 255k] ------------------ 286| 155k| longswapchar_utf(slst, word_utf, cpdsuggest, info); 287| 255k| else 288| 255k| longswapchar(slst, word, cpdsuggest, info); 289| 410k| } 290| 410k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_SUGGESTION_MS) ------------------ | | 105| 410k|#define TIMELIMIT_SUGGESTION_MS std::chrono::milliseconds(100) ------------------ | Branch (290:9): [True: 0, False: 410k] ------------------ 291| 0| return good_suggestion; 292| 410k| if (test_simplesug && slst.size()) ------------------ | Branch (292:9): [True: 0, False: 410k] | Branch (292:27): [True: 0, False: 0] ------------------ 293| 0| return true; 294| | 295| | // did we just hit the wrong key in place of a good char (case and keyboard) 296| 410k| if ((slst.size() < maxSug) && (!cpdsuggest || (slst.size() < oldSug + maxcpdsugs))) { ------------------ | Branch (296:9): [True: 410k, False: 21] | Branch (296:36): [True: 136k, False: 273k] | Branch (296:51): [True: 273k, False: 0] ------------------ 297| 410k| if (utf8) ------------------ | Branch (297:11): [True: 155k, False: 255k] ------------------ 298| 155k| badcharkey_utf(slst, word_utf, cpdsuggest, info); 299| 255k| else 300| 255k| badcharkey(slst, word, cpdsuggest, info); 301| 410k| } 302| 410k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_SUGGESTION_MS) ------------------ | | 105| 410k|#define TIMELIMIT_SUGGESTION_MS std::chrono::milliseconds(100) ------------------ | Branch (302:9): [True: 0, False: 410k] ------------------ 303| 0| return good_suggestion; 304| 410k| if (test_simplesug && slst.size()) ------------------ | Branch (304:9): [True: 0, False: 410k] | Branch (304:27): [True: 0, False: 0] ------------------ 305| 0| return true; 306| | 307| | // did we add a char that should not be there 308| 410k| if ((slst.size() < maxSug) && (!cpdsuggest || (slst.size() < oldSug + maxcpdsugs))) { ------------------ | Branch (308:9): [True: 410k, False: 21] | Branch (308:36): [True: 136k, False: 273k] | Branch (308:51): [True: 273k, False: 0] ------------------ 309| 410k| if (utf8) ------------------ | Branch (309:11): [True: 155k, False: 255k] ------------------ 310| 155k| extrachar_utf(slst, word_utf, cpdsuggest, info); 311| 255k| else 312| 255k| extrachar(slst, word, cpdsuggest, info); 313| 410k| } 314| 410k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_SUGGESTION_MS) ------------------ | | 105| 410k|#define TIMELIMIT_SUGGESTION_MS std::chrono::milliseconds(100) ------------------ | Branch (314:9): [True: 0, False: 410k] ------------------ 315| 0| return good_suggestion; 316| 410k| if (test_simplesug && slst.size()) ------------------ | Branch (316:9): [True: 0, False: 410k] | Branch (316:27): [True: 0, False: 0] ------------------ 317| 0| return true; 318| | 319| | // did we forgot a char 320| 410k| if ((slst.size() < maxSug) && (!cpdsuggest || (slst.size() < oldSug + maxcpdsugs))) { ------------------ | Branch (320:9): [True: 410k, False: 21] | Branch (320:36): [True: 136k, False: 273k] | Branch (320:51): [True: 273k, False: 0] ------------------ 321| 410k| if (utf8) ------------------ | Branch (321:11): [True: 155k, False: 255k] ------------------ 322| 155k| forgotchar_utf(slst, word_utf, cpdsuggest, info); 323| 255k| else 324| 255k| forgotchar(slst, word, cpdsuggest, info); 325| 410k| } 326| 410k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_SUGGESTION_MS) ------------------ | | 105| 410k|#define TIMELIMIT_SUGGESTION_MS std::chrono::milliseconds(100) ------------------ | Branch (326:9): [True: 0, False: 410k] ------------------ 327| 0| return good_suggestion; 328| 410k| if (test_simplesug && slst.size()) ------------------ | Branch (328:9): [True: 0, False: 410k] | Branch (328:27): [True: 0, False: 0] ------------------ 329| 0| return true; 330| | 331| | // did we move a char 332| 410k| if ((slst.size() < maxSug) && (!cpdsuggest || (slst.size() < oldSug + maxcpdsugs))) { ------------------ | Branch (332:9): [True: 410k, False: 21] | Branch (332:36): [True: 136k, False: 273k] | Branch (332:51): [True: 273k, False: 0] ------------------ 333| 410k| if (utf8) ------------------ | Branch (333:11): [True: 155k, False: 255k] ------------------ 334| 155k| movechar_utf(slst, word_utf, cpdsuggest, info); 335| 255k| else 336| 255k| movechar(slst, word, cpdsuggest, info); 337| 410k| } 338| 410k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_SUGGESTION_MS) ------------------ | | 105| 410k|#define TIMELIMIT_SUGGESTION_MS std::chrono::milliseconds(100) ------------------ | Branch (338:9): [True: 0, False: 410k] ------------------ 339| 0| return good_suggestion; 340| 410k| if (test_simplesug && slst.size()) ------------------ | Branch (340:9): [True: 0, False: 410k] | Branch (340:27): [True: 0, False: 0] ------------------ 341| 0| return true; 342| | 343| | // did we just hit the wrong key in place of a good char 344| 410k| if ((slst.size() < maxSug) && (!cpdsuggest || (slst.size() < oldSug + maxcpdsugs))) { ------------------ | Branch (344:9): [True: 410k, False: 21] | Branch (344:36): [True: 136k, False: 273k] | Branch (344:51): [True: 273k, False: 0] ------------------ 345| 410k| if (utf8) ------------------ | Branch (345:11): [True: 155k, False: 255k] ------------------ 346| 155k| badchar_utf(slst, word_utf, cpdsuggest, info); 347| 255k| else 348| 255k| badchar(slst, word, cpdsuggest, info); 349| 410k| } 350| 410k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_SUGGESTION_MS) ------------------ | | 105| 410k|#define TIMELIMIT_SUGGESTION_MS std::chrono::milliseconds(100) ------------------ | Branch (350:9): [True: 0, False: 410k] ------------------ 351| 0| return good_suggestion; 352| 410k| if (test_simplesug && slst.size()) ------------------ | Branch (352:9): [True: 0, False: 410k] | Branch (352:27): [True: 0, False: 0] ------------------ 353| 0| return true; 354| | 355| | // did we double two characters 356| 410k| if ((slst.size() < maxSug) && (!cpdsuggest || (slst.size() < oldSug + maxcpdsugs))) { ------------------ | Branch (356:9): [True: 410k, False: 21] | Branch (356:36): [True: 136k, False: 273k] | Branch (356:51): [True: 273k, False: 0] ------------------ 357| 410k| if (utf8) ------------------ | Branch (357:11): [True: 155k, False: 255k] ------------------ 358| 155k| doubletwochars_utf(slst, word_utf, cpdsuggest, info); 359| 255k| else 360| 255k| doubletwochars(slst, word, cpdsuggest, info); 361| 410k| } 362| 410k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_SUGGESTION_MS) ------------------ | | 105| 410k|#define TIMELIMIT_SUGGESTION_MS std::chrono::milliseconds(100) ------------------ | Branch (362:9): [True: 0, False: 410k] ------------------ 363| 0| return good_suggestion; 364| 410k| if (test_simplesug && slst.size()) ------------------ | Branch (364:9): [True: 0, False: 410k] | Branch (364:27): [True: 0, False: 0] ------------------ 365| 0| return true; 366| | 367| | // perhaps we forgot to hit space and two words ran together 368| | // (dictionary word pairs have top priority here, so 369| | // we always suggest them, in despite of nosplitsugs, and 370| | // drop compound word and other suggestions) 371| 410k| if (!cpdsuggest || (!nosplitsugs && slst.size() < oldSug + maxcpdsugs)) { ------------------ | Branch (371:9): [True: 136k, False: 273k] | Branch (371:25): [True: 273k, False: 0] | Branch (371:41): [True: 273k, False: 0] ------------------ 372| 410k| good_suggestion = twowords(slst, word, cpdsuggest, good_suggestion, info); 373| | 374| 410k| if (info & SPELL_BEST_SUG) ------------------ | | 86| 410k|#define SPELL_BEST_SUG (1 << 8) // limit suggestions for the best ones, i.e. ph: ------------------ | Branch (374:11): [True: 152, False: 410k] ------------------ 375| 152| return true; 376| 410k| } 377| 410k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_SUGGESTION_MS) ------------------ | | 105| 410k|#define TIMELIMIT_SUGGESTION_MS std::chrono::milliseconds(100) ------------------ | Branch (377:9): [True: 0, False: 410k] ------------------ 378| 0| return good_suggestion; 379| | 380| | // testing returns after the first loop 381| 410k| if (test_simplesug) ------------------ | Branch (381:9): [True: 0, False: 410k] ------------------ 382| 0| return slst.size() > 0; 383| | 384| | // don't need third loop, if the second loop was successful or 385| | // the first loop found a dictionary-based compound word 386| | // (we don't need more, likely worse and false 3-or-more-word compound words) 387| 410k| if (cpdsuggest == 1 && (slst.size() > oldSug || (info & SPELL_COMPOUND))) ------------------ | | 78| 136k|#define SPELL_COMPOUND (1 << 0) // the result is a compound word ------------------ | Branch (387:9): [True: 136k, False: 273k] | Branch (387:29): [True: 0, False: 136k] | Branch (387:53): [True: 2, False: 136k] ------------------ 388| 2| nocompoundtwowords = 1; 389| | 390| 410k| } // repeating ``for'' statement compounding support 391| | 392| 136k| if (!nocompoundtwowords && (!slst.empty()) && onlycompoundsug) ------------------ | Branch (392:7): [True: 136k, False: 35] | Branch (392:30): [True: 8.73k, False: 128k] | Branch (392:49): [True: 8.73k, False: 0] ------------------ 393| 8.73k| *onlycompoundsug = 1; 394| | 395| 136k| return good_suggestion; 396| 136k|} _ZN10SuggestMgr12capchars_utfERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKNS1_I6w_charNS5_ISB_EEEEiRi: 401| 155k| int cpdsuggest, int& info) { 402| 155k| std::vector candidate_utf(word); 403| 155k| mkallcap_utf(candidate_utf, langnum); 404| 155k| std::string candidate; 405| 155k| u16_u8(candidate, candidate_utf); 406| 155k| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 407| 155k|} _ZN10SuggestMgr8capcharsERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKS7_iRi: 412| 255k| int cpdsuggest, int& info) { 413| 255k| std::string candidate(word); 414| 255k| mkallcap(candidate, csconv); 415| 255k| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 416| 255k|} _ZN10SuggestMgr8mapcharsERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKS7_iRi: 421| 410k| int cpdsuggest, int& info) { 422| 410k| std::string candidate; 423| 410k| std::chrono::steady_clock::time_point timelimit; 424| 410k| int timer; 425| | 426| 410k| if (word.size() < 2 || !pAMgr) ------------------ | Branch (426:7): [True: 33.8k, False: 376k] | Branch (426:26): [True: 0, False: 376k] ------------------ 427| 33.8k| return wlst.size(); 428| | 429| 376k| const std::vector& maptable = pAMgr->get_maptable(); 430| 376k| if (maptable.empty()) ------------------ | Branch (430:7): [True: 361k, False: 15.3k] ------------------ 431| 361k| return wlst.size(); 432| | 433| 15.3k| timelimit = std::chrono::steady_clock::now(); 434| 15.3k| timer = MINTIMER; ------------------ | | 107| 15.3k|#define MINTIMER 100 ------------------ 435| 15.3k| return map_related(word, candidate, 0, wlst, cpdsuggest, 436| 15.3k| maptable, &timer, &timelimit, 0, info); 437| 376k|} _ZN10SuggestMgr11map_relatedERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEERS6_mRNS0_6vectorIS6_NS4_IS6_EEEEiRKNSA_ISC_NS4_ISC_EEEEPiPNS0_6chrono10time_pointINSJ_12steady_clockENSJ_8durationIxNS0_5ratioILl1ELl1000000000EEEEEEEiRi: 447| 1.63M| int depth, int& info) { 448| 1.63M| if (word.size() == wn) { ------------------ | Branch (448:7): [True: 199k, False: 1.44M] ------------------ 449| 199k| if (candidate == word) ------------------ | Branch (449:9): [True: 15.3k, False: 183k] ------------------ 450| 15.3k| return wlst.size(); 451| 183k| const int cwrd = std::find(wlst.begin(), wlst.end(), candidate) != wlst.end() ? 0 : 1; ------------------ | Branch (451:22): [True: 0, False: 183k] ------------------ 452| 183k| if ((cwrd) && checkword(candidate, cpdsuggest, timer, timelimit)) { ------------------ | Branch (452:9): [True: 183k, False: 0] | Branch (452:19): [True: 0, False: 183k] ------------------ 453| 0| if (wlst.size() < maxSug) { ------------------ | Branch (453:11): [True: 0, False: 0] ------------------ 454| 0| wlst.push_back(candidate); 455| 0| } 456| 0| } 457| 183k| return wlst.size(); 458| 199k| } 459| | 460| 1.44M| if (depth > 0x3F00) { ------------------ | Branch (460:7): [True: 0, False: 1.44M] ------------------ 461| 0| *timer = 0; 462| 0| return wlst.size(); 463| 0| } 464| | 465| 1.44M| int in_map = 0; 466| 5.76M| for (size_t j = 0; j < maptable.size(); ++j) { ------------------ | Branch (466:22): [True: 4.32M, False: 1.43M] ------------------ 467| 15.8M| for (size_t k = 0; k < maptable[j].size(); ++k) { ------------------ | Branch (467:24): [True: 11.5M, False: 4.32M] ------------------ 468| 11.5M| size_t len = maptable[j][k].size(); 469| 11.5M| if (len && word.compare(wn, len, maptable[j][k]) == 0) { ------------------ | Branch (469:11): [True: 11.5M, False: 0] | Branch (469:18): [True: 91.9k, False: 11.4M] ------------------ 470| 91.9k| in_map = 1; 471| 91.9k| size_t cn = candidate.size(); 472| 367k| for (size_t l = 0; l < maptable[j].size(); ++l) { ------------------ | Branch (472:28): [True: 275k, False: 91.8k] ------------------ 473| 275k| candidate.resize(cn); 474| 275k| candidate.append(maptable[j][l]); 475| 275k| map_related(word, candidate, wn + len, wlst, 476| 275k| cpdsuggest, maptable, timer, timelimit, depth + 1, info); 477| 275k| if (!(*timer)) ------------------ | Branch (477:15): [True: 38, False: 275k] ------------------ 478| 38| return wlst.size(); 479| 275k| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_SUGGESTION_MS) { ------------------ | | 105| 275k|#define TIMELIMIT_SUGGESTION_MS std::chrono::milliseconds(100) ------------------ | Branch (479:15): [True: 2, False: 275k] ------------------ 480| 2| *timer = 0; 481| 2| return wlst.size(); 482| 2| } 483| 275k| } 484| 91.9k| } 485| 11.5M| } 486| 4.32M| } 487| 1.43M| if (!in_map) { ------------------ | Branch (487:7): [True: 1.34M, False: 91.8k] ------------------ 488| 1.34M| candidate.push_back(word[wn]); 489| 1.34M| map_related(word, candidate, wn + 1, wlst, cpdsuggest, 490| 1.34M| maptable, timer, timelimit, depth + 1, info); 491| 1.34M| } 492| 1.43M| return wlst.size(); 493| 1.44M|} _ZN10SuggestMgr9replcharsERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKS7_iRi: 499| 410k| int cpdsuggest, int& info) { 500| 410k| std::string candidate; 501| 410k| int wl = word.size(); 502| 410k| if (wl < 2 || !pAMgr) ------------------ | Branch (502:7): [True: 33.8k, False: 376k] | Branch (502:17): [True: 0, False: 376k] ------------------ 503| 33.8k| return wlst.size(); 504| 376k| const std::vector& reptable = pAMgr->get_reptable(); 505| 376k| auto replchars_start = std::chrono::steady_clock::now(); 506| 376k| for (const auto& entry : reptable) { ------------------ | Branch (506:26): [True: 77.7k, False: 376k] ------------------ 507| 77.7k| size_t r = 0; 508| | // search every occurence of the pattern in the word 509| 77.7k| while ((r = word.find(entry.pattern, r)) != std::string::npos) { ------------------ | Branch (509:12): [True: 0, False: 77.7k] ------------------ 510| 0| if (std::chrono::steady_clock::now() - replchars_start > TIMELIMIT_SUGGESTION_MS) ------------------ | | 105| 0|#define TIMELIMIT_SUGGESTION_MS std::chrono::milliseconds(100) ------------------ | Branch (510:11): [True: 0, False: 0] ------------------ 511| 0| return wlst.size(); 512| 0| int type = (r == 0) ? 1 : 0; ------------------ | Branch (512:18): [True: 0, False: 0] ------------------ 513| 0| if (r + entry.pattern.size() == word.size()) ------------------ | Branch (513:11): [True: 0, False: 0] ------------------ 514| 0| type += 2; 515| 0| while (type && entry.outstrings[type].empty()) ------------------ | Branch (515:14): [True: 0, False: 0] | Branch (515:22): [True: 0, False: 0] ------------------ 516| 0| type = (type == 2 && r != 0) ? 