run_burl_normalize: 26| 1.73k| size_t in_len) { 27| 1.73k| int qs; 28| 1.73k| buffer_copy_string_len(psrc, in, in_len); 29| 1.73k| qs = burl_normalize(psrc, ptmp, flags); 30| 1.73k|} LLVMFuzzerTestOneInput: 32| 1.74k|int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) { 33| 1.74k| if (size <= 4) { ------------------ | Branch (33:9): [True: 4, False: 1.73k] ------------------ 34| 4| return 0; 35| 4| } 36| 1.73k| int flags = ((int*)data)[0]; 37| 1.73k| data += 4; 38| 1.73k| size -= 4; 39| 1.73k| char *new_str = (char *)malloc(size+1); 40| 1.73k| if (new_str == NULL){ ------------------ | Branch (40:9): [True: 0, False: 1.73k] ------------------ 41| 0| return 0; 42| 0| } 43| 1.73k| memcpy(new_str, data, size); 44| 1.73k| new_str[size] = '\0'; 45| | 46| | /* main fuzzer entrypoint for library */ 47| 1.73k| buffer *psrc = buffer_init(); 48| 1.73k| buffer *ptmp = buffer_init(); 49| 1.73k| run_burl_normalize(psrc, ptmp, flags, __LINE__, new_str, size); 50| 1.73k| buffer_urldecode_path(psrc); 51| | 52| 1.73k| buffer_free(psrc); 53| 1.73k| buffer_free(ptmp); 54| 1.73k| free(new_str); 55| 1.73k| return 0; 56| 1.73k|} buffer_init: 14| 3.47k|buffer* buffer_init(void) { 15| | #if 0 /* buffer_init() and chunk_init() can be hot, 16| | * so avoid the additional hop of indirection */ 17| | return ck_calloc(1, sizeof(buffer)); 18| | #else 19| 3.47k| buffer * const b = calloc(1, sizeof(*b)); 20| 3.47k| force_assert(b); ------------------ | | 448| 3.47k|#define force_assert(x) ck_assert(x) | | ------------------ | | | | 115| 3.47k| do { if (!(x)) ck_assert_failed(__FILE__, __LINE__, #x); } while (0) | | | | ------------------ | | | | | Branch (115:18): [True: 0, False: 3.47k] | | | | | Branch (115:75): [Folded - Ignored] | | | | ------------------ | | ------------------ ------------------ 21| 3.47k| return b; 22| 3.47k| #endif 23| 3.47k|} buffer_free: 25| 3.47k|void buffer_free(buffer *b) { 26| 3.47k| if (NULL == b) return; ------------------ | Branch (26:6): [True: 0, False: 3.47k] ------------------ 27| 3.47k| free(b->ptr); 28| 3.47k| free(b); 29| 3.47k|} buffer_string_prepare_copy: 82| 1.36k|char* buffer_string_prepare_copy(buffer * const b, const size_t size) { 83| 1.36k| b->used = 0; 84| | #ifdef __COVERITY__ /*(b->ptr is not NULL if b->size is not 0)*/ 85| | force_assert(size >= b->size || b->ptr); 86| | #endif 87| 1.36k| return (size < b->size) ------------------ | Branch (87:12): [True: 0, False: 1.36k] ------------------ 88| 1.36k| ? b->ptr 89| 1.36k| : buffer_alloc_replace(b, size); 90| 1.36k|} buffer_extend: 127| 135|{ 128| | /* extend buffer to append x (reallocate by power-2 (or larger), if needed) 129| | * (combine buffer_string_prepare_append() and buffer_commit()) 130| | * (future: might make buffer.h static inline func for HTTP/1.1 performance) 131| | * pre-sets '\0' byte and b->used (unlike buffer_string_prepare_append())*/ 132| | #if 0 133| | char * const s = buffer_string_prepare_append(b, x); 134| | b->used += x + (0 == b->used); 135| | #else 136| 135| const uint32_t len = b->used ? b->used-1 : 0; ------------------ | Branch (136:26): [True: 135, False: 0] ------------------ 137| 135| char * const s = (b->size - len >= x + 1) ------------------ | Branch (137:22): [True: 135, False: 0] ------------------ 138| 135| ? b->ptr + len 139| 135| : buffer_string_prepare_append_resize(b, x); 140| 135| b->used = len+x+1; 141| 135| #endif 142| 135| s[x] = '\0'; 143| 135| return s; 144| 135|} buffer_copy_string_len: 172| 3.23k|void buffer_copy_string_len(buffer * const restrict b, const char * const restrict s, const size_t len) { 173| 3.23k| b->used = len + 1; 174| 3.23k| char * const restrict d = (len < b->size) ------------------ | Branch (174:31): [True: 974, False: 2.26k] ------------------ 175| 3.23k| ? b->ptr 176| 