0 : type - 1; ------------------ | Branch (516:17): [True: 0, False: 0] | Branch (516:30): [True: 0, False: 0] ------------------ 517| 0| const std::string&out = entry.outstrings[type]; 518| 0| if (out.empty()) { ------------------ | Branch (518:11): [True: 0, False: 0] ------------------ 519| 0| ++r; 520| 0| continue; 521| 0| } 522| 0| candidate.assign(word, 0, r); 523| 0| candidate.append(entry.outstrings[type]); 524| 0| candidate.append(word, r + entry.pattern.size(), std::string::npos); 525| 0| size_t sp = candidate.find(' '); 526| 0| size_t oldns = wlst.size(); 527| 0| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 528| 0| if (oldns < wlst.size()) { ------------------ | Branch (528:11): [True: 0, False: 0] ------------------ 529| | // REP suggestions are the best, don't search other type of suggestions 530| 0| info |= SPELL_BEST_SUG; ------------------ | | 86| 0|#define SPELL_BEST_SUG (1 << 8) // limit suggestions for the best ones, i.e. ph: ------------------ 531| 0| } 532| | 533| | // check REP suggestions with space 534| 0| if (sp != std::string::npos) { ------------------ | Branch (534:11): [True: 0, False: 0] ------------------ 535| 0| size_t prev = 0; 536| 0| while (sp != std::string::npos) { ------------------ | Branch (536:16): [True: 0, False: 0] ------------------ 537| 0| std::string prev_chunk = candidate.substr(prev, sp - prev); 538| 0| if (checkword(prev_chunk, 0, nullptr, nullptr)) { ------------------ | Branch (538:15): [True: 0, False: 0] ------------------ 539| 0| oldns = wlst.size(); 540| 0| std::string post_chunk = candidate.substr(sp + 1); 541| 0| testsug(wlst, post_chunk, cpdsuggest, nullptr, nullptr, info); 542| 0| if (oldns < wlst.size()) { ------------------ | Branch (542:17): [True: 0, False: 0] ------------------ 543| 0| wlst[wlst.size() - 1] = candidate; 544| 0| } 545| 0| } 546| 0| prev = sp + 1; 547| 0| sp = candidate.find(' ', prev); 548| 0| } 549| 0| } 550| 0| r++; // search for the next letter 551| 0| } 552| 77.7k| } 553| 376k| return wlst.size(); 554| 376k|} _ZN10SuggestMgr14doubletwocharsERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKS7_iRi: 563| 255k| int cpdsuggest, int& info) { 564| 255k| size_t wl = word.size(); 565| 255k| if (wl < 5 || !pAMgr) ------------------ | Branch (565:7): [True: 60.4k, False: 194k] | Branch (565:17): [True: 0, False: 194k] ------------------ 566| 60.4k| return wlst.size(); 567| | 568| 194k| int state = 0; 569| 4.83M| for (size_t i = 2; i < wl; ++i) { ------------------ | Branch (569:22): [True: 4.63M, False: 194k] ------------------ 570| 4.63M| if (word[i] == word[i - 2]) { ------------------ | Branch (570:9): [True: 1.12M, False: 3.51M] ------------------ 571| 1.12M| state++; 572| 1.12M| if (state == 3 || (state == 2 && i >= 4)) { ------------------ | Branch (572:11): [True: 7.21k, False: 1.11M] | Branch (572:26): [True: 390k, False: 725k] | Branch (572:40): [True: 380k, False: 9.96k] ------------------ 573| 388k| auto word_iter = word.begin(); 574| 388k| std::string candidate(word_iter, word_iter + i - 1); 575| 388k| candidate.insert(candidate.end(), word_iter + i + 1, word.end()); 576| 388k| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 577| 388k| state = 0; 578| 388k| } 579| 3.51M| } else { 580| 3.51M| state = 0; 581| 3.51M| } 582| 4.63M| } 583| 194k| return wlst.size(); 584| 255k|} _ZN10SuggestMgr18doubletwochars_utfERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKNS1_I6w_charNS5_ISB_EEEEiRi: 593| 155k| int cpdsuggest, int& info) { 594| 155k| size_t wl = word.size(); 595| 155k| int state = 0; 596| 155k| if (wl < 5 || !pAMgr) ------------------ | Branch (596:7): [True: 57.0k, False: 98.0k] | Branch (596:17): [True: 0, False: 98.0k] ------------------ 597| 57.0k| return wlst.size(); 598| 1.58M| for (size_t i = 2; i < wl; ++i) { ------------------ | Branch (598:22): [True: 1.48M, False: 98.0k] ------------------ 599| 1.48M| if (word[i] == word[i - 2]) { ------------------ | Branch (599:9): [True: 445k, False: 1.03M] ------------------ 600| 445k| state++; 601| 445k| if (state == 3 || (state == 2 && i >= 4)) { ------------------ | Branch (601:11): [True: 3.97k, False: 441k] | Branch (601:26): [True: 154k, False: 286k] | Branch (601:40): [True: 148k, False: 5.77k] ------------------ 602| 152k| auto word_iter = word.begin(); 603| 152k| std::vector candidate_utf(word_iter, word_iter + i - 1); 604| 152k| candidate_utf.insert(candidate_utf.end(), word_iter + i + 1, word.end()); 605| 152k| std::string candidate; 606| 152k| u16_u8(candidate, candidate_utf); 607| 152k| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 608| 152k| state = 0; 609| 152k| } 610| 1.03M| } else { 611| 1.03M| state = 0; 612| 1.03M| } 613| 1.48M| } 614| 98.0k| return wlst.size(); 615| 155k|} _ZN10SuggestMgr10badcharkeyERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKS7_iRi: 621| 255k| int cpdsuggest, int& info) { 622| 255k| std::string candidate(word); 623| | 624| | // swap out each char one by one and try uppercase and neighbor 625| | // keyboard chars in its place to see if that makes a good word 626| 5.42M| for (size_t i = 0, wl = candidate.size(); i < wl; ++i) { ------------------ | Branch (626:45): [True: 5.17M, False: 255k] ------------------ 627| 5.17M| char tmpc = candidate[i]; 628| | // check with uppercase letters 629| 5.17M| candidate[i] = csconv[((unsigned char)tmpc)].cupper; 630| 5.17M| if (tmpc != candidate[i]) { ------------------ | Branch (630:9): [True: 1.21M, False: 3.95M] ------------------ 631| 1.21M| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 632| 1.21M| candidate[i] = tmpc; 633| 1.21M| } 634| | // check neighbor characters in keyboard string 635| 5.17M| if (ckey.empty()) ------------------ | Branch (635:9): [True: 0, False: 5.17M] ------------------ 636| 0| continue; 637| 5.17M| size_t loc = 0; 638| 137M| while ((loc < ckeyl) && ckey[loc] != tmpc) ------------------ | Branch (638:12): [True: 132M, False: 4.39M] | Branch (638:29): [True: 132M, False: 783k] ------------------ 639| 132M| ++loc; 640| 5.96M| while (loc < ckeyl) { ------------------ | Branch (640:12): [True: 794k, False: 5.17M] ------------------ 641| 794k| if ((loc > 0) && ckey[loc - 1] != '|') { ------------------ | Branch (641:11): [True: 632k, False: 162k] | Branch (641:24): [True: 600k, False: 31.5k] ------------------ 642| 600k| candidate[i] = ckey[loc - 1]; 643| 600k| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 644| 600k| } 645| 794k| if (((loc + 1) < ckeyl) && (ckey[loc + 1] != '|')) { ------------------ | Branch (645:11): [True: 737k, False: 56.7k] | Branch (645:34): [True: 646k, False: 91.6k] ------------------ 646| 646k| candidate[i] = ckey[loc + 1]; 647| 646k| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 648| 646k| } 649| 12.8M| do { 650| 12.8M| loc++; 651| 12.8M| } while ((loc < ckeyl) && ckey[loc] != tmpc); ------------------ | Branch (651:16): [True: 12.0M, False: 783k] | Branch (651:33): [True: 12.0M, False: 11.3k] ------------------ 652| 794k| } 653| 5.17M| candidate[i] = tmpc; 654| 5.17M| } 655| 255k| return wlst.size(); 656| 255k|} _ZN10SuggestMgr14badcharkey_utfERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKNS1_I6w_charNS5_ISB_EEEEiRi: 662| 155k| int cpdsuggest, int& info) { 663| 155k| std::string candidate; 664| 155k| std::vector candidate_utf(word); 665| | // swap out each char one by one and try all the tryme 666| | // chars in its place to see if that makes a good word 667| 1.98M| for (size_t i = 0, wl = word.size(); i < wl; ++i) { ------------------ | Branch (667:40): [True: 1.82M, False: 155k] ------------------ 668| 1.82M| w_char tmpc = candidate_utf[i]; 669| | // check with uppercase letters 670| 1.82M| candidate_utf[i] = upper_utf(candidate_utf[i], 1); 671| 1.82M| if (tmpc != candidate_utf[i]) { ------------------ | Branch (671:9): [True: 409k, False: 1.41M] ------------------ 672| 409k| u16_u8(candidate, candidate_utf); 673| 409k| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 674| 409k| candidate_utf[i] = tmpc; 675| 409k| } 676| | // check neighbor characters in keyboard string 677| 1.82M| if (ckey_utf.empty()) ------------------ | Branch (677:9): [True: 0, False: 1.82M] ------------------ 678| 0| continue; 679| 1.82M| size_t loc = 0; 680| 46.1M| while ((loc < ckeyl) && ckey_utf[loc] != tmpc) ------------------ | Branch (680:12): [True: 44.6M, False: 1.41M] | Branch (680:29): [True: 44.2M, False: 412k] ------------------ 681| 44.2M| ++loc; 682| 2.24M| while (loc < ckeyl) { ------------------ | Branch (682:12): [True: 419k, False: 1.82M] ------------------ 683| 419k| if ((loc > 0) && ckey_utf[loc - 1] != W_VLINE) { ------------------ | Branch (683:11): [True: 313k, False: 105k] | Branch (683:24): [True: 313k, False: 0] ------------------ 684| 313k| candidate_utf[i] = ckey_utf[loc - 1]; 685| 313k| u16_u8(candidate, candidate_utf); 686| 313k| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 687| 313k| } 688| 419k| if (((loc + 1) < ckeyl) && (ckey_utf[loc + 1] != W_VLINE)) { ------------------ | Branch (688:11): [True: 383k, False: 35.4k] | Branch (688:34): [True: 383k, False: 0] ------------------ 689| 383k| candidate_utf[i] = ckey_utf[loc + 1]; 690| 383k| u16_u8(candidate, candidate_utf); 691| 383k| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 692| 383k| } 693| 6.91M| do { 694| 6.91M| loc++; 695| 6.91M| } while ((loc < ckeyl) && ckey_utf[loc] != tmpc); ------------------ | Branch (695:16): [True: 6.50M, False: 412k] | Branch (695:33): [True: 6.50M, False: 6.02k] ------------------ 696| 419k| } 697| 1.82M| candidate_utf[i] = tmpc; 698| 1.82M| } 699| 155k| return wlst.size(); 700| 155k|} _ZN10SuggestMgr7badcharERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKS7_iRi: 705| 255k| int cpdsuggest, int& info) { 706| 255k| std::string candidate(word); 707| 255k| std::chrono::steady_clock::time_point timelimit = std::chrono::steady_clock::now(); 708| 255k| int timer = MINTIMER; ------------------ | | 107| 255k|#define MINTIMER 100 ------------------ 709| | // swap out each char one by one and try all the tryme 710| | // chars in its place to see if that makes a good word 711| 255k| for (size_t j = 0; j < ctryl; ++j) { ------------------ | Branch (711:22): [True: 0, False: 255k] ------------------ 712| 0| for (auto aI = candidate.rbegin(), aEnd = candidate.rend(); aI != aEnd; ++aI) { ------------------ | Branch (712:65): [True: 0, False: 0] ------------------ 713| 0| char tmpc = *aI; 714| 0| if (ctry[j] == tmpc) ------------------ | Branch (714:11): [True: 0, False: 0] ------------------ 715| 0| continue; 716| 0| *aI = ctry[j]; 717| 0| testsug(wlst, candidate, cpdsuggest, &timer, &timelimit, info); 718| 0| if (!timer) ------------------ | Branch (718:11): [True: 0, False: 0] ------------------ 719| 0| return wlst.size(); 720| 0| *aI = tmpc; 721| 0| } 722| 0| } 723| 255k| return wlst.size(); 724| 255k|} _ZN10SuggestMgr11badchar_utfERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKNS1_I6w_charNS5_ISB_EEEEiRi: 729| 155k| int cpdsuggest, int& info) { 730| 155k| std::vector candidate_utf(word); 731| 155k| std::string candidate; 732| 155k| std::chrono::steady_clock::time_point timelimit = std::chrono::steady_clock::now(); 733| 155k| int timer = MINTIMER; ------------------ | | 107| 155k|#define MINTIMER 100 ------------------ 734| | // swap out each char one by one and try all the tryme 735| | // chars in its place to see if that makes a good word 736| 173k| for (size_t j = 0; j < ctryl; ++j) { ------------------ | Branch (736:22): [True: 18.4k, False: 155k] ------------------ 737| 237k| for (auto aI = candidate_utf.rbegin(), aEnd = candidate_utf.rend(); aI != aEnd; ++aI) { ------------------ | Branch (737:73): [True: 218k, False: 18.4k] ------------------ 738| 218k| w_char tmpc = *aI; 739| 218k| if (tmpc == ctry_utf[j]) ------------------ | Branch (739:11): [True: 0, False: 218k] ------------------ 740| 0| continue; 741| 218k| *aI = ctry_utf[j]; 742| 218k| u16_u8(candidate, candidate_utf); 743| 218k| testsug(wlst, candidate, cpdsuggest, &timer, &timelimit, info); 744| 218k| if (!timer) ------------------ | Branch (744:11): [True: 0, False: 218k] ------------------ 745| 0| return wlst.size(); 746| 218k| *aI = tmpc; 747| 218k| } 748| 18.4k| } 749| 155k| return wlst.size(); 750| 155k|} _ZN10SuggestMgr13extrachar_utfERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKNS1_I6w_charNS5_ISB_EEEEiRi: 755| 155k| int cpdsuggest, int& info) { 756| 155k| std::vector candidate_utf(word); 757| 155k| if (candidate_utf.size() < 2) ------------------ | Branch (757:7): [True: 15.5k, False: 139k] ------------------ 758| 15.5k| return wlst.size(); 759| | // try omitting one char of word at a time 760| 1.95M| for (size_t i = 0; i < candidate_utf.size(); ++i) { ------------------ | Branch (760:22): [True: 1.81M, False: 139k] ------------------ 761| 1.81M| size_t index = candidate_utf.size() - 1 - i; 762| 1.81M| w_char tmpc = candidate_utf[index]; 763| 1.81M| candidate_utf.erase(candidate_utf.begin() + index); 764| 1.81M| std::string candidate; 765| 1.81M| u16_u8(candidate, candidate_utf); 766| 1.81M| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 767| 1.81M| candidate_utf.insert(candidate_utf.begin() + index, tmpc); 768| 1.81M| } 769| 139k| return wlst.size(); 770| 155k|} _ZN10SuggestMgr9extracharERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKS7_iRi: 775| 255k| int cpdsuggest, int& info) { 776| 255k| std::string candidate(word); 777| 255k| if (candidate.size() < 2) ------------------ | Branch (777:7): [True: 18.7k, False: 236k] ------------------ 778| 18.7k| return wlst.size(); 779| | // try omitting one char of word at a time 780| 5.39M| for (size_t i = 0; i < candidate.size(); ++i) { ------------------ | Branch (780:22): [True: 5.15M, False: 236k] ------------------ 781| 5.15M| size_t index = candidate.size() - 1 - i; 782| 5.15M| char tmpc = candidate[index]; 783| 5.15M| candidate.erase(candidate.begin() + index); 784| 5.15M| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 785| 5.15M| candidate.insert(candidate.begin() + index, tmpc); 786| 5.15M| } 787| 236k| return wlst.size(); 788| 255k|} _ZN10SuggestMgr10forgotcharERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKS7_iRi: 793| 255k| int cpdsuggest, int& info) { 794| 255k| std::string candidate(word); 795| 255k| std::chrono::steady_clock::time_point timelimit = std::chrono::steady_clock::now(); 796| 255k| int timer = MINTIMER; ------------------ | | 107| 255k|#define MINTIMER 100 ------------------ 797| | 798| | // try inserting a tryme character before every letter (and the null 799| | // terminator) 800| 255k| for (size_t k = 0; k < ctryl; ++k) { ------------------ | Branch (800:22): [True: 0, False: 255k] ------------------ 801| 0| for (size_t i = 0; i <= candidate.size(); ++i) { ------------------ | Branch (801:24): [True: 0, False: 0] ------------------ 802| 0| size_t index = candidate.size() - i; 803| 0| candidate.insert(candidate.begin() + index, ctry[k]); 804| 0| testsug(wlst, candidate, cpdsuggest, &timer, &timelimit, info); 805| 0| if (!timer) ------------------ | Branch (805:11): [True: 0, False: 0] ------------------ 806| 0| return wlst.size(); 807| 0| candidate.erase(candidate.begin() + index); 808| 0| } 809| 0| } 810| 255k| return wlst.size(); 811| 255k|} _ZN10SuggestMgr14forgotchar_utfERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKNS1_I6w_charNS5_ISB_EEEEiRi: 816| 155k| int cpdsuggest, int& info) { 817| 155k| std::vector candidate_utf(word); 818| 155k| std::chrono::steady_clock::time_point timelimit = std::chrono::steady_clock::now(); 819| 155k| int timer = MINTIMER; ------------------ | | 107| 155k|#define MINTIMER 100 ------------------ 820| | 821| | // try inserting a tryme character at the end of the word and before every 822| | // letter 823| 173k| for (size_t k = 0; k < ctryl; ++k) { ------------------ | Branch (823:22): [True: 18.4k, False: 155k] ------------------ 824| 255k| for (size_t i = 0; i <= candidate_utf.size(); ++i) { ------------------ | Branch (824:24): [True: 237k, False: 18.4k] ------------------ 825| 237k| size_t index = candidate_utf.size() - i; 826| 237k| candidate_utf.insert(candidate_utf.begin() + index, ctry_utf[k]); 827| 237k| std::string candidate; 828| 237k| u16_u8(candidate, candidate_utf); 829| 237k| testsug(wlst, candidate, cpdsuggest, &timer, &timelimit, info); 830| 237k| if (!timer) ------------------ | Branch (830:11): [True: 0, False: 237k] ------------------ 831| 0| return wlst.size(); 832| 237k| candidate_utf.erase(candidate_utf.begin() + index); 833| 237k| } 834| 18.4k| } 835| 155k| return wlst.size(); 836| 155k|} _ZN10SuggestMgr8twowordsERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKS7_ibRi: 846| 410k| bool good, int& info) { 847| 410k| int c2, forbidden = 0, cwrd, wl = word.size(); 848| 410k| if (wl < 3) ------------------ | Branch (848:7): [True: 57.0k, False: 353k] ------------------ 849| 57.0k| return false; 850| | 851| 353k| if (langnum == LANG_hu) ------------------ | Branch (851:7): [True: 14.4k, False: 338k] ------------------ 852| 14.4k| forbidden = check_forbidden(word); 853| | 854| 353k| char* candidate = new char[wl + 2]; 855| 353k| memcpy(candidate + 1, word.data(), wl); 856| 353k| candidate[wl + 1] = 0; 857| | 858| | // split the string into two pieces after every char 859| | // if both pieces are good words make them a suggestion 860| 4.42M| for (char* p = candidate + 1; p[1] != '\0'; p++) { ------------------ | Branch (860:33): [True: 4.10M, False: 317k] ------------------ 861| 4.10M| p[-1] = *p; 862| | // go to end of the UTF-8 character 863| 4.36M| while (utf8 && is_utf8_cont(p[1])) { ------------------ | Branch (863:12): [True: 1.39M, False: 2.96M] | Branch (863:20): [True: 254k, False: 1.14M] ------------------ 864| 254k| *p = p[1]; 865| 254k| p++; 866| 254k| } 867| 4.10M| if (utf8 && p[1] == '\0') ------------------ | Branch (867:9): [True: 1.14M, False: 2.96M] | Branch (867:17): [True: 35.6k, False: 1.10M] ------------------ 868| 35.6k| break; // last UTF-8 character 869| | 870| | // Suggest only word pairs, if they are listed in the dictionary. 871| | // For example, adding "a lot" to the English dic file will 872| | // result only "alot" -> "a lot" suggestion instead of 873| | // "alto, slot, alt, lot, allot, aloft, aloe, clot, plot, blot, a lot". 874| | // Note: using "ph:alot" keeps the other suggestions: 875| | // a lot ph:alot 876| | // alot -> a lot, alto, slot... 877| 4.07M| *p = ' '; 878| 4.07M| if (!cpdsuggest && checkword(candidate, cpdsuggest, nullptr, nullptr)) { ------------------ | Branch (878:9): [True: 1.35M, False: 2.71M] | Branch (878:9): [True: 1.56k, False: 4.07M] | Branch (878:24): [True: 1.56k, False: 1.35M] ------------------ 879| | // best solution 880| 1.56k| info |= SPELL_BEST_SUG; ------------------ | | 86| 1.56k|#define SPELL_BEST_SUG (1 << 8) // limit suggestions for the best ones, i.e. ph: ------------------ 881| | 882| | // remove not word pair suggestions 883| 1.56k| if (!good) { ------------------ | Branch (883:11): [True: 152, False: 1.41k] ------------------ 884| 152| good = true; 885| 152| wlst.clear(); 886| 152| } 887| | 888| 1.56k| wlst.insert(wlst.begin(), candidate); 889| 1.56k| } 890| | 891| | // word pairs with dash? 892| 4.07M| if (lang_with_dash_usage) { ------------------ | Branch (892:9): [True: 0, False: 4.07M] ------------------ 893| 0| *p = '-'; 894| | 895| 0| if (!cpdsuggest && checkword(candidate, cpdsuggest, nullptr, nullptr)) { ------------------ | Branch (895:11): [True: 0, False: 0] | Branch (895:11): [True: 0, False: 0] | Branch (895:26): [True: 0, False: 0] ------------------ 896| | // best solution 897| 0| info |= SPELL_BEST_SUG; ------------------ | | 86| 0|#define SPELL_BEST_SUG (1 << 8) // limit suggestions for the best ones, i.e. ph: ------------------ 898| | 899| | // remove not word pair suggestions 900| 0| if (!good) { ------------------ | Branch (900:13): [True: 0, False: 0] ------------------ 901| 0| good = true; 902| 0| wlst.clear(); 903| 0| } 904| 0| wlst.insert(wlst.begin(), candidate); 905| 0| } 906| 0| } 907| | 908| 4.07M| if (wlst.size() < maxSug && !nosplitsugs && !good) { ------------------ | Branch (908:9): [True: 4.07M, False: 598] | Branch (908:33): [True: 4.07M, False: 0] | Branch (908:49): [True: 4.07M, False: 967] ------------------ 909| 4.07M| *p = '\0'; 910| 4.07M| int c1 = checkword(candidate, cpdsuggest, nullptr, nullptr); 911| 4.07M| if (c1) { ------------------ | Branch (911:11): [True: 19.4k, False: 4.05M] ------------------ 912| 19.4k| c2 = checkword((p + 1), cpdsuggest, nullptr, nullptr); 913| 19.4k| if (c2) { ------------------ | Branch (913:13): [True: 678, False: 18.7k] ------------------ 914| | // spec. Hungarian code (TODO need a better compound word support) 915| 678| if ((langnum == LANG_hu) && !forbidden && ------------------ | Branch (915:15): [True: 15, False: 663] | Branch (915:39): [True: 15, False: 0] ------------------ 916| | // if 3 repeating letter, use - instead of space 917| 15| (((p[-1] == p[1]) && ------------------ | Branch (917:17): [True: 3, False: 12] ------------------ 918| 3| (((p > candidate + 1) && (p[-1] == p[-2])) || (p[-1] == p[2]))) || ------------------ | Branch (918:17): [True: 1, False: 2] | Branch (918:40): [True: 0, False: 1] | Branch (918:61): [True: 0, False: 3] ------------------ 919| | // or multiple compounding, with more, than 6 syllables 920| 15| ((c1 == 3) && (c2 >= 2)))) ------------------ | Branch (920:16): [True: 0, False: 15] | Branch (920:29): [True: 0, False: 0] ------------------ 921| 0| *p = '-'; 922| 678| else 923| 678| *p = ' '; 924| | 925| 678| cwrd = std::find(wlst.begin(), wlst.end(), candidate) != wlst.end() ? 0 : 1; ------------------ | Branch (925:18): [True: 16, False: 662] ------------------ 926| | 927| 678| if (cwrd && (wlst.size() < maxSug)) ------------------ | Branch (927:15): [True: 662, False: 16] | Branch (927:23): [True: 662, False: 0] ------------------ 928| 662| wlst.emplace_back(candidate); 929| | 930| | // add two word suggestion with dash, depending on the language 931| | // Note that cwrd doesn't modified for REP twoword sugg. 932| 678| if ( !nosplitsugs && lang_with_dash_usage && ------------------ | Branch (932:16): [True: 678, False: 0] | Branch (932:32): [True: 0, False: 678] ------------------ 933| 0| mystrlen(p + 1) > 1 && mystrlen(candidate) - mystrlen(p) > 1) { ------------------ | Branch (933:15): [True: 0, False: 0] | Branch (933:38): [True: 0, False: 0] ------------------ 934| 0| *p = '-'; 935| 0| for (auto& k : wlst) { ------------------ | Branch (935:26): [True: 0, False: 0] ------------------ 936| 0| if (k == candidate) { ------------------ | Branch (936:19): [True: 0, False: 0] ------------------ 937| 0| cwrd = 0; 938| 0| break; 939| 0| } 940| 0| } 941| | 942| 0| if ((wlst.size() < maxSug) && cwrd) ------------------ | Branch (942:17): [True: 0, False: 0] | Branch (942:43): [True: 0, False: 0] ------------------ 943| 0| wlst.emplace_back(candidate); 944| 0| } 945| 678| } 946| 19.4k| } 947| 4.07M| } 948| 4.07M| } 949| 353k| delete[] candidate; 950| 353k| return good; 951| 410k|} _ZN10SuggestMgr8swapcharERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKS7_iRi: 956| 255k| int cpdsuggest, int& info) { 957| 255k| if (word.size() < 2) ------------------ | Branch (957:7): [True: 18.7k, False: 236k] ------------------ 958| 18.7k| return wlst.size(); 959| | 960| 236k| std::string candidate(word); 961| | 962| | // try swapping adjacent chars one by one 963| 5.15M| for (size_t i = 0; i < candidate.size() - 1; ++i) { ------------------ | Branch (963:22): [True: 4.91M, False: 236k] ------------------ 964| 4.91M| std::swap(candidate[i], candidate[i+1]); 965| 4.91M| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 966| 4.91M| std::swap(candidate[i], candidate[i+1]); 967| 4.91M| } 968| | 969| | // try double swaps for short words 970| | // ahev -> have, owudl -> would 971| 236k| if (candidate.size() == 4 || candidate.size() == 5) { ------------------ | Branch (971:7): [True: 16.3k, False: 220k] | Branch (971:32): [True: 10.7k, False: 209k] ------------------ 972| 27.0k| candidate[0] = word[1]; 973| 27.0k| candidate[1] = word[0]; 974| 27.0k| candidate[2] = word[2]; 975| 27.0k| candidate[candidate.size() - 2] = word[candidate.size() - 1]; 976| 27.0k| candidate[candidate.size() - 1] = word[candidate.size() - 2]; 977| 27.0k| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 978| 27.0k| if (candidate.size() == 5) { ------------------ | Branch (978:9): [True: 10.7k, False: 16.3k] ------------------ 979| 10.7k| candidate[0] = word[0]; 980| 10.7k| candidate[1] = word[2]; 981| 10.7k| candidate[2] = word[1]; 982| 10.7k| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 983| 10.7k| } 984| 27.0k| } 985| | 986| 236k| return wlst.size(); 987| 255k|} _ZN10SuggestMgr12swapchar_utfERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKNS1_I6w_charNS5_ISB_EEEEiRi: 992| 155k| int cpdsuggest, int& info) { 993| 155k| if (word.size() < 2) ------------------ | Branch (993:7): [True: 15.5k, False: 139k] ------------------ 994| 15.5k| return wlst.size(); 995| | 996| 139k| std::vector candidate_utf(word); 997| | 998| 139k| std::string candidate; 999| | // try swapping adjacent chars one by one 1000| 1.81M| for (size_t i = 0; i < candidate_utf.size() - 1; ++i) { ------------------ | Branch (1000:22): [True: 1.67M, False: 139k] ------------------ 1001| 1.67M| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_SUGGESTION_MS) ------------------ | | 105| 1.67M|#define TIMELIMIT_SUGGESTION_MS std::chrono::milliseconds(100) ------------------ | Branch (1001:9): [True: 0, False: 1.67M] ------------------ 1002| 0| return wlst.size(); 1003| 1.67M| std::swap(candidate_utf[i], candidate_utf[i+1]); 1004| 1.67M| u16_u8(candidate, candidate_utf); 1005| 1.67M| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 1006| 1.67M| std::swap(candidate_utf[i], candidate_utf[i+1]); 1007| 1.67M| } 1008| | 1009| | // try double swaps for short words 1010| | // ahev -> have, owudl -> would, suodn -> sound 1011| 139k| if (candidate_utf.size() == 4 || candidate_utf.size() == 5) { ------------------ | Branch (1011:7): [True: 19.4k, False: 120k] | Branch (1011:36): [True: 9.99k, False: 110k] ------------------ 1012| 29.4k| candidate_utf[0] = word[1]; 1013| 29.4k| candidate_utf[1] = word[0]; 1014| 29.4k| candidate_utf[2] = word[2]; 1015| 29.4k| candidate_utf[candidate_utf.size() - 2] = word[candidate_utf.size() - 1]; 1016| 29.4k| candidate_utf[candidate_utf.size() - 1] = word[candidate_utf.size() - 2]; 1017| 29.4k| u16_u8(candidate, candidate_utf); 1018| 29.4k| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 1019| 29.4k| if (candidate_utf.size() == 5) { ------------------ | Branch (1019:9): [True: 9.99k, False: 19.4k] ------------------ 1020| 9.99k| candidate_utf[0] = word[0]; 1021| 9.99k| candidate_utf[1] = word[2]; 1022| 9.99k| candidate_utf[2] = word[1]; 1023| 9.99k| u16_u8(candidate, candidate_utf); 1024| 9.99k| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 1025| 9.99k| } 1026| 29.4k| } 1027| 139k| return wlst.size(); 1028| 139k|} _ZN10SuggestMgr12longswapcharERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKS7_iRi: 1033| 255k| int cpdsuggest, int& info) { 1034| 255k| std::string candidate(word); 1035| | // try swapping not adjacent chars one by one 1036| 5.42M| for (auto p = candidate.begin(); p < candidate.end(); ++p) { ------------------ | Branch (1036:36): [True: 5.17M, False: 255k] ------------------ 1037| 190M| for (auto q = candidate.begin(); q < candidate.end(); ++q) { ------------------ | Branch (1037:38): [True: 185M, False: 5.17M] ------------------ 1038| 185M| const auto distance = std::abs(std::distance(q, p)); 1039| 185M| if (distance > 1 && distance <= MAX_CHAR_DISTANCE && *p != *q) { ------------------ | | 84| 356M|#define MAX_CHAR_DISTANCE 4 ------------------ | Branch (1039:11): [True: 170M, False: 15.0M] | Branch (1039:27): [True: 26.7M, False: 143M] | Branch (1039:60): [True: 20.6M, False: 6.17M] ------------------ 1040| 20.6M| std::swap(*p, *q); 1041| 20.6M| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 1042| 20.6M| std::swap(*p, *q); 1043| 20.6M| } 1044| 185M| } 1045| 5.17M| } 1046| 255k| return wlst.size(); 1047| 255k|} _ZN10SuggestMgr16longswapchar_utfERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKNS1_I6w_charNS5_ISB_EEEEiRi: 1052| 155k| int cpdsuggest, int& info) { 1053| 155k| std::vector candidate_utf(word); 1054| | // try swapping not adjacent chars 1055| 1.98M| for (auto p = candidate_utf.begin(); p < candidate_utf.end(); ++p) { ------------------ | Branch (1055:40): [True: 1.82M, False: 155k] ------------------ 1056| 49.0M| for (auto q = candidate_utf.begin(); q < candidate_utf.end(); ++q) { ------------------ | Branch (1056:42): [True: 47.2M, False: 1.82M] ------------------ 1057| 47.2M| const auto distance = std::abs(std::distance(q, p)); 1058| 47.2M| if (distance > 1 && distance <= MAX_CHAR_DISTANCE && *p != *q) { ------------------ | | 84| 89.2M|#define MAX_CHAR_DISTANCE 4 ------------------ | Branch (1058:11): [True: 42.0M, False: 5.17M] | Branch (1058:27): [True: 8.45M, False: 33.5M] | Branch (1058:60): [True: 6.04M, False: 2.40M] ------------------ 1059| 6.04M| std::swap(*p, *q); 1060| 6.04M| std::string candidate; 1061| 6.04M| u16_u8(candidate, candidate_utf); 1062| 6.04M| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 1063| 6.04M| std::swap(*p, *q); 1064| 6.04M| } 1065| 47.2M| } 1066| 1.82M| } 1067| 155k| return wlst.size(); 1068| 155k|} _ZN10SuggestMgr8movecharERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKS7_iRi: 1073| 255k| int cpdsuggest, int& info) { 1074| 255k| if (word.size() < 2) ------------------ | Branch (1074:7): [True: 18.7k, False: 236k] ------------------ 1075| 18.7k| return wlst.size(); 1076| | 1077| 236k| std::string candidate(word); 1078| | 1079| | // try moving a char 1080| 5.39M| for (auto p = candidate.begin(); p < candidate.end(); ++p) { ------------------ | Branch (1080:36): [True: 5.15M, False: 236k] ------------------ 1081| 23.4M| for (auto q = p + 1; q < candidate.end() && std::distance(p, q) <= MAX_CHAR_DISTANCE; ++q) { ------------------ | | 84| 22.3M|#define MAX_CHAR_DISTANCE 4 ------------------ | Branch (1081:26): [True: 22.3M, False: 1.10M] | Branch (1081:26): [True: 18.3M, False: 5.15M] | Branch (1081:49): [True: 18.3M, False: 4.05M] ------------------ 1082| 18.3M| std::swap(*q, *(q - 1)); 1083| 18.3M| if (std::distance(p, q) < 2) ------------------ | Branch (1083:11): [True: 4.91M, False: 13.3M] ------------------ 1084| 4.91M| continue; // omit swap char 1085| 13.3M| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 1086| 13.3M| } 1087| 5.15M| const auto word_iter = word.begin(); 1088| 5.15M| std::copy_n(word_iter, candidate.size(), candidate.begin()); 1089| 5.15M| } 1090| | 1091| 5.15M| for (auto p = candidate.rbegin(), pEnd = candidate.rend() - 1; p != pEnd; ++p) { ------------------ | Branch (1091:66): [True: 4.91M, False: 236k] ------------------ 1092| 23.2M| for (auto q = p + 1, qEnd = candidate.rend(); q != qEnd && std::distance(p, q) <= MAX_CHAR_DISTANCE; ++q) { ------------------ | | 84| 22.3M|#define MAX_CHAR_DISTANCE 4 ------------------ | Branch (1092:51): [True: 22.3M, False: 865k] | Branch (1092:64): [True: 18.3M, False: 4.05M] ------------------ 1093| 18.3M| std::swap(*q, *(q - 1)); 1094| 18.3M| if (std::distance(p, q) < 2) ------------------ | Branch (1094:11): [True: 4.91M, False: 13.3M] ------------------ 1095| 4.91M| continue; // omit swap char 1096| 13.3M| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 1097| 13.3M| } 1098| 4.91M| const auto word_iter = word.begin(); 1099| 4.91M| std::copy_n(word_iter, candidate.size(), candidate.begin()); 1100| 4.91M| } 1101| | 1102| 236k| return wlst.size(); 1103| 255k|} _ZN10SuggestMgr12movechar_utfERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKNS1_I6w_charNS5_ISB_EEEEiRi: 1108| 155k| int cpdsuggest, int& info) { 1109| 155k| if (word.size() < 2) ------------------ | Branch (1109:7): [True: 15.5k, False: 139k] ------------------ 1110| 15.5k| return wlst.size(); 1111| | 1112| 139k| std::vector candidate_utf(word); 1113| | 1114| | // try moving a char 1115| 1.95M| for (auto p = candidate_utf.begin(); p < candidate_utf.end(); ++p) { ------------------ | Branch (1115:40): [True: 1.81M, False: 139k] ------------------ 1116| 7.71M| for (auto q = p + 1; q < candidate_utf.end() && std::distance(p, q) <= MAX_CHAR_DISTANCE; ++q) { ------------------ | | 84| 7.09M|#define MAX_CHAR_DISTANCE 4 ------------------ | Branch (1116:26): [True: 7.09M, False: 622k] | Branch (1116:26): [True: 5.90M, False: 1.81M] | Branch (1116:53): [True: 5.90M, False: 1.19M] ------------------ 1117| 5.90M| std::swap(*q, *(q - 1)); 1118| 5.90M| if (std::distance(p, q) < 2) ------------------ | Branch (1118:11): [True: 1.67M, False: 4.22M] ------------------ 1119| 1.67M| continue; // omit swap char 1120| 4.22M| std::string candidate; 1121| 4.22M| u16_u8(candidate, candidate_utf); 1122| 4.22M| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 1123| 4.22M| } 1124| 1.81M| const auto word_iter = word.begin(); 1125| 1.81M| std::copy_n(word_iter, candidate_utf.size(), candidate_utf.begin()); 