3.23k| : buffer_alloc_replace(b, len); 177| 3.23k| d[len] = '\0'; 178| 3.23k| memcpy(d, s, len); 179| 3.23k|} buffer_append_string_len: 198| 135|void buffer_append_string_len(buffer * const restrict b, const char * const restrict s, const size_t len) { 199| 135| memcpy(buffer_extend(b, len), s, len); 200| 135|} hex2int: 386| 99.4M|char hex2int(unsigned char hex) { 387| 99.4M| unsigned char n; 388| 99.4M| return li_cton(hex,n) ? (char)n : 0xFF; ------------------ | | 381| 99.4M| (((n) = (c) - '0') <= 9 || (((n) = ((c)&0xdf) - 'A') <= 5 ? ((n) += 10) : 0)) | | ------------------ | | | Branch (381:4): [True: 77.3M, False: 22.1M] | | | Branch (381:30): [True: 22.1M, False: 0] | | | Branch (381:31): [True: 22.1M, False: 0] | | ------------------ ------------------ 389| 99.4M|} buffer_urldecode_path: 792| 1.73k|void buffer_urldecode_path(buffer * const b) { 793| 1.73k| const size_t len = buffer_clen(b); 794| 1.73k| char *src = len ? memchr(b->ptr, '%', len) : NULL; ------------------ | Branch (794:17): [True: 1.73k, False: 4] ------------------ 795| 1.73k| if (NULL == src) return; ------------------ | Branch (795:9): [True: 343, False: 1.39k] ------------------ 796| | 797| 1.39k| char *dst = src; 798| 49.7M| do { 799| | /* *src == '%' */ 800| 49.7M| unsigned char high = ((unsigned char *)src)[1]; 801| 49.7M| unsigned char low = high ? hex2int(((unsigned char *)src)[2]) : 0xFF; ------------------ | Branch (801:29): [True: 49.7M, False: 0] ------------------ 802| 49.7M| if (0xFF != (high = hex2int(high)) && 0xFF != low) { ------------------ | Branch (802:13): [True: 49.7M, False: 0] | Branch (802:47): [True: 49.7M, False: 0] ------------------ 803| 49.7M| high = (high << 4) | low; /* map ctrls to '_' */ 804| 49.7M| *dst = (high >= 32 && high != 127) ? high : '_'; ------------------ | Branch (804:21): [True: 18.6M, False: 31.1M] | Branch (804:35): [True: 16.2M, False: 2.37M] ------------------ 805| 49.7M| src += 2; 806| 49.7M| } /* else ignore this '%'; leave as-is and move on */ 807| | 808| 55.0M| while ((*++dst = *++src) != '%' && *src) ; ------------------ | Branch (808:16): [True: 5.34M, False: 49.7M] | Branch (808:44): [True: 5.34M, False: 1.39k] ------------------ 809| 49.7M| } while (*src); ------------------ | Branch (809:14): [True: 49.7M, False: 1.39k] ------------------ 810| 1.39k| b->used = (dst - b->ptr) + 1; 811| 1.39k|} buffer_path_simplify: 864| 378|{ 865| 378| char *out = b->ptr; 866| 378| char * const end = b->ptr + b->used - 1; 867| | 868| 378| if (__builtin_expect( (buffer_is_blank(b)), 0)) { ------------------ | Branch (868:9): [True: 0, False: 378] ------------------ 869| 0| buffer_blank(b); 870| 0| return; 871| 0| } 872| | 873| | #if defined(_WIN32) || defined(__CYGWIN__) 874| | /* cygwin is treating \ and / the same, so we have to that too */ 875| | for (char *p = b->ptr; *p; p++) { 876| | if (*p == '\\') *p = '/'; 877| | } 878| | #endif 879| | 880| 378| *end = '/'; /*(end of path modified to avoid need to check '\0')*/ 881| | 882| 378| char *walk = out; 883| 378| if (__builtin_expect( (*walk == '/'), 1)) { ------------------ | Branch (883:9): [True: 130, False: 248] ------------------ 884| | /* scan to detect (potential) need for path simplification 885| | * (repeated '/' or "/.") */ 886| 352| do { 887| 352| if (*++walk == '.' || *walk == '/') ------------------ | Branch (887:17): [True: 62, False: 290] | Branch (887:35): [True: 68, False: 222] ------------------ 888| 130| break; 889| 187k| do { ++walk; } while (*walk != '/'); ------------------ | Branch (889:35): [True: 187k, False: 222] ------------------ 890| 222| } while (walk != end); ------------------ | Branch (890:18): [True: 222, False: 0] ------------------ 891| 130| if (__builtin_expect( (walk == end), 1)) { ------------------ | Branch (891:13): [True: 0, False: 130] ------------------ 892| | /* common case: no repeated '/' or "/." */ 893| 0| *end = '\0'; /* overwrite extra '/' added to end of path */ 894| 0| return; 895| 0| } 896| 130| out = walk-1; 897| 130| } 898| 248| else { 899| 248| if (walk[0] == '.' && walk[1] == '/') ------------------ | Branch (899:13): [True: 142, False: 106] | Branch (899:31): [True: 75, False: 67] ------------------ 900| 75| *out = *++walk; 901| 173| else if (walk[0] == '.' && walk[1] == '.' && walk[2] == '/') ------------------ | Branch (901:18): [True: 67, False: 106] | Branch (901:36): [True: 16, False: 51] | Branch (901:54): [True: 1, False: 15] ------------------ 902| 1| *out = *(walk += 2); 903| 172| else { 904| 9.02M| while (*++walk != '/') ; ------------------ | Branch (904:20): [True: 9.02M, False: 172] ------------------ 905| 172| out = walk; 906| 172| } 907| 248| ++walk; 908| 248| } 909| | 910| 9.67k| while (walk <= end) { ------------------ | Branch (910:12): [True: 9.44k, False: 227] ------------------ 911| | /* previous char is '/' at this point (or start of string w/o '/') */ 912| 9.44k| if (__builtin_expect( (walk[0] == '/'), 0)) { ------------------ | Branch (912:13): [True: 1.58k, False: 7.86k] ------------------ 913| | /* skip repeated '/' (e.g. "///" -> "/") */ 914| 1.58k| if (++walk < end) ------------------ | Branch (914:17): [True: 1.53k, False: 56] ------------------ 915| 1.53k| continue; 916| 56| else { 917| 56| ++out; 918| 56| break; 919| 56| } 920| 1.58k| } 921| 7.86k| else if (__builtin_expect( (walk[0] == '.'), 0)) { ------------------ | Branch (921:18): [True: 3.26k, False: 4.59k] ------------------ 922| | /* handle "./" and "../" */ 923| 3.26k| if (walk[1] == '.' && walk[2] == '/') { ------------------ | Branch (923:17): [True: 1.33k, False: 1.93k] | Branch (923:35): [True: 622, False: 716] ------------------ 924| | /* handle "../" */ 925| 1.68M| while (out > b->ptr && *--out != '/') ; ------------------ | Branch (925:24): [True: 1.68M, False: 421] | Branch (925:40): [True: 1.68M, False: 201] ------------------ 926| 622| *out = '/'; /*(in case path had not started with '/')*/ 927| 622| if ((walk += 3) >= end) { ------------------ | Branch (927:21): [True: 35, False: 587] ------------------ 928| 35| ++out; 929| 35| break; 930| 35| } 931| 587| else 932| 587| continue; 933| 622| } 934| 2.64k| else if (walk[1] == '/') { ------------------ | Branch (934:22): [True: 725, False: 1.92k] ------------------ 935| | /* handle "./" */ 936| 725| if ((walk += 2) >= end) { ------------------ | Branch (936:21): [True: 60, False: 665] ------------------ 937| 60| ++out; 938| 60| break; 939| 60| } 940| 665| continue; 941| 725| } 942| 1.92k| else { 943| | /* accept "." if not part of "../" or "./" */ 944| 1.92k| *++out = '.'; 945| 1.92k| ++walk; 946| 1.92k| } 947| 3.26k| } 948| | 949| 16.5M| while ((*++out = *walk++) != '/') ; ------------------ | Branch (949:16): [True: 16.5M, False: 6.51k] ------------------ 950| 6.51k| } 951| 378| *out = *end = '\0'; /* overwrite extra '/' added to end of path */ 952| 378| b->used = (out - b->ptr) + 1; 953| | /*buffer_truncate(b, out - b->ptr);*/ 954| 378|} buffer.c:buffer_alloc_replace: 71| 3.63k|static char* buffer_alloc_replace(buffer * const restrict b, const size_t size) { 72| | /*(discard old data so realloc() does not copy)*/ 73| 3.63k| if (NULL != b->ptr) { ------------------ | Branch (73:9): [True: 523, False: 3.10k] ------------------ 74| 523| free(b->ptr); 75| 523| b->ptr = NULL; 76| 523| } 77| | /*(note: if size larger than one lshift, use size instead of power-2)*/ 78| 3.63k| const size_t bsize2x = (b->size & ~1uL) << 1; 79| 3.63k| return buffer_realloc(b, bsize2x > size ? bsize2x-1 : size); ------------------ | Branch (79:30): [True: 285, False: 3.34k] ------------------ 80| 