1126| 1.81M| } 1127| | 1128| 1.95M| for (auto p = candidate_utf.rbegin(); p < candidate_utf.rend(); ++p) { ------------------ | Branch (1128:41): [True: 1.81M, False: 139k] ------------------ 1129| 7.71M| for (auto q = p + 1; q < candidate_utf.rend() && std::distance(p, q) <= MAX_CHAR_DISTANCE; ++q) { ------------------ | | 84| 7.09M|#define MAX_CHAR_DISTANCE 4 ------------------ | Branch (1129:26): [True: 7.09M, False: 622k] | Branch (1129:26): [True: 5.90M, False: 1.81M] | Branch (1129:54): [True: 5.90M, False: 1.19M] ------------------ 1130| 5.90M| std::swap(*q, *(q - 1)); 1131| 5.90M| if (std::distance(p, q) < 2) ------------------ | Branch (1131:11): [True: 1.67M, False: 4.22M] ------------------ 1132| 1.67M| continue; // omit swap char 1133| 4.22M| std::string candidate; 1134| 4.22M| u16_u8(candidate, candidate_utf); 1135| 4.22M| testsug(wlst, candidate, cpdsuggest, nullptr, nullptr, info); 1136| 4.22M| } 1137| 1.81M| const auto word_iter = word.begin(); 1138| 1.81M| std::copy_n(word_iter, candidate_utf.size(), candidate_utf.begin()); 1139| 1.81M| } 1140| | 1141| 139k| return wlst.size(); 1142| 155k|} _ZN10SuggestMgr9ngsuggestERNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEEPKcRKNS1_INS0_10unique_ptrI7HashMgrNS0_14default_deleteISE_EEEENS5_ISH_EEEEi: 1177| 62.3k| int captype) { 1178| 62.3k| int lval, sc, lp, lpphon, nonbmp = 0; 1179| | 1180| | // exhaustively search through all root words 1181| | // keeping track of the MAX_ROOTS most similar root words 1182| 62.3k| struct hentry* roots[MAX_ROOTS]; 1183| 62.3k| char* rootsphon[MAX_ROOTS]; 1184| 62.3k| int scores[MAX_ROOTS]; 1185| 62.3k| int scoresphon[MAX_ROOTS]; 1186| 6.29M| for (int i = 0; i < MAX_ROOTS; i++) { ------------------ | | 74| 6.29M|#define MAX_ROOTS 100 ------------------ | Branch (1186:19): [True: 6.23M, False: 62.3k] ------------------ 1187| 6.23M| roots[i] = nullptr; 1188| 6.23M| scores[i] = -100 * i; 1189| 6.23M| rootsphon[i] = nullptr; 1190| 6.23M| scoresphon[i] = -100 * i; 1191| 6.23M| } 1192| 62.3k| bool has_roots = false; 1193| 62.3k| bool has_rootsphon = false; 1194| 62.3k| lp = MAX_ROOTS - 1; ------------------ | | 74| 62.3k|#define MAX_ROOTS 100 ------------------ 1195| 62.3k| lpphon = MAX_ROOTS - 1; ------------------ | | 74| 62.3k|#define MAX_ROOTS 100 ------------------ 1196| 62.3k| int low = NGRAM_LOWERING; ------------------ | | 83| 62.3k|#define NGRAM_LOWERING (1 << 2) ------------------ 1197| | 1198| 62.3k| std::string w2; 1199| 62.3k| const char* word = w; 1200| | 1201| 62.3k| int nc; 1202| | // word reversing wrapper for complex prefixes 1203| 62.3k| if (complexprefixes) { ------------------ | Branch (1203:7): [True: 3.57k, False: 58.7k] ------------------ 1204| 3.57k| w2.assign(w); 1205| 3.57k| if (utf8) ------------------ | Branch (1205:9): [True: 0, False: 3.57k] ------------------ 1206| 0| reverseword_utf(w2); 1207| 3.57k| else 1208| 3.57k| reverseword(w2); 1209| 3.57k| word = w2.c_str(); 1210| 3.57k| nc = (int)w2.size(); 1211| 58.7k| } else { 1212| 58.7k| nc = (int)strlen(word); 1213| 58.7k| } 1214| | 1215| 62.3k| std::vector u8; 1216| 62.3k| int n = (utf8) ? u8_u16(u8, word) : nc; ------------------ | Branch (1216:11): [True: 26.0k, False: 36.3k] ------------------ 1217| | 1218| | // set character based ngram suggestion for words with non-BMP Unicode 1219| | // characters 1220| 62.3k| const struct cs_info* origconv = csconv; 1221| 62.3k| if (n == -1) { ------------------ | Branch (1221:7): [True: 1.75k, False: 60.5k] ------------------ 1222| 1.75k| utf8 = 0; // XXX not state-free 1223| 1.75k| if (!csconv) ------------------ | Branch (1223:9): [True: 1.75k, False: 0] ------------------ 1224| 1.75k| csconv = get_current_cs("iso88591"); // XXX not state-free 1225| 1.75k| n = nc; 1226| 1.75k| nonbmp = 1; 1227| 1.75k| low = 0; 1228| 1.75k| } 1229| 62.3k| ngsuggest_guard restore_state(nonbmp, origconv, &utf8, &csconv); 1230| 62.3k| int max_word_len = (utf8) ? MAXWORDUTF8LEN : MAXWORDLEN; ------------------ | | 85| 24.2k|#define MAXWORDUTF8LEN (MAXWORDLEN * 3) | | ------------------ | | | | 89| 24.2k|#define MAXWORDLEN 100 | | ------------------ ------------------ int max_word_len = (utf8) ? MAXWORDUTF8LEN : MAXWORDLEN; ------------------ | | 89| 38.0k|#define MAXWORDLEN 100 ------------------ | Branch (1230:22): [True: 24.2k, False: 38.0k] ------------------ 1231| | // ofz#59067 a replist entry can generate a very long word, abandon 1232| | // ngram if that odd-edge case arises 1233| 62.3k| if (n > max_word_len * 4) ------------------ | Branch (1233:7): [True: 0, False: 62.3k] ------------------ 1234| 0| return; 1235| | 1236| 62.3k| struct hentry* hp = nullptr; 1237| 62.3k| int col = -1; 1238| 62.3k| phonetable* ph = (pAMgr) ? pAMgr->get_phonetable() : nullptr; ------------------ | Branch (1238:20): [True: 62.3k, False: 0] ------------------ 1239| 62.3k| std::string target; 1240| 62.3k| std::string candidate; 1241| 62.3k| std::vector w_candidate; 1242| 62.3k| if (ph) { ------------------ | Branch (1242:7): [True: 3.54k, False: 58.7k] ------------------ 1243| 3.54k| if (utf8) { ------------------ | Branch (1243:9): [True: 0, False: 3.54k] ------------------ 1244| 0| u8_u16(w_candidate, word); 1245| 0| mkallcap_utf(w_candidate, langnum); 1246| 0| u16_u8(candidate, w_candidate); 1247| 3.54k| } else { 1248| 3.54k| candidate.assign(word); 1249| 3.54k| if (!nonbmp) ------------------ | Branch (1249:11): [True: 3.54k, False: 0] ------------------ 1250| 3.54k| mkallcap(candidate, csconv); 1251| 3.54k| } 1252| 3.54k| target = phonet(candidate, *ph); // XXX phonet() is 8-bit (nc, not n) 1253| 3.54k| } 1254| | 1255| 62.3k| FLAG forbiddenword = pAMgr ? pAMgr->get_forbiddenword() : FLAG_NULL; ------------------ | | 95| 62.3k|#define FLAG unsigned short ------------------ FLAG forbiddenword = pAMgr ? pAMgr->get_forbiddenword() : FLAG_NULL; ------------------ | | 96| 62.3k|#define FLAG_NULL 0x00 ------------------ | Branch (1255:24): [True: 62.3k, False: 0] ------------------ 1256| 62.3k| FLAG nosuggest = pAMgr ? pAMgr->get_nosuggest() : FLAG_NULL; ------------------ | | 95| 62.3k|#define FLAG unsigned short ------------------ FLAG nosuggest = pAMgr ? pAMgr->get_nosuggest() : FLAG_NULL; ------------------ | | 96| 62.3k|#define FLAG_NULL 0x00 ------------------ | Branch (1256:20): [True: 62.3k, False: 0] ------------------ 1257| 62.3k| FLAG nongramsuggest = pAMgr ? pAMgr->get_nongramsuggest() : FLAG_NULL; ------------------ | | 95| 62.3k|#define FLAG unsigned short ------------------ FLAG nongramsuggest = pAMgr ? pAMgr->get_nongramsuggest() : FLAG_NULL; ------------------ | | 96| 62.3k|#define FLAG_NULL 0x00 ------------------ | Branch (1257:25): [True: 62.3k, False: 0] ------------------ 1258| 62.3k| FLAG onlyincompound = pAMgr ? pAMgr->get_onlyincompound() : FLAG_NULL; ------------------ | | 95| 62.3k|#define FLAG unsigned short ------------------ FLAG onlyincompound = pAMgr ? pAMgr->get_onlyincompound() : FLAG_NULL; ------------------ | | 96| 62.3k|#define FLAG_NULL 0x00 ------------------ | Branch (1258:25): [True: 62.3k, False: 0] ------------------ 1259| | 1260| 62.3k| std::vector w_word, w_target; 1261| 62.3k| if (utf8) { ------------------ | Branch (1261:7): [True: 24.2k, False: 38.0k] ------------------ 1262| 24.2k| u8_u16(w_word, word); 1263| 24.2k| u8_u16(w_target, target); 1264| 24.2k| } 1265| | 1266| 62.3k| std::string f; 1267| 62.3k| std::vector w_f; 1268| | 1269| 62.3k| for (const auto& i : rHMgr) { ------------------ | Branch (1269:22): [True: 62.3k, False: 62.3k] ------------------ 1270| 2.08M| while (nullptr != (hp = i->walk_hashtable(col, hp))) { ------------------ | Branch (1270:12): [True: 2.02M, False: 62.3k] ------------------ 1271| | // skip exceptions 1272| 2.02M| if ( 1273| | // skip it, if the word length different by 5 or 1274| | // more characters (to avoid strange suggestions) 1275| | // (except Unicode characters over BMP) 1276| 2.02M| (((abs(n - hp->clen) > 4) && !nonbmp)) || ------------------ | Branch (1276:12): [True: 1.40M, False: 621k] | Branch (1276:14): [True: 1.26M, False: 759k] | Branch (1276:41): [True: 1.21M, False: 45.1k] ------------------ 1277| | // don't suggest capitalized dictionary words for 1278| | // lower case misspellings in ngram suggestions, except 1279| | // - PHONE usage, or 1280| | // - in the case of German, where not only proper 1281| | // nouns are capitalized, or 1282| | // - the capitalized word has special pronunciation 1283| 805k| ((captype == NOCAP) && (hp->var & H_OPT_INITCAP) && ------------------ | | 92| 805k|#define NOCAP 0 ------------------ ((captype == NOCAP) && (hp->var & H_OPT_INITCAP) && ------------------ | | 50| 190k|#define H_OPT_INITCAP (1 << 3) // is dictionary word capitalized? ------------------ | Branch (1283:13): [True: 190k, False: 614k] | Branch (1283:35): [True: 84.5k, False: 106k] ------------------ 1284| 84.5k| !ph && (langnum != LANG_de) && !(hp->var & H_OPT_PHON)) || ------------------ | | 49| 82.3k|#define H_OPT_PHON (1 << 2) // is there ph: field in the morphological data? ------------------ | Branch (1284:15): [True: 82.3k, False: 2.25k] | Branch (1284:22): [True: 82.3k, False: 0] | Branch (1284:46): [True: 82.3k, False: 0] ------------------ 1285| | // or it has one of the following special flags 1286| 722k| ((hp->astr) && (pAMgr) && ------------------ | Branch (1286:13): [True: 200k, False: 522k] | Branch (1286:27): [True: 200k, False: 0] ------------------ 1287| 200k| (TESTAFF(hp->astr, forbiddenword, hp->alen) || ------------------ | | 99| 400k|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 2.15k, False: 198k] | | ------------------ ------------------ 1288| 198k| TESTAFF(hp->astr, ONLYUPCASEFLAG, hp->alen) || ------------------ | | 99| 398k|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 99.4k, False: 98.8k] | | ------------------ ------------------ 1289| 98.8k| TESTAFF(hp->astr, nosuggest, hp->alen) || ------------------ | | 99| 299k|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 98.8k] | | ------------------ ------------------ 1290| 98.8k| TESTAFF(hp->astr, nongramsuggest, hp->alen) || ------------------ | | 99| 299k|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 98.8k] | | ------------------ ------------------ 1291| 98.8k| TESTAFF(hp->astr, onlyincompound, hp->alen))) ------------------ | | 99| 98.8k|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 98.8k] | | ------------------ ------------------ 1292| 2.02M| ) 1293| 1.40M| continue; 1294| | 1295| 621k| if (utf8) { ------------------ | Branch (1295:11): [True: 270k, False: 350k] ------------------ 1296| 270k| u8_u16(w_f, HENTRY_WORD(hp)); ------------------ | | 54| 270k|#define HENTRY_WORD(h) &(h->word[0]) ------------------ 1297| | 1298| 270k| int leftcommon = leftcommonsubstring(w_word, w_f); 1299| 270k| if (low) { ------------------ | Branch (1299:13): [True: 270k, False: 0] ------------------ 1300| | // lowering dictionary word 1301| 270k| mkallsmall_utf(w_f, langnum); 1302| 270k| } 1303| 270k| sc = ngram(3, w_word, w_f, NGRAM_LONGER_WORSE) + leftcommon; ------------------ | | 81| 270k|#define NGRAM_LONGER_WORSE (1 << 0) ------------------ 1304| 350k| } else { 1305| 350k| f.assign(HENTRY_WORD(hp)); ------------------ | | 54| 350k|#define HENTRY_WORD(h) &(h->word[0]) ------------------ 1306| | 1307| 350k| int leftcommon = leftcommonsubstring(word, f.c_str()); 1308| 350k| if (low) { ------------------ | Branch (1308:13): [True: 328k, False: 22.0k] ------------------ 1309| | // lowering dictionary word 1310| 328k| mkallsmall(f, csconv); 1311| 328k| } 1312| 350k| sc = ngram(3, word, f, NGRAM_LONGER_WORSE) + leftcommon; ------------------ | | 81| 350k|#define NGRAM_LONGER_WORSE (1 << 0) ------------------ 1313| 350k| } 1314| | 1315| | // check special pronunciation 1316| 621k| f.clear(); 1317| 621k| if ((hp->var & H_OPT_PHON) && ------------------ | | 49| 621k|#define H_OPT_PHON (1 << 2) // is there ph: field in the morphological data? ------------------ | Branch (1317:11): [True: 826, False: 620k] | Branch (1317:11): [True: 826, False: 620k] ------------------ 1318| 826| copy_field(f, HENTRY_DATA(hp), MORPH_PHON)) { ------------------ | | 114| 826|#define MORPH_PHON "ph:" ------------------ | Branch (1318:11): [True: 826, False: 0] ------------------ 1319| 826| int sc2; 1320| 826| if (utf8) { ------------------ | Branch (1320:13): [True: 0, False: 826] ------------------ 1321| 0| u8_u16(w_f, f); 1322| | 1323| 0| int leftcommon = leftcommonsubstring(w_word, w_f); 1324| 0| if (low) { ------------------ | Branch (1324:15): [True: 0, False: 0] ------------------ 1325| | // lowering dictionary word 1326| 0| mkallsmall_utf(w_f, langnum); 1327| 0| } 1328| 0| sc2 = ngram(3, w_word, w_f, NGRAM_LONGER_WORSE) + leftcommon; ------------------ | | 81| 0|#define NGRAM_LONGER_WORSE (1 << 0) ------------------ 1329| 826| } else { 1330| 826| int leftcommon = leftcommonsubstring(word, f.c_str()); 1331| 826| if (low) { ------------------ | Branch (1331:15): [True: 826, False: 0] ------------------ 1332| | // lowering dictionary word 1333| 826| mkallsmall(f, csconv); 1334| 826| } 1335| 826| sc2 = ngram(3, word, f, NGRAM_LONGER_WORSE) + leftcommon; ------------------ | | 81| 826|#define NGRAM_LONGER_WORSE (1 << 0) ------------------ 1336| 826| } 1337| 826| if (sc2 > sc) ------------------ | Branch (1337:13): [True: 389, False: 437] ------------------ 1338| 389| sc = sc2; 1339| 826| } 1340| | 1341| 621k| int scphon = -20000; 1342| 621k| if (ph && (sc > 2) && (abs(n - (int)hp->clen) <= 3)) { ------------------ | Branch (1342:11): [True: 36.7k, False: 584k] | Branch (1342:17): [True: 8.70k, False: 28.0k] | Branch (1342:29): [True: 7.38k, False: 1.31k] ------------------ 1343| 7.38k| if (utf8) { ------------------ | Branch (1343:13): [True: 0, False: 7.38k] ------------------ 1344| 0| u8_u16(w_candidate, HENTRY_WORD(hp)); ------------------ | | 54| 0|#define HENTRY_WORD(h) &(h->word[0]) ------------------ 1345| 0| mkallcap_utf(w_candidate, langnum); 1346| 0| u16_u8(candidate, w_candidate); 1347| 7.38k| } else { 1348| 7.38k| candidate = HENTRY_WORD(hp); ------------------ | | 54| 7.38k|#define HENTRY_WORD(h) &(h->word[0]) ------------------ 1349| 7.38k| mkallcap(candidate, csconv); 1350| 7.38k| } 1351| 7.38k| f = phonet(candidate, *ph); 1352| 7.38k| if (utf8) { ------------------ | Branch (1352:13): [True: 0, False: 7.38k] ------------------ 1353| 0| u8_u16(w_f, f); 1354| 0| scphon = 2 * ngram(3, w_target, w_f, 1355| 0| NGRAM_LONGER_WORSE); ------------------ | | 81| 0|#define NGRAM_LONGER_WORSE (1 << 0) ------------------ 1356| 7.38k| } else { 1357| 7.38k| scphon = 2 * ngram(3, target, f, 1358| 7.38k| NGRAM_LONGER_WORSE); ------------------ | | 81| 7.38k|#define NGRAM_LONGER_WORSE (1 << 0) ------------------ 1359| 7.38k| } 1360| 7.38k| } 1361| | 1362| 621k| if (sc > scores[lp]) { ------------------ | Branch (1362:11): [True: 621k, False: 0] ------------------ 1363| 621k| scores[lp] = sc; 1364| 621k| roots[lp] = hp; 1365| 621k| has_roots = true; 1366| 621k| lval = sc; 1367| 62.7M| for (int j = 0; j < MAX_ROOTS; j++) ------------------ | | 74| 62.7M|#define MAX_ROOTS 100 ------------------ | Branch (1367:25): [True: 62.1M, False: 621k] ------------------ 1368| 62.1M| if (scores[j] < lval) { ------------------ | Branch (1368:15): [True: 51.0M, False: 11.0M] ------------------ 1369| 51.0M| lp = j; 1370| 51.0M| lval = scores[j]; 1371| 51.0M| } 1372| 621k| } 1373| | 1374| 621k| if (scphon > scoresphon[lpphon]) { ------------------ | Branch (1374:11): [True: 7.38k, False: 613k] ------------------ 1375| 7.38k| scoresphon[lpphon] = scphon; 1376| 7.38k| rootsphon[lpphon] = HENTRY_WORD(hp); ------------------ | | 54| 7.38k|#define HENTRY_WORD(h) &(h->word[0]) ------------------ 1377| 7.38k| has_rootsphon = true; 1378| 7.38k| lval = scphon; 1379| 746k| for (int j = 0; j < MAX_ROOTS; j++) ------------------ | | 74| 746k|#define MAX_ROOTS 100 ------------------ | Branch (1379:25): [True: 738k, False: 7.38k] ------------------ 1380| 738k| if (scoresphon[j] < lval) { ------------------ | Branch (1380:15): [True: 648k, False: 90.0k] ------------------ 1381| 648k| lpphon = j; 1382| 648k| lval = scoresphon[j]; 1383| 648k| } 1384| 7.38k| } 1385| 621k| } 1386| 62.3k| } 1387| | 1388| 62.3k| if (!has_roots && !has_rootsphon) { ------------------ | Branch (1388:7): [True: 13.8k, False: 48.4k] | Branch (1388:21): [True: 13.8k, False: 0] ------------------ 1389| | // with no roots there will be no guesses and no point running 1390| | // ngram 1391| 13.8k| return; 1392| 13.8k| } 1393| | 1394| | // find minimum threshold for a passable suggestion 1395| | // mangle original word three differnt ways 1396| | // and score them to generate a minimum acceptable score 1397| 48.4k| std::vector w_mw; 1398| 48.4k| int thresh = 0; 1399| 193k| for (int sp = 1; sp < 4; sp++) { ------------------ | Branch (1399:20): [True: 145k, False: 48.4k] ------------------ 1400| 145k| if (utf8) { ------------------ | Branch (1400:9): [True: 59.5k, False: 85.9k] ------------------ 1401| 59.5k| w_mw = w_word; 1402| 110k| for (int k = sp; k < n; k += 4) { ------------------ | Branch (1402:24): [True: 50.6k, False: 59.5k] ------------------ 1403| 50.6k| w_mw[k].l = '*'; 1404| 50.6k| w_mw[k].h = 0; 1405| 50.6k| } 1406| | 1407| 59.5k| if (low) { ------------------ | Branch (1407:11): [True: 59.5k, False: 0] ------------------ 1408| | // lowering dictionary word 1409| 59.5k| mkallsmall_utf(w_mw, langnum); 1410| 59.5k| } 1411| | 1412| 59.5k| thresh += ngram(n, w_word, w_mw, NGRAM_ANY_MISMATCH); ------------------ | | 82| 59.5k|#define NGRAM_ANY_MISMATCH (1 << 1) ------------------ 1413| 85.9k| } else { 1414| 85.9k| std::string mw = word; 1415| 195k| for (int k = sp; k < n; k += 4) ------------------ | Branch (1415:24): [True: 110k, False: 85.9k] ------------------ 1416| 110k| mw[k] = '*'; 1417| | 1418| 85.9k| if (low) { ------------------ | Branch (1418:11): [True: 80.6k, False: 5.27k] ------------------ 1419| | // lowering dictionary word 1420| 80.6k| mkallsmall(mw, csconv); 1421| 80.6k| } 1422| | 1423| 85.9k| thresh += ngram(n, word, mw, NGRAM_ANY_MISMATCH); ------------------ | | 82| 85.9k|#define NGRAM_ANY_MISMATCH (1 << 1) ------------------ 1424| 85.9k| } 1425| 145k| } 1426| 48.4k| thresh = thresh / 3; 1427| 48.4k| thresh--; 1428| | 1429| | // now expand affixes on each of these root words and 1430| | // and use length adjusted ngram scores to select 1431| | // possible suggestions 1432| 48.4k| char* guess[MAX_GUESS]; 1433| 48.4k| char* guessorig[MAX_GUESS]; 1434| 48.4k| int gscore[MAX_GUESS]; 1435| 9.74M| for (int i = 0; i < MAX_GUESS; i++) { ------------------ | | 76| 9.74M|#define MAX_GUESS 200 ------------------ | Branch (1435:19): [True: 9.69M, False: 48.4k] ------------------ 1436| 9.69M| guess[i] = nullptr; 1437| 9.69M| guessorig[i] = nullptr; 1438| 9.69M| gscore[i] = -100 * i; 1439| 9.69M| } 1440| | 1441| 48.4k| lp = MAX_GUESS - 1; ------------------ | | 76| 48.4k|#define MAX_GUESS 200 ------------------ 1442| | 1443| 48.4k| std::vector glst(MAX_WORDS); ------------------ | | 75| 48.4k|#define MAX_WORDS 100 ------------------ 1444| | 1445| 4.84M| for (auto& root : roots) { ------------------ | Branch (1445:19): [True: 4.84M, False: 48.4k] ------------------ 1446| 4.84M| if (root) { ------------------ | Branch (1446:9): [True: 621k, False: 4.22M] ------------------ 1447| 621k| struct hentry* rp = root; 1448| | 1449| 621k| f.clear(); 1450| 621k| const char* field = nullptr; 1451| 621k| if ((rp->var & H_OPT_PHON) && copy_field(f, HENTRY_DATA(rp), MORPH_PHON)) ------------------ | | 49| 621k|#define H_OPT_PHON (1 << 2) // is there ph: field in the morphological data? ------------------ if ((rp->var & H_OPT_PHON) && copy_field(f, HENTRY_DATA(rp), MORPH_PHON)) ------------------ | | 114| 826|#define MORPH_PHON "ph:" ------------------ | Branch (1451:11): [True: 826, False: 620k] | Branch (1451:11): [True: 826, False: 620k] | Branch (1451:37): [True: 826, False: 0] ------------------ 1452| 826| field = f.c_str(); 1453| 621k| int nw = pAMgr->expand_rootword( 1454| 621k| glst.data(), MAX_WORDS, HENTRY_WORD(rp), rp->blen, rp->astr, rp->alen, word, ------------------ | | 75| 621k|#define MAX_WORDS 100 ------------------ glst.data(), MAX_WORDS, HENTRY_WORD(rp), rp->blen, rp->astr, rp->alen, word, ------------------ | | 54| 621k|#define HENTRY_WORD(h) &(h->word[0]) ------------------ 1455| 621k| nc, field); 1456| | 1457| 1.24M| for (int k = 0; k < nw; k++) { ------------------ | Branch (1457:23): [True: 619k, False: 621k] ------------------ 1458| 619k| if (utf8) { ------------------ | Branch (1458:13): [True: 270k, False: 349k] ------------------ 1459| 270k| u8_u16(w_f, glst[k].word); 1460| | 1461| 270k| int leftcommon = leftcommonsubstring(w_word, w_f); 1462| 270k| if (low) { ------------------ | Branch (1462:15): [True: 270k, False: 0] ------------------ 1463| | // lowering dictionary word 1464| 270k| mkallsmall_utf(w_f, langnum); 1465| 270k| } 1466| | 1467| 270k| sc = ngram(n, w_word, w_f, NGRAM_ANY_MISMATCH) + leftcommon; ------------------ | | 82| 270k|#define NGRAM_ANY_MISMATCH (1 << 1) ------------------ 1468| 349k| } else { 1469| 349k| f = glst[k].word; 1470| | 1471| 349k| int leftcommon = leftcommonsubstring(word, f.c_str()); 1472| 349k| if (low) { ------------------ | Branch (1472:15): [True: 326k, False: 22.0k] ------------------ 1473| | // lowering dictionary word 1474| 326k| mkallsmall(f, csconv); 1475| 326k| } 1476| | 1477| 349k| sc = ngram(n, word, f, NGRAM_ANY_MISMATCH) + leftcommon; ------------------ | | 82| 349k|#define NGRAM_ANY_MISMATCH (1 << 1) ------------------ 1478| 349k| } 1479| | 1480| 619k| if (sc > thresh) { ------------------ | Branch (1480:13): [True: 102k, False: 517k] ------------------ 1481| 102k| if (sc > gscore[lp]) { ------------------ | Branch (1481:15): [True: 102k, False: 0] ------------------ 1482| 102k| if (guess[lp]) { ------------------ | Branch (1482:17): [True: 0, False: 102k] ------------------ 1483| 0| delete[] guess[lp]; 1484| 0| if (guessorig[lp]) { ------------------ | Branch (1484:19): [True: 0, False: 0] ------------------ 1485| 0| delete[] guessorig[lp]; 1486| 0| guessorig[lp] = nullptr; 1487| 0| } 1488| 0| } 1489| 102k| gscore[lp] = sc; 1490| 102k| guess[lp] = glst[k].word; 1491| 102k| guessorig[lp] = glst[k].orig; 1492| 102k| lval = sc; 1493| 20.5M| for (int j = 0; j < MAX_GUESS; j++) ------------------ | | 76| 20.5M|#define MAX_GUESS 200 ------------------ | Branch (1493:29): [True: 20.4M, False: 102k] ------------------ 1494| 20.4M| if (gscore[j] < lval) { ------------------ | Branch (1494:19): [True: 19.5M, False: 913k] ------------------ 1495| 19.5M| lp = j; 1496| 19.5M| lval = gscore[j]; 1497| 19.5M| } 1498| 102k| } else { 1499| 0| delete[] glst[k].word; 1500| 0| delete[] glst[k].orig; 1501| 0| } 1502| 517k| } else { 1503| 517k| delete[] glst[k].word; 1504| 517k| delete[] glst[k].orig; 1505| 517k| } 1506| 619k| } 1507| 621k| } 1508| 4.84M| } 1509| 48.4k| glst.clear(); 1510| | 1511| | // now we are done generating guesses 1512| | // sort in order of decreasing score 1513| | 1514| 48.4k| bubblesort(&guess[0], &guessorig[0], &gscore[0], MAX_GUESS); ------------------ | | 76| 48.4k|#define MAX_GUESS 200 ------------------ 1515| 48.4k| if (ph) ------------------ | Branch (1515:7): [True: 2.82k, False: 45.6k] ------------------ 1516| 2.82k| bubblesort(&rootsphon[0], nullptr, &scoresphon[0], MAX_ROOTS); ------------------ | | 74| 2.82k|#define MAX_ROOTS 100 ------------------ 1517| | 1518| | // weight suggestions with a similarity index, based on 1519| | // the longest common subsequent algorithm and resort 1520| | 1521| 48.4k| int is_swap = 0; 1522| 48.4k| int re = 0; 1523| 48.4k| double fact = 1.0; 1524| 48.4k| if (pAMgr) { ------------------ | Branch (1524:7): [True: 48.4k, False: 0] ------------------ 1525| 48.4k| int maxd = pAMgr->get_maxdiff(); 1526| 48.4k| if (maxd >= 0) ------------------ | Branch (1526:9): [True: 0, False: 48.4k] ------------------ 1527| 0| fact = (10.0 - maxd) / 5.0; 1528| 48.4k| } 1529| | 1530| 48.4k| std::vector w_gl; 1531| 9.30M| for (int i = 0; i < MAX_GUESS; i++) { ------------------ | | 76| 9.30M|#define MAX_GUESS 200 ------------------ | Branch (1531:19): [True: 9.25M, False: 46.2k] ------------------ 1532| 9.25M| if (guess[i]) { ------------------ | Branch (1532:9): [True: 81.9k, False: 9.17M] ------------------ 1533| | // lowering guess[i] 1534| 81.9k| std::string gl; 1535| 81.9k| int len; 1536| 81.9k| if (utf8) { ------------------ | Branch (1536:11): [True: 23.8k, False: 58.1k] ------------------ 1537| 23.8k| len = u8_u16(w_gl, guess[i]); 1538| 23.8k| mkallsmall_utf(w_gl, langnum); 1539| 23.8k| u16_u8(gl, w_gl); 1540| 58.1k| } else { 1541| 58.1k| gl.assign(guess[i]); 1542| 58.1k| if (!nonbmp) ------------------ | Branch (1542:13): [True: 58.1k, False: 0] ------------------ 1543| 58.1k| mkallsmall(gl, csconv); 1544| 58.1k| len = strlen(guess[i]); 1545| 58.1k| } 1546| | 1547| 81.9k| int _lcs = lcslen(word, gl.c_str()); 1548| | 1549| | // same characters with different casing 1550| 81.9k| if ((n == len) && (n == _lcs)) { ------------------ | Branch (1550:11): [True: 17.9k, False: 64.0k] | Branch (1550:25): [True: 2.23k, False: 15.7k] ------------------ 1551| 2.23k| gscore[i] += 2000; 1552| 2.23k| break; 1553| 2.23k| } 1554| | // using 2-gram instead of 3, and other weightening 1555| | 1556| 79.7k| if (utf8) { ------------------ | Branch (1556:11): [True: 22.0k, False: 57.6k] ------------------ 1557| 22.0k| u8_u16(w_gl, gl); 1558| | //w_gl is lowercase already at this point 1559| 22.0k| re = ngram(2, w_word, w_gl, NGRAM_ANY_MISMATCH | NGRAM_WEIGHTED); ------------------ | | 82| 22.0k|#define NGRAM_ANY_MISMATCH (1 << 1) ------------------ re = ngram(2, w_word, w_gl, NGRAM_ANY_MISMATCH | NGRAM_WEIGHTED); ------------------ | | 84| 22.0k|#define NGRAM_WEIGHTED (1 << 3) ------------------ 1560| 22.0k| if (low) { ------------------ | Branch (1560:13): [True: 22.0k, False: 0] ------------------ 1561| 22.0k| w_f = w_word; 1562| | // lowering dictionary word 1563| 22.0k| mkallsmall_utf(w_f, langnum); 1564| 22.0k| re += ngram(2, w_gl, w_f, NGRAM_ANY_MISMATCH | NGRAM_WEIGHTED); ------------------ | | 82| 22.0k|#define NGRAM_ANY_MISMATCH (1 << 1) ------------------ re += ngram(2, w_gl, w_f, NGRAM_ANY_MISMATCH | NGRAM_WEIGHTED); ------------------ | | 84| 22.0k|#define NGRAM_WEIGHTED (1 << 3) ------------------ 1565| 22.0k| } else { 1566| 0| re += ngram(2, w_gl, w_word, NGRAM_ANY_MISMATCH | NGRAM_WEIGHTED); ------------------ | | 82| 0|#define NGRAM_ANY_MISMATCH (1 << 1) ------------------ re += ngram(2, w_gl, w_word, NGRAM_ANY_MISMATCH | NGRAM_WEIGHTED); ------------------ | | 84| 0|#define NGRAM_WEIGHTED (1 << 3) ------------------ 1567| 0| } 1568| 57.6k| } else { 1569| | //gl is lowercase already at this point 1570| 57.6k| re = ngram(2, word, gl, NGRAM_ANY_MISMATCH | NGRAM_WEIGHTED); ------------------ | | 82| 57.6k|#define NGRAM_ANY_MISMATCH (1 << 1) ------------------ re = ngram(2, word, gl, NGRAM_ANY_MISMATCH | NGRAM_WEIGHTED); ------------------ | | 84| 57.6k|#define NGRAM_WEIGHTED (1 << 3) ------------------ 1571| 57.6k| if (low) { ------------------ | Branch (1571:13): [True: 57.6k, False: 0] ------------------ 1572| 57.6k| f = word; 1573| | // lowering dictionary word 1574| 57.6k| mkallsmall(f, csconv); 1575| 57.6k| re += ngram(2, gl, f, NGRAM_ANY_MISMATCH | NGRAM_WEIGHTED); ------------------ | | 82| 57.6k|#define NGRAM_ANY_MISMATCH (1 << 1) ------------------ re += ngram(2, gl, f, NGRAM_ANY_MISMATCH | NGRAM_WEIGHTED); ------------------ | | 84| 57.6k|#define NGRAM_WEIGHTED (1 << 3) ------------------ 1576| 57.6k| } else { 1577| 0| re += ngram(2, gl, word, NGRAM_ANY_MISMATCH | NGRAM_WEIGHTED); ------------------ | | 82| 0|#define NGRAM_ANY_MISMATCH (1 << 1) ------------------ re += ngram(2, gl, word, NGRAM_ANY_MISMATCH | NGRAM_WEIGHTED); ------------------ | | 84| 0|#define NGRAM_WEIGHTED (1 << 3) ------------------ 1578| 0| } 1579| 57.6k| } 1580| | 1581| 79.7k| int ngram_score, leftcommon_score; 1582| 79.7k| if (utf8) { ------------------ | Branch (1582:11): [True: 22.0k, False: 57.6k] ------------------ 1583| | //w_gl is lowercase already at this point 1584| 22.0k| ngram_score = ngram(4, w_word, w_gl, NGRAM_ANY_MISMATCH); ------------------ | | 82| 22.0k|#define NGRAM_ANY_MISMATCH (1 << 1) ------------------ 1585| 22.0k| leftcommon_score = leftcommonsubstring(w_word, w_gl); 1586| 57.6k| } else { 1587| | //gl is lowercase already at this point 1588| 57.6k| ngram_score = ngram(4, word, gl, NGRAM_ANY_MISMATCH); ------------------ | | 82| 57.6k|#define NGRAM_ANY_MISMATCH (1 << 1) ------------------ 1589| 57.6k| leftcommon_score = leftcommonsubstring(word, gl.c_str()); 1590| 57.6k| } 1591| 79.7k| gscore[i] = 1592| | // length of longest common subsequent minus length difference 1593| 79.7k| 2 * _lcs - abs((int)(n - len)) + 1594| | // weight length of the left common substring 1595| 79.7k| leftcommon_score + 1596| | // weight equal character positions 1597| 79.7k| (!nonbmp && commoncharacterpositions(word, gl.c_str(), &is_swap) ------------------ | Branch (1597:12): [True: 79.7k, False: 0] | Branch (1597:23): [True: 65.8k, False: 13.9k] ------------------ 1598| 79.7k| ? 1 1599| 79.7k| : 0) + 1600| | // swap character (not neighboring) 1601| 79.7k| ((is_swap) ? 10 : 0) + ------------------ | Branch (1601:12): [True: 161, False: 79.5k] ------------------ 1602| | // ngram 1603| 79.7k| ngram_score + 1604| | // weighted ngrams 1605| 79.7k| re + 1606| | // different limit for dictionaries with PHONE rules 1607| 79.7k| (ph ? (re < len * fact ? -1000 : 0) ------------------ | Branch (1607:12): [True: 5.19k, False: 74.5k] | Branch (1607:18): [True: 3.94k, False: 1.24k] ------------------ 1608| 79.7k| : (re < (n + len) * fact ? -1000 : 0)); ------------------ | Branch (1608:18): [True: 68.9k, False: 5.63k] ------------------ 1609| 79.7k| } 1610| 9.25M| } 1611| | 1612| 48.4k| bubblesort(&guess[0], &guessorig[0], &gscore[0], MAX_GUESS); ------------------ | | 76| 48.4k|#define MAX_GUESS 200 ------------------ 1613| | 1614| | // phonetic version 1615| 48.4k| if (ph) ------------------ | Branch (1615:7): [True: 2.82k, False: 45.6k] ------------------ 1616| 284k| for (int i = 0; i < MAX_ROOTS; i++) { ------------------ | | 74| 284k|#define MAX_ROOTS 100 ------------------ | Branch (1616:21): [True: 282k, False: 2.82k] ------------------ 1617| 282k| if (rootsphon[i]) { ------------------ | Branch (1617:11): [True: 7.38k, False: 274k] ------------------ 1618| | // lowering rootphon[i] 1619| 7.38k| std::string gl; 1620| 7.38k| int len; 1621| 7.38k| if (utf8) { ------------------ | Branch (1621:13): [True: 0, False: 7.38k] ------------------ 1622| 0| len = u8_u16(w_gl, rootsphon[i]); 1623| 0| mkallsmall_utf(w_gl, langnum); 1624| 0| u16_u8(gl, w_gl); 1625| 7.38k| } else { 1626| 7.38k| gl.assign(rootsphon[i]); 1627| 7.38k| if (!nonbmp) ------------------ | Branch (1627:15): [True: 7.38k, False: 0] ------------------ 1628| 7.38k| mkallsmall(gl, csconv); 1629| 7.38k| len = strlen(rootsphon[i]); 1630| 7.38k| } 1631| | 1632| | // weight length of the left common substring 1633| 7.38k| int leftcommon_score; 1634| 7.38k| if (utf8) ------------------ | Branch (1634:13): [True: 0, False: 7.38k] ------------------ 1635| 0| leftcommon_score = leftcommonsubstring(w_word, w_gl); 1636| 7.38k| else 1637| 7.38k| leftcommon_score = leftcommonsubstring(word, gl.c_str()); 1638| | // heuristic weigthing of ngram scores 1639| 7.38k| scoresphon[i] += 2 * lcslen(word, gl) - abs((int)(n - len)) + 1640| 7.38k| leftcommon_score; 1641| 7.38k| } 1642| 282k| } 1643| | 1644| 48.4k| if (ph) ------------------ | Branch (1644:7): [True: 2.82k, False: 45.6k] ------------------ 1645| 2.82k| bubblesort(&rootsphon[0], nullptr, &scoresphon[0], MAX_ROOTS); ------------------ | | 74| 2.82k|#define MAX_ROOTS 100 ------------------ 1646| | 1647| | // copy over 1648| 48.4k| size_t oldns = wlst.size(); 1649| | 1650| 48.4k| int same = 0; 1651| 9.74M| for (int i = 0; i < MAX_GUESS; i++) { ------------------ | | 76| 9.74M|#define MAX_GUESS 200 ------------------ | Branch (1651:19): [True: 9.69M, False: 48.4k] ------------------ 1652| 9.69M| if (guess[i]) { ------------------ | Branch (1652:9): [True: 102k, False: 9.59M] ------------------ 1653| 102k| if ((wlst.size() < oldns + maxngramsugs) && (wlst.size() < maxSug) && ------------------ | Branch (1653:11): [True: 93.5k, False: 8.83k] | Branch (1653:51): [True: 93.5k, False: 0] ------------------ 1654| 93.5k| (!same || (gscore[i] > 1000))) { ------------------ | Branch (1654:12): [True: 15.7k, False: 77.7k] | Branch (1654:21): [True: 0, False: 77.7k] ------------------ 1655| 15.7k| int unique = 1; 1656| | // leave only excellent suggestions, if exists 1657| 15.7k| if (gscore[i] > 1000) ------------------ | Branch (1657:13): [True: 2.23k, False: 13.5k] ------------------ 1658| 2.23k| same = 1; 1659| 13.5k| else if (gscore[i] < -100) { ------------------ | Branch (1659:18): [True: 8.72k, False: 4.83k] ------------------ 1660| 8.72k| same = 1; 1661| | // keep the best ngram suggestions, unless in ONLYMAXDIFF mode 1662| 8.72k| if (wlst.size() > oldns || (pAMgr && pAMgr->get_onlymaxdiff())) { ------------------ | Branch (1662:15): [True: 1.15k, False: 7.56k] | Branch (1662:39): [True: 7.56k, False: 0] | Branch (1662:48): [True: 0, False: 7.56k] ------------------ 1663| 1.15k| delete[] guess[i]; 1664| 1.15k| delete[] guessorig[i]; 1665| 1.15k| continue; 1666| 1.15k| } 1667| 8.72k| } 1668| 14.6k| for (auto& j : wlst) { ------------------ | Branch (1668:22): [True: 7.52k, False: 13.6k] ------------------ 1669| | // don't suggest previous suggestions or a previous suggestion with 1670| | // prefixes or affixes 1671| 7.52k| if ((!guessorig[i] && strstr(guess[i], j.c_str())) || (guessorig[i] && strstr(guessorig[i], j.c_str())) || ------------------ | Branch (1671:15): [True: 961, False: 6.56k] | Branch (1671:16): [True: 7.52k, False: 0] | Branch (1671:33): [True: 961, False: 6.56k] | Branch (1671:66): [True: 0, False: 6.56k] | Branch (1671:82): [True: 0, False: 0] ------------------ 1672| | // check forbidden words 1673| 6.56k| !checkword(guess[i], 0, nullptr, nullptr)) { ------------------ | Branch (1673:15): [True: 0, False: 6.56k] ------------------ 1674| 961| unique = 0; 1675| 961| break; 1676| 961| } 1677| 7.52k| } 1678| 14.6k| if (unique) { ------------------ | Branch (1678:13): [True: 13.6k, False: 961] ------------------ 1679| 13.6k| if (guessorig[i]) { ------------------ | Branch (1679:15): [True: 0, False: 13.6k] ------------------ 1680| 0| wlst.emplace_back(guessorig[i]); 1681| 13.6k| } else { 1682| 13.6k| wlst.emplace_back(guess[i]); 1683| 13.6k| } 1684| 13.6k| } 1685| 14.6k| delete[] guess[i]; 1686| 14.6k| delete[] guessorig[i]; 1687| 86.5k| } else { 1688| 86.5k| delete[] guess[i]; 1689| 86.5k| delete[] guessorig[i]; 1690| 86.5k| } 1691| 102k| } 1692| 9.69M| } 1693| | 1694| 48.4k| oldns = wlst.size(); 1695| 48.4k| if (ph) ------------------ | Branch (1695:7): [True: 2.82k, False: 45.6k] ------------------ 1696| 282k| for (auto& i : rootsphon) { ------------------ | Branch (1696:18): [True: 282k, False: 2.82k] ------------------ 1697| 282k| if (i) { ------------------ | Branch (1697:11): [True: 7.38k, False: 274k] ------------------ 1698| 7.38k| if ((wlst.size() < oldns + MAXPHONSUGS) && (wlst.size() < maxSug)) { ------------------ | | 78| 7.38k|#define MAXPHONSUGS 2 ------------------ | Branch (1698:13): [True: 1.84k, False: 5.54k] | Branch (1698:52): [True: 1.84k, False: 0] ------------------ 1699| 1.84k| int unique = 1; 1700| 3.39k| for (auto& j : wlst) { ------------------ | Branch (1700:24): [True: 3.39k, False: 997] ------------------ 1701| | // don't suggest previous suggestions or a previous suggestion with 1702| | // prefixes or affixes 1703| 3.39k| if (strstr(i, j.c_str()) || ------------------ | Branch (1703:17): [True: 847, False: 2.54k] | Branch (1703:17): [True: 847, False: 2.54k] ------------------ 1704| | // check forbidden words 1705| 2.54k| !checkword(i, 0, nullptr, nullptr)) { ------------------ | Branch (1705:17): [True: 0, False: 2.54k] ------------------ 1706| 847| unique = 0; 1707| 847| break; 1708| 847| } 1709| 3.39k| } 1710| 1.84k| if (unique) { ------------------ | Branch (1710:15): [True: 997, False: 847] ------------------ 1711| 997| wlst.emplace_back(i); 1712| 997| } 1713| 1.84k| } 1714| 7.38k| } 1715| 282k| } 1716| 48.4k|} _ZN10SuggestMgr9checkwordERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEEiPiPNS0_6chrono10time_pointINSA_12steady_clockENSA_8durationIxNS0_5ratioILl1ELl1000000000EEEEEEE: 1727| 86.1M| std::chrono::steady_clock::time_point* timelimit) { 1728| | // check overall suggestion time limit 1729| 86.1M| if (std::chrono::steady_clock::now() - suggest_start > TIMELIMIT_SUGGESTION_MS) ------------------ | | 105| 86.1M|#define TIMELIMIT_SUGGESTION_MS std::chrono::milliseconds(100) ------------------ | Branch (1729:7): [True: 2, False: 86.1M] ------------------ 1730| 2| return 0; 1731| | 1732| | // check time limit 1733| 86.1M| if (timer) { ------------------ | Branch (1733:7): [True: 639k, False: 85.4M] ------------------ 1734| 639k| (*timer)--; 1735| 639k| if (!(*timer) && timelimit) { ------------------ | Branch (1735:9): [True: 1.81k, False: 637k] | Branch (1735:22): [True: 1.81k, False: 0] ------------------ 1736| 1.81k| if (std::chrono::steady_clock::now() - *timelimit > TIMELIMIT_MS) ------------------ | | 106| 1.81k|#define TIMELIMIT_MS std::chrono::milliseconds(50) ------------------ | Branch (1736:11): [True: 2, False: 1.80k] ------------------ 1737| 2| return 0; 1738| 1.80k| *timer = MAXPLUSTIMER; ------------------ | | 108| 1.80k|#define MAXPLUSTIMER 100 ------------------ 1739| 1.80k| } 1740| 639k| } 1741| | 1742| 86.1M| if (pAMgr) { ------------------ | Branch (1742:7): [True: 86.1M, False: 0] ------------------ 1743| 86.1M| struct hentry* rv = nullptr; 1744| 86.1M| int nosuffix = 0; 1745| | 1746| 86.1M| if (cpdsuggest >= 1) { ------------------ | Branch (1746:9): [True: 56.4M, False: 29.6M] ------------------ 1747| 56.4M| if (pAMgr->get_compound()) { ------------------ | Branch (1747:11): [True: 14.3M, False: 42.1M] ------------------ 1748| 14.3M| struct hentry* rv2 = nullptr; 1749| 14.3M| struct hentry* rwords[100] = {}; // buffer for COMPOUND pattern checking 1750| 14.3M| int info = (cpdsuggest == 1) ? SPELL_COMPOUND_2 : 0; ------------------ | | 85| 7.16M|#define SPELL_COMPOUND_2 (1 << 7) // permit only 2 dictionary words in the compound ------------------ | Branch (1750:20): [True: 7.16M, False: 7.16M] ------------------ 1751| 14.3M| AffixScratch scratch; 1752| 14.3M| rv = pAMgr->compound_check(word, 0, 0, 100, 0, nullptr, (hentry**)&rwords, 0, 1, &info, scratch); // EXT 1753| | // TODO filter 3-word or more compound words, as in spell() 1754| | // (it's too slow to call suggest() here for all possible compound words) 1755| 14.3M| if (rv && ------------------ | Branch (1755:13): [True: 0, False: 14.3M] | Branch (1755:13): [True: 0, False: 14.3M] ------------------ 1756| 0| (!(rv2 = pAMgr->lookup(word.c_str(), word.size())) || !rv2->astr || ------------------ | Branch (1756:14): [True: 0, False: 0] | Branch (1756:67): [True: 0, False: 0] ------------------ 1757| 0| !(TESTAFF(rv2->astr, pAMgr->get_forbiddenword(), rv2->alen) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1758| 0| TESTAFF(rv2->astr, pAMgr->get_nosuggest(), rv2->alen)))) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1759| 0| return 3; // XXX obsolote categorisation + only ICONV needs affix 1760| | // flag check? 1761| 14.3M| } 1762| 56.4M| return 0; 1763| 56.4M| } 1764| | 1765| 29.6M| rv = pAMgr->lookup(word.c_str(), word.size()); 1766| | 1767| 29.6M| if (rv) { ------------------ | Branch (1767:9): [True: 32.2k, False: 29.6M] ------------------ 1768| 32.2k| if ((rv->astr) && ------------------ | Branch (1768:11): [True: 372, False: 31.8k] | Branch (1768:11): [True: 0, False: 32.2k] ------------------ 1769| 372| (TESTAFF(rv->astr, pAMgr->get_forbiddenword(), rv->alen) || ------------------ | | 99| 744|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 372] | | ------------------ ------------------ 1770| 372| TESTAFF(rv->astr, pAMgr->get_nosuggest(), rv->alen) || ------------------ | | 99| 744|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 372] | | ------------------ ------------------ 1771| 372| TESTAFF(rv->astr, pAMgr->get_substandard(), rv->alen))) ------------------ | | 99| 372|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 372] | | ------------------ ------------------ 1772| 0| return 0; 1773| 32.6k| while (rv) { ------------------ | Branch (1773:14): [True: 32.2k, False: 362] ------------------ 1774| 32.2k| if (rv->astr && ------------------ | Branch (1774:13): [True: 372, False: 31.8k] | Branch (1774:13): [True: 362, False: 31.8k] ------------------ 1775| 372| (TESTAFF(rv->astr, pAMgr->get_needaffix(), rv->alen) || ------------------ | | 99| 744|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 372] | | ------------------ ------------------ 1776| 372| TESTAFF(rv->astr, ONLYUPCASEFLAG, rv->alen) || ------------------ | | 99| 744|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 362, False: 10] | | ------------------ ------------------ 1777| 362| TESTAFF(rv->astr, pAMgr->get_onlyincompound(), rv->alen))) { ------------------ | | 99| 10|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 10] | | ------------------ ------------------ 1778| 362| rv = rv->next_homonym; 1779| 362| } else 1780| 31.8k| break; 1781| 32.2k| } 1782| 29.6M| } else { 1783| 29.6M| AffixScratch scratch; 1784| 29.6M| rv = pAMgr->prefix_check(word, 0, word.size(), 1785| 29.6M| 0, scratch); // only prefix, and prefix + suffix XXX 1786| 29.6M| } 1787| | 1788| 29.6M| if (rv) { ------------------ | Branch (1788:9): [True: 31.8k, False: 29.6M] ------------------ 1789| 31.8k| nosuffix = 1; 1790| 29.6M| } else { 1791| 29.6M| AffixScratch scratch; 1792| 29.6M| rv = pAMgr->suffix_check(word, 0, word.size(), 0, nullptr, scratch, FLAG_NULL, FLAG_NULL, IN_CPD_NOT); // only suffix ------------------ | | 96| 29.6M|#define FLAG_NULL 0x00 ------------------ rv = pAMgr->suffix_check(word, 0, word.size(), 0, nullptr, scratch, FLAG_NULL, FLAG_NULL, IN_CPD_NOT); // only suffix ------------------ | | 96| 29.6M|#define FLAG_NULL 0x00 ------------------ rv = pAMgr->suffix_check(word, 0, word.size(), 0, nullptr, scratch, FLAG_NULL, FLAG_NULL, IN_CPD_NOT); // only suffix ------------------ | | 72| 29.6M|#define IN_CPD_NOT 0 ------------------ 1793| 29.6M| } 1794| | 1795| 29.6M| if (!rv && pAMgr->have_contclass()) { ------------------ | Branch (1795:9): [True: 29.6M, False: 31.8k] | Branch (1795:16): [True: 10.7M, False: 18.8M] ------------------ 1796| 10.7M| AffixScratch scratch; 1797| 10.7M| rv = pAMgr->suffix_check_twosfx(word, 0, word.size(), 0, nullptr, scratch, FLAG_NULL); ------------------ | | 96| 10.7M|#define FLAG_NULL 0x00 ------------------ 1798| 10.7M| if (!rv) ------------------ | Branch (1798:11): [True: 10.7M, False: 0] ------------------ 1799| 10.7M| rv = pAMgr->prefix_check_twosfx(word, 0, word.size(), 0, scratch, FLAG_NULL); ------------------ | | 96| 10.7M|#define FLAG_NULL 0x00 ------------------ 1800| 10.7M| } 1801| | 1802| | // check forbidden words 1803| 29.6M| if ((rv) && (rv->astr) && ------------------ | Branch (1803:9): [True: 31.8k, False: 29.6M] | Branch (1803:9): [True: 0, False: 29.6M] | Branch (1803:17): [True: 12, False: 31.8k] ------------------ 1804| 12| (TESTAFF(rv->astr, pAMgr->get_forbiddenword(), rv->alen) || ------------------ | | 99| 24|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 12] | | ------------------ ------------------ 1805| 12| TESTAFF(rv->astr, ONLYUPCASEFLAG, rv->alen) || ------------------ | | 99| 24|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 12] | | ------------------ ------------------ 1806| 12| TESTAFF(rv->astr, pAMgr->get_nosuggest(), rv->alen) || ------------------ | | 99| 24|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 12] | | ------------------ ------------------ 1807| 12| TESTAFF(rv->astr, pAMgr->get_onlyincompound(), rv->alen))) ------------------ | | 99| 12|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 12] | | ------------------ ------------------ 1808| 0| return 0; 1809| | 1810| 29.6M| if (rv) { // XXX obsolote ------------------ | Branch (1810:9): [True: 31.8k, False: 29.6M] ------------------ 1811| 31.8k| if ((pAMgr->get_compoundflag()) && ------------------ | Branch (1811:11): [True: 7.38k, False: 24.5k] | Branch (1811:11): [True: 8, False: 31.8k] ------------------ 1812| 7.38k| TESTAFF(rv->astr, pAMgr->get_compoundflag(), rv->alen)) ------------------ | | 99| 7.38k|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 8, False: 7.37k] | | ------------------ ------------------ 1813| 8| return 2 + nosuffix; 1814| 31.8k| return 1; 1815| 31.8k| } 1816| 29.6M| } 1817| 29.6M| return 0; 1818| 86.1M|} _ZN10SuggestMgr15check_forbiddenERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEE: 1820| 14.4k|int SuggestMgr::check_forbidden(const std::string& word) { 1821| 14.4k| if (pAMgr) { ------------------ | Branch (1821:7): [True: 14.4k, False: 0] ------------------ 1822| 14.4k| struct hentry* rv = pAMgr->lookup(word.c_str(), word.size()); 1823| 14.4k| if (rv && rv->astr && ------------------ | Branch (1823:9): [True: 0, False: 14.4k] | Branch (1823:9): [True: 0, False: 14.4k] | Branch (1823:15): [True: 0, False: 0] ------------------ 1824| 0| (TESTAFF(rv->astr, pAMgr->get_needaffix(), rv->alen) || ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1825| 0| TESTAFF(rv->astr, pAMgr->get_onlyincompound(), rv->alen))) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1826| 0| rv = nullptr; 1827| 14.4k| size_t len = word.size(); 1828| 14.4k| AffixScratch scratch; 1829| 14.4k| if (!(pAMgr->prefix_check(word, 0, len, 1, scratch))) ------------------ | Branch (1829:9): [True: 14.4k, False: 0] ------------------ 1830| 14.4k| rv = pAMgr->suffix_check(word, 0, len, 0, nullptr, scratch, FLAG_NULL, FLAG_NULL, IN_CPD_NOT); // prefix+suffix, suffix ------------------ | | 96| 14.4k|#define FLAG_NULL 0x00 ------------------ rv = pAMgr->suffix_check(word, 0, len, 0, nullptr, scratch, FLAG_NULL, FLAG_NULL, IN_CPD_NOT); // prefix+suffix, suffix ------------------ | | 96| 14.4k|#define FLAG_NULL 0x00 ------------------ rv = pAMgr->suffix_check(word, 0, len, 0, nullptr, scratch, FLAG_NULL, FLAG_NULL, IN_CPD_NOT); // prefix+suffix, suffix ------------------ | | 72| 14.4k|#define IN_CPD_NOT 0 ------------------ 1831| | // check forbidden words 1832| 14.4k| if ((rv) && (rv->astr) && ------------------ | Branch (1832:9): [True: 0, False: 14.4k] | Branch (1832:9): [True: 0, False: 14.4k] | Branch (1832:17): [True: 0, False: 0] ------------------ 1833| 0| TESTAFF(rv->astr, pAMgr->get_forbiddenword(), rv->alen)) ------------------ | | 99| 0|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 0] | | ------------------ ------------------ 1834| 0| return 1; 1835| 14.4k| } 1836| 14.4k| return 0; 1837| 14.4k|} _ZN10SuggestMgr13suggest_morphERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEE: 1839| 12.7k|std::string SuggestMgr::suggest_morph(const std::string& in_w) { 1840| 12.7k| std::string result; 1841| | 1842| 12.7k| struct hentry* rv = nullptr; 1843| | 1844| 12.7k| if (!pAMgr) ------------------ | Branch (1844:7): [True: 0, False: 12.7k] ------------------ 1845| 0| return {}; 1846| | 1847| 12.7k| std::string w(in_w); 1848| | 1849| | // word reversing wrapper for complex prefixes 1850| 12.7k| if (complexprefixes) { ------------------ | Branch (1850:7): [True: 648, False: 12.0k] ------------------ 1851| 648| if (utf8) ------------------ | Branch (1851:9): [True: 0, False: 648] ------------------ 1852| 0| reverseword_utf(w); 1853| 648| else 1854| 648| reverseword(w); 1855| 648| } 1856| | 1857| 12.7k| rv = pAMgr->lookup(w.c_str(), w.size()); 1858| | 1859| 15.4k| while (rv) { ------------------ | Branch (1859:10): [True: 2.73k, False: 12.7k] ------------------ 1860| 2.73k| if ((!rv->astr) || ------------------ | Branch (1860:9): [True: 2.43k, False: 304] | Branch (1860:9): [True: 2.73k, False: 0] ------------------ 1861| 304| !(TESTAFF(rv->astr, pAMgr->get_forbiddenword(), rv->alen) || ------------------ | | 99| 608|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 304] | | ------------------ ------------------ 1862| 304| TESTAFF(rv->astr, pAMgr->get_needaffix(), rv->alen) || ------------------ | | 99| 608|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 304] | | ------------------ ------------------ 1863| 2.73k| TESTAFF(rv->astr, pAMgr->get_onlyincompound(), rv->alen))) { ------------------ | | 99| 304|#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) | | ------------------ | | | Branch (99:26): [True: 0, False: 304] | | ------------------ ------------------ 1864| 2.73k| if (!HENTRY_FIND(rv, MORPH_STEM)) { ------------------ | | 103| 2.73k|#define MORPH_STEM "st:" ------------------ | Branch (1864:11): [True: 2.73k, False: 0] ------------------ 1865| 2.73k| result.push_back(MSEP_FLD); ------------------ | | 121| 2.73k|#define MSEP_FLD ' ' ------------------ 1866| 2.73k| result.append(MORPH_STEM); ------------------ | | 103| 2.73k|#define MORPH_STEM "st:" ------------------ 1867| 2.73k| result.append(w); 1868| 2.73k| } 1869| 2.73k| if (HENTRY_DATA(rv)) { ------------------ | Branch (1869:11): [True: 0, False: 2.73k] ------------------ 1870| 0| result.push_back(MSEP_FLD); ------------------ | | 121| 0|#define MSEP_FLD ' ' ------------------ 1871| 0| result.append(HENTRY_DATA2(rv)); 1872| 0| } 1873| 2.73k| result.push_back(MSEP_REC); ------------------ | | 122| 2.73k|#define MSEP_REC '\n' ------------------ 1874| 2.73k| } 1875| 2.73k| rv = rv->next_homonym; 1876| 2.73k| } 1877| | 1878| 12.7k| AffixScratch scratch; 1879| 12.7k| std::string st = pAMgr->affix_check_morph(w, 0, w.size(), scratch); 1880| 12.7k| if (!st.empty()) { ------------------ | Branch (1880:7): [True: 0, False: 12.7k] ------------------ 1881| 0| result.append(st); 1882| 0| } 1883| | 1884| 12.7k| if (pAMgr->get_compound() && result.empty()) { ------------------ | Branch (1884:7): [True: 2.64k, False: 10.0k] | Branch (1884:32): [True: 2.06k, False: 582] ------------------ 1885| 2.06k| struct hentry* rwords[100] = {}; // buffer for COMPOUND pattern checking 1886| 2.06k| pAMgr->compound_check_morph(w, 0, 0, 100, 0, nullptr, (hentry**)&rwords, 0, result, nullptr, scratch); 1887| 2.06k| } 1888| | 1889| 12.7k| line_uniq(result, MSEP_REC); ------------------ | | 122| 12.7k|#define MSEP_REC '\n' ------------------ 1890| | 1891| 12.7k| return result; 1892| 12.7k|} _ZN10SuggestMgr18suggest_hentry_genEP6hentryPKc: 1917| 558k|std::string SuggestMgr::suggest_hentry_gen(hentry* rv, const char* pattern) { 1918| 558k| std::string result; 1919| 558k| int sfxcount = get_sfxcount(pattern); 1920| | 1921| 558k| if (get_sfxcount(HENTRY_DATA(rv)) > sfxcount) ------------------ | Branch (1921:7): [True: 0, False: 558k] ------------------ 1922| 0| return result; 1923| | 1924| 558k| if (HENTRY_DATA(rv)) { ------------------ | Branch (1924:7): [True: 0, False: 558k] ------------------ 1925| 0| std::string aff = pAMgr->morphgen(HENTRY_WORD(rv), rv->blen, rv->astr, rv->alen, ------------------ | | 54| 0|#define HENTRY_WORD(h) &(h->word[0]) ------------------ 1926| 0| HENTRY_DATA(rv), pattern, 0); 1927| 0| if (!aff.empty()) { ------------------ | Branch (1927:9): [True: 0, False: 0] ------------------ 1928| 0| result.append(aff); 1929| 0| result.push_back(MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 1930| 0| } 1931| 0| } 1932| | 1933| | // check all allomorphs 1934| 558k| char* p = nullptr; 1935| 558k| if (HENTRY_DATA(rv)) ------------------ | Branch (1935:7): [True: 0, False: 558k] ------------------ 1936| 0| p = (char*)strstr(HENTRY_DATA2(rv), MORPH_ALLOMORPH); ------------------ | | 104| 0|#define MORPH_ALLOMORPH "al:" ------------------ 1937| 558k| while (p) { ------------------ | Branch (1937:10): [True: 0, False: 558k] ------------------ 1938| 0| p += MORPH_TAG_LEN; ------------------ | | 119| 0|#define MORPH_TAG_LEN strlen(MORPH_STEM) | | ------------------ | | | | 103| 0|#define MORPH_STEM "st:" | | ------------------ ------------------ 1939| 0| int plen = fieldlen(p); 1940| 0| std::string allomorph(p, plen); 1941| 0| struct hentry* rv2 = pAMgr->lookup(allomorph.c_str(), allomorph.size()); 1942| 0| while (rv2) { ------------------ | Branch (1942:12): [True: 0, False: 0] ------------------ 1943| | // if (HENTRY_DATA(rv2) && get_sfxcount(HENTRY_DATA(rv2)) <= 1944| | // sfxcount) { 1945| 0| if (HENTRY_DATA(rv2)) { ------------------ | Branch (1945:11): [True: 0, False: 0] ------------------ 1946| 0| char* st = (char*)strstr(HENTRY_DATA2(rv2), MORPH_STEM); ------------------ | | 103| 0|#define MORPH_STEM "st:" ------------------ 1947| 0| if (st && (strncmp(st + MORPH_TAG_LEN, HENTRY_WORD(rv), ------------------ | | 119| 0|#define MORPH_TAG_LEN strlen(MORPH_STEM) | | ------------------ | | | | 103| 0|#define MORPH_STEM "st:" | | ------------------ ------------------ if (st && (strncmp(st + MORPH_TAG_LEN, HENTRY_WORD(rv), ------------------ | | 54| 0|#define HENTRY_WORD(h) &(h->word[0]) ------------------ | Branch (1947:13): [True: 0, False: 0] | Branch (1947:19): [True: 0, False: 0] ------------------ 1948| 0| fieldlen(st + MORPH_TAG_LEN)) == 0)) { ------------------ | | 119| 0|#define MORPH_TAG_LEN strlen(MORPH_STEM) | | ------------------ | | | | 103| 0|#define MORPH_STEM "st:" | | ------------------ ------------------ 1949| 0| std::string aff = pAMgr->morphgen(HENTRY_WORD(rv2), rv2->blen, rv2->astr, ------------------ | | 54| 0|#define HENTRY_WORD(h) &(h->word[0]) ------------------ 1950| 0| rv2->alen, HENTRY_DATA(rv2), pattern, 0); 1951| 0| if (!aff.empty()) { ------------------ | Branch (1951:15): [True: 0, False: 0] ------------------ 1952| 0| result.append(aff); 1953| 0| result.push_back(MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 1954| 0| } 1955| 0| } 1956| 0| } 1957| 0| rv2 = rv2->next_homonym; 1958| 0| } 1959| 0| p = strstr(p + plen, MORPH_ALLOMORPH); ------------------ | | 104| 0|#define MORPH_ALLOMORPH "al:" ------------------ 1960| 0| } 1961| | 1962| 558k| return result; 1963| 558k|} _ZN10SuggestMgr11suggest_genERKNSt3__16vectorINS0_12basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEENS5_IS7_EEEERKS7_NS0_6chrono10time_pointINSE_12steady_clockENSE_8durationIxNS0_5ratioILl1ELl1000000000EEEEEEE: 1966| 371k| std::chrono::steady_clock::time_point start_time) { 1967| 371k| if (desc.empty() || !pAMgr) ------------------ | Branch (1967:7): [True: 208k, False: 163k] | Branch (1967:23): [True: 0, False: 163k] ------------------ 1968| 208k| return {}; 1969| | 1970| 163k| const char* pattern = in_pattern.c_str(); 1971| 163k| std::string result2; 1972| 163k| std::string newpattern; 1973| 163k| struct hentry* rv = nullptr; 1974| | 1975| | // search affixed forms with and without derivational suffixes 1976| 262k| while (true) { ------------------ | Branch (1976:10): [True: 262k, Folded] ------------------ 1977| 278k| for (const auto& k : desc) { ------------------ | Branch (1977:24): [True: 278k, False: 262k] ------------------ 1978| 278k| if (std::chrono::steady_clock::now() - start_time > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 278k|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (1978:11): [True: 0, False: 278k] ------------------ 1979| 0| return result2; 1980| 278k| std::string result; 1981| | 1982| | // add compound word parts (except the last one) 1983| 278k| const char* s = k.c_str(); 1984| 278k| const char* part = strstr(s, MORPH_PART); ------------------ | | 116| 278k|#define MORPH_PART "pa:" ------------------ 1985| 278k| if (part) { ------------------ | Branch (1985:11): [True: 0, False: 278k] ------------------ 1986| 0| const char* nextpart = strstr(part + 1, MORPH_PART); ------------------ | | 116| 0|#define MORPH_PART "pa:" ------------------ 1987| 0| while (nextpart) { ------------------ | Branch (1987:16): [True: 0, False: 0] ------------------ 1988| 0| std::string field; 1989| 0| copy_field(field, part, MORPH_PART); ------------------ | | 116| 0|#define MORPH_PART "pa:" ------------------ 1990| 0| result.append(field); 1991| 0| part = nextpart; 1992| 0| nextpart = strstr(part + 1, MORPH_PART); ------------------ | | 116| 0|#define MORPH_PART "pa:" ------------------ 1993| 0| } 1994| 0| s = part; 1995| 0| } 1996| | 1997| 278k| std::string tok(s); 1998| 278k| size_t pos = tok.find(" | "); 1999| 278k| while (pos != std::string::npos) { ------------------ | Branch (1999:14): [True: 0, False: 278k] ------------------ 2000| 0| tok[pos + 1] = MSEP_ALT; ------------------ | | 123| 0|#define MSEP_ALT '\v' ------------------ 2001| 0| pos = tok.find(" | ", pos); 2002| 0| } 2003| 278k| std::vector pl = line_tok(tok, MSEP_ALT); ------------------ | | 123| 278k|#define MSEP_ALT '\v' ------------------ 2004| 278k| for (auto& i : pl) { ------------------ | Branch (2004:20): [True: 278k, False: 278k] ------------------ 2005| | // remove inflectional and terminal suffixes 2006| 278k| size_t is = i.find(MORPH_INFL_SFX); ------------------ | | 110| 278k|#define MORPH_INFL_SFX "is:" ------------------ 2007| 278k| if (is != std::string::npos) ------------------ | Branch (2007:13): [True: 0, False: 278k] ------------------ 2008| 0| i.resize(is); 2009| 278k| size_t ts = i.find(MORPH_TERM_SFX); ------------------ | | 111| 278k|#define MORPH_TERM_SFX "ts:" ------------------ 2010| 278k| while (ts != std::string::npos) { ------------------ | Branch (2010:16): [True: 0, False: 278k] ------------------ 2011| 0| i[ts] = '_'; 2012| 0| ts = i.find(MORPH_TERM_SFX); ------------------ | | 111| 0|#define MORPH_TERM_SFX "ts:" ------------------ 2013| 0| } 2014| 278k| const char* st = strstr(s, MORPH_STEM); ------------------ | | 103| 278k|#define MORPH_STEM "st:" ------------------ 2015| 278k| if (st) { ------------------ | Branch (2015:13): [True: 263k, False: 14.7k] ------------------ 2016| 263k| copy_field(tok, st, MORPH_STEM); ------------------ | | 103| 263k|#define MORPH_STEM "st:" ------------------ 2017| 263k| rv = pAMgr->lookup(tok.c_str(), tok.size()); 2018| 542k| while (rv) { ------------------ | Branch (2018:18): [True: 279k, False: 263k] ------------------ 2019| 279k| if (std::chrono::steady_clock::now() - start_time > TIMELIMIT_GLOBAL_MS) ------------------ | | 104| 279k|#define TIMELIMIT_GLOBAL_MS std::chrono::milliseconds(250) ------------------ | Branch (2019:17): [True: 0, False: 279k] ------------------ 2020| 0| return result2; 2021| 279k| std::string newpat(i); 2022| 279k| newpat.append(pattern); 2023| 279k| std::string sg = suggest_hentry_gen(rv, newpat.c_str()); 2024| 279k| if (sg.empty()) ------------------ | Branch (2024:17): [True: 279k, False: 0] ------------------ 2025| 279k| sg = suggest_hentry_gen(rv, pattern); 2026| 279k| if (!sg.empty()) { ------------------ | Branch (2026:17): [True: 0, False: 279k] ------------------ 2027| 0| std::vector gen = line_tok(sg, MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 2028| 0| for (auto& j : gen) { ------------------ | Branch (2028:28): [True: 0, False: 0] ------------------ 2029| 0| result2.push_back(MSEP_REC); ------------------ | | 122| 0|#define MSEP_REC '\n' ------------------ 2030| 0| result2.append(result); 2031| 0| if (i.find(MORPH_SURF_PFX) != std::string::npos) { ------------------ | | 112| 0|#define MORPH_SURF_PFX "sp:" ------------------ | Branch (2031:21): [True: 0, False: 0] ------------------ 2032| 0| std::string field; 2033| 0| copy_field(field, i, MORPH_SURF_PFX); ------------------ | | 112| 0|#define MORPH_SURF_PFX "sp:" ------------------ 2034| 0| result2.append(field); 2035| 0| } 2036| 0| result2.append(j); 2037| 0| } 2038| 0| } 2039| 279k| rv = rv->next_homonym; 2040| 279k| } 2041| 263k| } 2042| 278k| } 2043| 278k| } 2044| | 2045| 262k| if (!result2.empty() || !strstr(pattern, MORPH_DERI_SFX)) ------------------ | | 109| 262k|#define MORPH_DERI_SFX "ds:" ------------------ | Branch (2045:9): [True: 0, False: 262k] | Branch (2045:29): [True: 163k, False: 99.1k] ------------------ 2046| 163k| break; 2047| | 2048| 99.1k| newpattern.assign(pattern); 2049| 99.1k| mystrrep(newpattern, MORPH_DERI_SFX, MORPH_TERM_SFX); ------------------ | | 109| 99.1k|#define MORPH_DERI_SFX "ds:" ------------------ mystrrep(newpattern, MORPH_DERI_SFX, MORPH_TERM_SFX); ------------------ | | 111| 99.1k|#define MORPH_TERM_SFX "ts:" ------------------ 2050| 99.1k| pattern = newpattern.c_str(); 2051| 99.1k| } 2052| 163k| return result2; 2053| 163k|} _ZN10SuggestMgr5ngramEiRKNSt3__16vectorI6w_charNS0_9allocatorIS2_EEEES7_i: 2059| 667k| int opt) { 2060| 667k| int nscore = 0, ns, l1 = su1.size(), l2 = su2.size(); 2061| | 2062| 667k| if (l2 == 0) ------------------ | Branch (2062:7): [True: 6.50k, False: 660k] ------------------ 2063| 6.50k| return 0; 2064| 1.01M| for (int j = 1; j <= n; j++) { ------------------ | Branch (2064:19): [True: 964k, False: 53.0k] ------------------ 2065| 964k| ns = 0; 2066| 4.53M| for (int i = 0; i <= (l1 - j); i++) { ------------------ | Branch (2066:21): [True: 3.57M, False: 964k] ------------------ 2067| 3.57M| int k = 0; 2068| 14.7M| for (int l = 0; l <= (l2 - j); l++) { ------------------ | Branch (2068:23): [True: 12.3M, False: 2.41M] ------------------ 2069| 14.7M| for (k = 0; k < j; k++) { ------------------ | Branch (2069:21): [True: 13.5M, False: 1.15M] ------------------ 2070| 13.5M| if (su1[i + k] != su2[l + k]) ------------------ | Branch (2070:15): [True: 11.2M, False: 2.33M] ------------------ 2071| 11.2M| break; 2072| 13.5M| } 2073| 12.3M| if (k == j) { ------------------ | Branch (2073:13): [True: 1.15M, False: 11.2M] ------------------ 2074| 1.15M| ns++; 2075| 1.15M| break; 2076| 1.15M| } 2077| 12.3M| } 2078| 3.57M| if (k != j && opt & NGRAM_WEIGHTED) { ------------------ | | 84| 2.41M|#define NGRAM_WEIGHTED (1 << 3) ------------------ | Branch (2078:11): [True: 2.41M, False: 1.15M] | Branch (2078:21): [True: 94.8k, False: 2.31M] ------------------ 2079| 94.8k| ns--; 2080| 94.8k| if (i == 0 || i == l1 - j) ------------------ | Branch (2080:13): [True: 22.9k, False: 71.8k] | Branch (2080:23): [True: 37.5k, False: 34.3k] ------------------ 2081| 60.4k| ns--; // side weight 2082| 94.8k| } 2083| 3.57M| } 2084| 964k| nscore = nscore + ns; 2085| 964k| if (ns < 2 && !(opt & NGRAM_WEIGHTED)) ------------------ | | 84| 672k|#define NGRAM_WEIGHTED (1 << 3) ------------------ | Branch (2085:9): [True: 672k, False: 292k] | Branch (2085:19): [True: 607k, False: 64.5k] ------------------ 2086| 607k| break; 2087| 964k| } 2088| | 2089| 660k| ns = 0; 2090| 660k| if (opt & NGRAM_LONGER_WORSE) ------------------ | | 81| 660k|#define NGRAM_LONGER_WORSE (1 << 0) ------------------ | Branch (2090:7): [True: 270k, False: 390k] ------------------ 2091| 270k| ns = (l2 - l1) - 2; 2092| 660k| if (opt & NGRAM_ANY_MISMATCH) ------------------ | | 82| 660k|#define NGRAM_ANY_MISMATCH (1 << 1) ------------------ | Branch (2092:7): [True: 390k, False: 270k] ------------------ 2093| 390k| ns = abs(l2 - l1) - 2; 2094| 660k| ns = (nscore - ((ns > 0) ? ns : 0)); ------------------ | Branch (2094:19): [True: 113k, False: 547k] ------------------ 2095| 660k| return ns; 2096| 667k|} _ZN10SuggestMgr5ngramEiRKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEES8_i: 2102| 966k| int opt) { 2103| 966k| int nscore = 0, ns, l1, l2 = s2.size(); 2104| | 2105| 966k| if (l2 == 0) ------------------ | Branch (2105:7): [True: 13.4k, False: 953k] ------------------ 2106| 13.4k| return 0; 2107| 953k| l1 = s1.size(); 2108| 1.77M| for (int j = 1; j <= n; j++) { ------------------ | Branch (2108:19): [True: 1.62M, False: 151k] ------------------ 2109| 1.62M| ns = 0; 2110| 11.3M| for (int i = 0; i <= (l1 - j); i++) { ------------------ | Branch (2110:21): [True: 9.68M, False: 1.62M] ------------------ 2111| | //s2 is haystack, s1[i..i+j) is needle 2112| 9.68M| if (s2.find(s1.c_str()+i, 0, j) != std::string::npos) { ------------------ | Branch (2112:11): [True: 3.65M, False: 6.02M] ------------------ 2113| 3.65M| ns++; 2114| 6.02M| } else if (opt & NGRAM_WEIGHTED) { ------------------ | | 84| 6.02M|#define NGRAM_WEIGHTED (1 << 3) ------------------ | Branch (2114:18): [True: 398k, False: 5.63M] ------------------ 2115| 398k| ns--; 2116| 398k| if (i == 0 || i == l1 - j) ------------------ | Branch (2116:13): [True: 45.9k, False: 353k] | Branch (2116:23): [True: 124k, False: 228k] ------------------ 2117| 170k| ns--; // side weight 2118| 398k| } 2119| 9.68M| } 2120| 1.62M| nscore = nscore + ns; 2121| 1.62M| if (ns < 2 && !(opt & NGRAM_WEIGHTED)) ------------------ | | 84| 955k|#define NGRAM_WEIGHTED (1 << 3) ------------------ | Branch (2121:9): [True: 955k, False: 665k] | Branch (2121:19): [True: 801k, False: 154k] ------------------ 2122| 801k| break; 2123| 1.62M| } 2124| | 2125| 953k| ns = 0; 2126| 953k| if (opt & NGRAM_LONGER_WORSE) ------------------ | | 81| 953k|#define NGRAM_LONGER_WORSE (1 << 0) ------------------ | Branch (2126:7): [True: 356k, False: 596k] ------------------ 2127| 356k| ns = (l2 - l1) - 2; 2128| 953k| if (opt & NGRAM_ANY_MISMATCH) ------------------ | | 82| 953k|#define NGRAM_ANY_MISMATCH (1 << 1) ------------------ | Branch (2128:7): [True: 596k, False: 356k] ------------------ 2129| 596k| ns = abs(l2 - l1) - 2; 2130| 953k| ns = (nscore - ((ns > 0) ? ns : 0)); ------------------ | Branch (2130:19): [True: 214k, False: 738k] ------------------ 2131| 953k| return ns; 2132| 966k|} _ZN10SuggestMgr19leftcommonsubstringERKNSt3__16vectorI6w_charNS0_9allocatorIS2_EEEES7_: 2137| 563k| const std::vector& su2) { 2138| 563k| int l1 = su1.size(), l2 = su2.size(); 2139| | // decapitalize dictionary word 2140| 563k| if (complexprefixes) { ------------------ | Branch (2140:7): [True: 0, False: 563k] ------------------ 2141| 0| if (l1 && l2 && su1[l1 - 1] == su2[l2 - 1]) ------------------ | Branch (2141:9): [True: 0, False: 0] | Branch (2141:15): [True: 0, False: 0] | Branch (2141:21): [True: 0, False: 0] ------------------ 2142| 0| return 1; 2143| 563k| } else { 2144| 563k| unsigned short idx = su2.empty() ? 0 : (unsigned short)(su2[0]), ------------------ | Branch (2144:26): [True: 0, False: 563k] ------------------ 2145| 563k| otheridx = su1.empty() ? 0 : (unsigned short)(su1[0]); ------------------ | Branch (2145:31): [True: 20.1k, False: 543k] ------------------ 2146| 563k| if (otheridx != idx && (otheridx != unicodetolower(idx, langnum))) ------------------ | Branch (2146:9): [True: 527k, False: 35.8k] | Branch (2146:28): [True: 383k, False: 143k] ------------------ 2147| 383k| return 0; 2148| 179k| int i; 2149| 261k| for (i = 1; (i < l1) && (i < l2) && (su1[i] == su2[i]); ------------------ | Branch (2149:17): [True: 250k, False: 11.0k] | Branch (2149:29): [True: 238k, False: 11.3k] | Branch (2149:41): [True: 81.7k, False: 157k] ------------------ 2150| 179k| i++) 2151| 81.7k| ; 2152| 179k| return i; 2153| 563k| } 2154| 0| return 0; 2155| 563k|} _ZN10SuggestMgr19leftcommonsubstringEPKcS1_: 2160| 765k| const char* s2) { 2161| 765k| if (complexprefixes) { ------------------ | Branch (2161:7): [True: 64.6k, False: 700k] ------------------ 2162| 64.6k| int l1 = strlen(s1), l2 = strlen(s2); 2163| 64.6k| if (l1 && l1 <= l2 && s1[l1 - 1] == s2[l2 - 1]) ------------------ | Branch (2163:9): [True: 62.1k, False: 2.42k] | Branch (2163:15): [True: 30.6k, False: 31.5k] | Branch (2163:27): [True: 2.66k, False: 28.0k] ------------------ 2164| 2.66k| return 1; 2165| 700k| } else if (csconv) { ------------------ | Branch (2165:14): [True: 700k, False: 0] ------------------ 2166| 700k| const char* olds = s1; 2167| | // decapitalise dictionary word 2168| 700k| if ((*s1 != *s2) && (((unsigned char)*s1) != csconv[((unsigned char)*s2)].clower)) ------------------ | Branch (2168:9): [True: 633k, False: 67.0k] | Branch (2168:25): [True: 446k, False: 187k] ------------------ 2169| 446k| return 0; 2170| 515k| do { 2171| 515k| s1++; 2172| 515k| s2++; 2173| 515k| } while ((*s1 == *s2) && (*s1 != '\0')); ------------------ | Branch (2173:14): [True: 261k, False: 253k] | Branch (2173:30): [True: 261k, False: 381] ------------------ 2174| 254k| return (int)(s1 - olds); 2175| 700k| } 2176| 61.9k| return 0; 2177| 765k|} _ZN10SuggestMgr24commoncharacterpositionsEPKcS1_Pi: 2181| 79.7k| int* is_swap) { 2182| 79.7k| int num = 0, diff = 0, diffpos[2]; 2183| 79.7k| *is_swap = 0; 2184| 79.7k| if (utf8) { ------------------ | Branch (2184:7): [True: 22.0k, False: 57.6k] ------------------ 2185| 22.0k| std::vector su1; 2186| 22.0k| std::vector su2; 2187| 22.0k| int l1 = u8_u16(su1, s1), l2 = u8_u16(su2, s2); 2188| | 2189| 22.0k| if (l1 <= 0 || l2 <= 0) ------------------ | Branch (2189:9): [True: 3.72k, False: 18.3k] | Branch (2189:20): [True: 0, False: 18.3k] ------------------ 2190| 3.72k| return 0; 2191| | 2192| | // decapitalize dictionary word 2193| 18.3k| if (complexprefixes) { ------------------ | Branch (2193:9): [True: 0, False: 18.3k] ------------------ 2194| 0| su2[l2 - 1] = lower_utf(su2[l2 - 1], langnum); 2195| 18.3k| } else { 2196| 18.3k| su2[0] = lower_utf(su2[0], langnum); 2197| 18.3k| } 2198| 67.5k| for (int i = 0; (i < l1) && (i < l2); i++) { ------------------ | Branch (2198:21): [True: 54.1k, False: 13.3k] | Branch (2198:33): [True: 49.1k, False: 4.97k] ------------------ 2199| 49.1k| if (su1[i] == su2[i]) { ------------------ | Branch (2199:11): [True: 26.9k, False: 22.2k] ------------------ 2200| 26.9k| num++; 2201| 26.9k| } else { 2202| 22.2k| if (diff < 2) ------------------ | Branch (2202:13): [True: 20.3k, False: 1.96k] ------------------ 2203| 20.3k| diffpos[diff] = i; 2204| 22.2k| diff++; 2205| 22.2k| } 2206| 49.1k| } 2207| 18.3k| if ((diff == 2) && (l1 == l2) && ------------------ | Branch (2207:9): [True: 3.88k, False: 14.4k] | Branch (2207:24): [True: 1.22k, False: 2.65k] ------------------ 2208| 1.22k| (su1[diffpos[0]] == su2[diffpos[1]]) && ------------------ | Branch (2208:9): [True: 197, False: 1.02k] ------------------ 2209| 197| (su1[diffpos[1]] == su2[diffpos[0]])) ------------------ | Branch (2209:9): [True: 71, False: 126] ------------------ 2210| 71| *is_swap = 1; 2211| 57.6k| } else { 2212| 57.6k| size_t i; 2213| 57.6k| std::string t(s2); 2214| | // decapitalize dictionary word 2215| 57.6k| if (complexprefixes) { ------------------ | Branch (2215:9): [True: 3.45k, False: 54.2k] ------------------ 2216| 3.45k| size_t l2 = t.size(); 2217| 3.45k| t[l2 - 1] = csconv[(unsigned char)t[l2 - 1]].clower; 2218| 54.2k| } else { 2219| 54.2k| mkallsmall(t, csconv); 2220| 54.2k| } 2221| 295k| for (i = 0; i < t.size() && (*(s1 + i) != 0); ++i) { ------------------ | Branch (2221:17): [True: 267k, False: 27.4k] | Branch (2221:33): [True: 237k, False: 30.1k] ------------------ 2222| 237k| if (*(s1 + i) == t[i]) { ------------------ | Branch (2222:11): [True: 129k, False: 108k] ------------------ 2223| 129k| num++; 2224| 129k| } else { 2225| 108k| if (diff < 2) ------------------ | Branch (2225:13): [True: 71.0k, False: 37.2k] ------------------ 2226| 71.0k| diffpos[diff] = i; 2227| 108k| diff++; 2228| 108k| } 2229| 237k| } 2230| 57.6k| if ((diff == 2) && (*(s1 + i) == 0) && i == t.size() && ------------------ | Branch (2230:9): [True: 13.8k, False: 43.7k] | Branch (2230:24): [True: 10.7k, False: 3.10k] | Branch (2230:44): [True: 3.67k, False: 7.09k] ------------------ 2231| 3.67k| (*(s1 + diffpos[0]) == t[diffpos[1]]) && ------------------ | Branch (2231:9): [True: 610, False: 3.06k] ------------------ 2232| 610| (*(s1 + diffpos[1]) == t[diffpos[0]])) ------------------ | Branch (2232:9): [True: 90, False: 520] ------------------ 2233| 90| *is_swap = 1; 2234| 57.6k| } 2235| 76.0k| return num; 2236| 79.7k|} _ZN10SuggestMgr10bubblesortEPPcS1_Pii: 2247| 102k|void SuggestMgr::bubblesort(char** rword, char** rword2, int* rsc, int n) { 2248| 102k| int m = 1; 2249| 19.9M| while (m < n) { ------------------ | Branch (2249:10): [True: 19.8M, False: 102k] ------------------ 2250| 19.8M| int j = m; 2251| 40.3M| while (j > 0) { ------------------ | Branch (2251:12): [True: 40.2M, False: 28.2k] ------------------ 2252| 40.2M| if (rsc[j - 1] < rsc[j]) { ------------------ | Branch (2252:11): [True: 20.4M, False: 19.8M] ------------------ 2253| 20.4M| int sctmp = rsc[j - 1]; 2254| 20.4M| char* wdtmp = rword[j - 1]; 2255| 20.4M| rsc[j - 1] = rsc[j]; 2256| 20.4M| rword[j - 1] = rword[j]; 2257| 20.4M| rsc[j] = sctmp; 2258| 20.4M| rword[j] = wdtmp; 2259| 20.4M| if (rword2) { ------------------ | Branch (2259:13): [True: 19.8M, False: 602k] ------------------ 2260| 19.8M| wdtmp = rword2[j - 1]; 2261| 19.8M| rword2[j - 1] = rword2[j]; 2262| 19.8M| rword2[j] = wdtmp; 2263| 19.8M| } 2264| 20.4M| j--; 2265| 20.4M| } else 2266| 19.8M| break; 2267| 40.2M| } 2268| 19.8M| m++; 2269| 19.8M| } 2270| 102k|} _ZN10SuggestMgr3lcsEPKcS1_PiS2_: 2276| 89.3k| int* l2) { 2277| 89.3k| int n, m, i, j; 2278| 89.3k| std::vector su; 2279| 89.3k| std::vector su2; 2280| 89.3k| if (utf8) { ------------------ | Branch (2280:7): [True: 23.8k, False: 65.5k] ------------------ 2281| 23.8k| m = u8_u16(su, s); 2282| 23.8k| n = u8_u16(su2, s2); 2283| 65.5k| } else { 2284| 65.5k| m = strlen(s); 2285| 65.5k| n = strlen(s2); 2286| 65.5k| } 2287| 89.3k| if (m <= 0 || n <= 0) { ------------------ | Branch (2287:7): [True: 8.79k, False: 80.5k] | Branch (2287:17): [True: 0, False: 80.5k] ------------------ 2288| 8.79k| *l1 = 0; 2289| 8.79k| *l2 = 0; 2290| 8.79k| return nullptr; 2291| 8.79k| } 2292| 80.5k| char* c = new char[(m + 1) * (n + 1)](); 2293| 80.5k| char* b = new char[(m + 1) * (n + 1)](); 2294| 462k| for (i = 1; i <= m; i++) { ------------------ | Branch (2294:15): [True: 381k, False: 80.5k] ------------------ 2295| 3.46M| for (j = 1; j <= n; j++) { ------------------ | Branch (2295:17): [True: 3.08M, False: 381k] ------------------ 2296| 3.08M| if (((utf8) && (su[i - 1] == su2[j - 1])) || ------------------ | Branch (2296:12): [True: 236k, False: 2.84M] | Branch (2296:22): [True: 55.7k, False: 180k] ------------------ 2297| 3.02M| ((!utf8) && (s[i - 1] == s2[j - 1]))) { ------------------ | Branch (2297:12): [True: 2.84M, False: 180k] | Branch (2297:23): [True: 465k, False: 2.38M] ------------------ 2298| 521k| c[i * (n + 1) + j] = c[(i - 1) * (n + 1) + j - 1] + 1; 2299| 521k| b[i * (n + 1) + j] = LCS_UPLEFT; 2300| 2.56M| } else if (c[(i - 1) * (n + 1) + j] >= c[i * (n + 1) + j - 1]) { ------------------ | Branch (2300:18): [True: 1.76M, False: 791k] ------------------ 2301| 1.76M| c[i * (n + 1) + j] = c[(i - 1) * (n + 1) + j]; 2302| 1.76M| b[i * (n + 1) + j] = LCS_UP; 2303| 1.76M| } else { 2304| 791k| c[i * (n + 1) + j] = c[i * (n + 1) + j - 1]; 2305| 791k| b[i * (n + 1) + j] = LCS_LEFT; 2306| 791k| } 2307| 3.08M| } 2308| 381k| } 2309| 80.5k| delete[] c; 2310| 80.5k| *l1 = m; 2311| 80.5k| *l2 = n; 2312| 80.5k| return b; 2313| 89.3k|} _ZN10SuggestMgr6lcslenEPKcS1_: 2315| 89.3k|int SuggestMgr::lcslen(const char* s, const char* s2) { 2316| 89.3k| int m, n, len = 0; 2317| 89.3k| char* result = lcs(s, s2, &m, &n); 2318| 89.3k| if (!result) ------------------ | Branch (2318:7): [True: 8.79k, False: 80.5k] ------------------ 2319| 8.79k| return 0; 2320| 80.5k| int i = m, j = n; 2321| 618k| while ((i != 0) && (j != 0)) { ------------------ | Branch (2321:10): [True: 545k, False: 73.0k] | Branch (2321:22): [True: 537k, False: 7.50k] ------------------ 2322| 537k| if (result[i * (n + 1) + j] == LCS_UPLEFT) { ------------------ | Branch (2322:9): [True: 217k, False: 320k] ------------------ 2323| 217k| len++; 2324| 217k| i--; 2325| 217k| j--; 2326| 320k| } else if (result[i * (n + 1) + j] == LCS_UP) { ------------------ | Branch (2326:16): [True: 151k, False: 169k] ------------------ 2327| 151k| i--; 2328| 151k| } else 2329| 169k| j--; 2330| 537k| } 2331| 80.5k| delete[] result; 2332| 80.5k| return len; 2333| 89.3k|} _ZN10SuggestMgr6lcslenERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEES8_: 2335| 7.38k|int SuggestMgr::lcslen(const std::string& s, const std::string& s2) { 2336| 7.38k| return lcslen(s.c_str(), s2.c_str()); 2337| 7.38k|} suggestmgr.cxx:_ZN12_GLOBAL__N_115ngsuggest_guardC2EbPK7cs_infoPiPS3_: 1156| 62.3k| : m_nonbmp(nonbmp) 1157| 62.3k| , m_origconv(origconv) 1158| 62.3k| , m_utf8(utf8) 1159| 62.3k| , m_csconv(csconv) 1160| 62.3k| { 1161| 62.3k| } suggestmgr.cxx:_ZN12_GLOBAL__N_115ngsuggest_guardD2Ev: 1164| 62.3k| { 1165| 62.3k| if (m_nonbmp) { ------------------ | Branch (1165:11): [True: 1.75k, False: 60.5k] ------------------ 1166| 1.75k| *m_csconv = m_origconv; 1167| 1.75k| *m_utf8 = 1; 1168| 1.75k| } 1169| 62.3k| } suggestmgr.cxx:_ZL12get_sfxcountPKc: 1894| 1.11M|static int get_sfxcount(const char* morph) { 1895| 1.11M| if (!morph || !*morph) ------------------ | Branch (1895:7): [True: 558k, False: 558k] | Branch (1895:17): [True: 0, False: 558k] ------------------ 1896| 558k| return 0; 1897| 558k| int n = 0; 1898| 558k| const char* old = morph; 1899| 558k| morph = strstr(morph, MORPH_DERI_SFX); ------------------ | | 109| 558k|#define MORPH_DERI_SFX "ds:" ------------------ 1900| 558k| if (!morph) ------------------ | Branch (1900:7): [True: 347k, False: 211k] ------------------ 1901| 347k| morph = strstr(old, MORPH_INFL_SFX); ------------------ | | 110| 347k|#define MORPH_INFL_SFX "is:" ------------------ 1902| 558k| if (!morph) ------------------ | Branch (1902:7): [True: 346k, False: 211k] ------------------ 1903| 346k| morph = strstr(old, MORPH_TERM_SFX); ------------------ | | 111| 346k|#define MORPH_TERM_SFX "ts:" ------------------ 1904| 1.16M| while (morph) { ------------------ | Branch (1904:10): [True: 603k, False: 558k] ------------------ 1905| 603k| n++; 1906| 603k| old = morph; 1907| 603k| morph = strstr(morph + 1, MORPH_DERI_SFX); ------------------ | | 109| 603k|#define MORPH_DERI_SFX "ds:" ------------------ 1908| 603k| if (!morph) ------------------ | Branch (1908:9): [True: 513k, False: 90.3k] ------------------ 1909| 513k| morph = strstr(old + 1, MORPH_INFL_SFX); ------------------ | | 110| 513k|#define MORPH_INFL_SFX "is:" ------------------ 1910| 603k| if (!morph) ------------------ | Branch (1910:9): [True: 513k, False: 90.5k] ------------------ 1911| 513k| morph = strstr(old + 1, MORPH_TERM_SFX); ------------------ | | 111| 513k|#define MORPH_TERM_SFX "ts:" ------------------ 1912| 603k| } 1913| 558k| return n; 1914| 1.11M|} _ZNK6w_charcvtEv: 57| 207M| { 58| 207M|#if defined(_WIN32) || (defined(__BYTE_ORDER__) && (__BYTE_ORDER__==__ORDER_LITTLE_ENDIAN__)) || defined(__LITTLE_ENDIAN__) 59| | //use little-endian optimized version 60| |#if (__cplusplus >= 202002L || (defined(_MSVC_LANG) && _MSVC_LANG >= 202002L)) && defined __cpp_lib_bit_cast && __cpp_lib_bit_cast >= 201806L 61| | return std::bit_cast(*this); 62| |#else 63| 207M| unsigned short u; 64| 207M| memcpy(&u, this, sizeof(unsigned short)); 65| 207M| return u; 66| 207M|#endif 67| | 68| |#else 69| | return ((unsigned short)h << 8) | (unsigned short)l; 70| |#endif 71| 207M| } _Zlt6w_charS_: 73| 13.0M| friend bool operator<(const w_char a, const w_char b) { 74| 13.0M| return (unsigned short)a < (unsigned short)b; 75| 13.0M| } _Zeq6w_charS_: 77| 77.9M| friend bool operator==(const w_char a, const w_char b) { 78| 77.9M| return (unsigned short)a == (unsigned short)b; 79| 77.9M| } _Zne6w_charS_: 81| 75.7M| friend bool operator!=(const w_char a, const w_char b) { 82| 75.7M| return !(a == b); 83| 75.7M| } _Z8endswithRKNSt3__112basic_stringIcNS_11char_traitsIcEENS_9allocatorIcEEEES7_: 40| 199|{ 41| 199| return str.size() >= suffix.size() && ------------------ | Branch (41:12): [True: 191, False: 8] ------------------ 42| 191| str.compare(str.size() - suffix.size(), suffix.size(), suffix) == 0; ------------------ | Branch (42:12): [True: 38, False: 153] ------------------ 43| 199|} LLVMFuzzerInitialize: 46| 2|{ 47| 2| char* exe_path = (*argv)[0]; 48| | // dirname() can modify its argument. 49| 2| char* exe_path_copy = strdup(exe_path); 50| 2| char* dir = dirname(exe_path_copy); 51| 2| DIR* d = opendir(dir); 52| 2| struct dirent *direntry; 53| 2| std::vector affs; 54| 201| while ((direntry = readdir(d)) != NULL) ------------------ | Branch (54:12): [True: 199, False: 2] ------------------ 55| 199| { 56| 199| std::string entry(direntry->d_name); 57| 199| if (endswith(entry, ".aff")) ------------------ | Branch (57:13): [True: 38, False: 161] ------------------ 58| 38| affs.push_back(std::move(entry)); 59| 199| } 60| 2| closedir(d); 61| 2| free(exe_path_copy); 62| | 63| | // sort so dict iteration order is reproducible across runs regardless of 64| | // readdir ordering; a failing input then reproduces against the same dict 65| 2| std::sort(affs.begin(), affs.end()); 66| 2| for (const auto& entry : affs) ------------------ | Branch (66:28): [True: 38, False: 2] ------------------ 67| 38| { 68| 38| std::string dic = entry.substr(0, entry.size() - 4) + ".dic"; 69| 38| dictionaries.emplace_back(new Hunspell(entry.c_str(), dic.c_str())); 70| 38| } 71| | 72| 2| return 0; 73| 2|} LLVMFuzzerTestOneInput: 76| 3.17k|{ 77| 3.17k| std::string word(data, size); 78| 3.17k| for (auto& dict : dictionaries) ------------------ | Branch (78:21): [True: 60.3k, False: 3.17k] ------------------ 79| 60.3k| { 80| 60.3k| if (!dict->spell(word)) ------------------ | Branch (80:13): [True: 59.9k, False: 367] ------------------ 81| 59.9k| dict->suggest(word); 82| 60.3k| } 83| 3.17k| return 0; 84| 3.17k|}