3.63k|} buffer.c:buffer_realloc: 49| 3.63k|static char* buffer_realloc(buffer * const restrict b, const size_t len) { 50| 3.63k| #define BUFFER_PIECE_SIZE 64uL /*(must be power-of-2)*/ 51| 3.63k| size_t sz = (len + 1 + BUFFER_PIECE_SIZE-1) & ~(BUFFER_PIECE_SIZE-1); ------------------ | | 50| 3.63k| #define BUFFER_PIECE_SIZE 64uL /*(must be power-of-2)*/ ------------------ size_t sz = (len + 1 + BUFFER_PIECE_SIZE-1) & ~(BUFFER_PIECE_SIZE-1); ------------------ | | 50| 3.63k| #define BUFFER_PIECE_SIZE 64uL /*(must be power-of-2)*/ ------------------ 52| 3.63k| force_assert(sz > len); ------------------ | | 448| 3.63k|#define force_assert(x) ck_assert(x) | | ------------------ | | | | 115| 3.63k| do { if (!(x)) ck_assert_failed(__FILE__, __LINE__, #x); } while (0) | | | | ------------------ | | | | | Branch (115:18): [True: 0, False: 3.63k] | | | | | Branch (115:75): [Folded - Ignored] | | | | ------------------ | | ------------------ ------------------ 53| 3.63k| if ((sz & (sz-1)) && sz < INT_MAX) {/* not power-2; huge val not expected */ ------------------ | Branch (53:9): [True: 1.26k, False: 2.36k] | Branch (53:26): [True: 1.26k, False: 0] ------------------ 54| | /*(optimizer should recognize this and use ffs or clz or equivalent)*/ 55| 1.26k| const size_t psz = sz; 56| 8.64k| for (sz = 256; sz < psz; sz <<= 1) ; ------------------ | Branch (56:24): [True: 7.37k, False: 1.26k] ------------------ 57| 1.26k| } 58| 3.63k| sz |= 1; /*(extra +1 for '\0' when needed buffer size is exact power-2)*/ 59| | 60| 3.63k| b->size = sz; 61| 3.63k| b->ptr = realloc(b->ptr, sz); 62| | 63| 3.63k| force_assert(NULL != b->ptr); ------------------ | | 448| 3.63k|#define force_assert(x) ck_assert(x) | | ------------------ | | | | 115| 3.63k| do { if (!(x)) ck_assert_failed(__FILE__, __LINE__, #x); } while (0) | | | | ------------------ | | | | | Branch (115:18): [True: 0, False: 3.63k] | | | | | Branch (115:75): [Folded - Ignored] | | | | ------------------ | | ------------------ ------------------ 64| 3.63k| return b->ptr; 65| 3.63k|} burl.c:buffer_truncate: 349| 413|static inline void buffer_truncate(buffer *b, uint32_t len) { 350| 413| b->ptr[len] = '\0'; /* b->ptr must exist; use buffer_blank() for trunc 0 */ 351| 413| b->used = len + 1; 352| 413|} burl.c:light_isalnum: 232| 9.68k|static inline int light_isalnum(int c) { 233| 9.68k| return light_isdigit(c) || light_isalpha(c); ------------------ | Branch (233:9): [True: 483, False: 9.20k] | Branch (233:29): [True: 121, False: 9.08k] ------------------ 234| 9.68k|} burl.c:light_isdigit: 214| 9.68k|static inline int light_isdigit(int c) { 215| 9.68k| return ((uint32_t)c-'0' <= '9'-'0'); 216| 9.68k|} burl.c:light_isalpha: 226| 9.20k|static inline int light_isalpha(int c) { 227| 9.20k| return (((uint32_t)c | 0x20)-'a' <= 'z'-'a'); 228| 9.20k|} burl.c:buffer_clen: 323| 6.46k|static inline uint32_t buffer_clen (const buffer *b) { 324| 6.46k| return b->used - (0 != b->used); 325| 6.46k|} buffer.c:buffer_clen: 323| 1.73k|static inline uint32_t buffer_clen (const buffer *b) { 324| 1.73k| return b->used - (0 != b->used); 325| 1.73k|} buffer.c:buffer_is_blank: 315| 378|static inline int buffer_is_blank(const buffer *b) { 316| 378| return b->used < 2; /* buffer_is_blank() || buffer_is_unset() */ 317| 378|} burl_normalize: 338| 1.73k|{ 339| 1.73k| int qs; 340| | 341| | #if defined(_WIN32) || defined(__CYGWIN__) 342| | /* Windows and Cygwin treat '\\' as '/' if '\\' is present in path; 343| | * convert to '/' for consistency before percent-encoding 344| | * normalization which will convert '\\' to "%5C" in the URL. 345| | * (Clients still should not be sending '\\' unencoded in requests.) */ 346| | if (flags & HTTP_PARSEOPT_URL_NORMALIZE_PATH_BACKSLASH_TRANS) { 347| | for (char *p = b->ptr; *p != '?' && *p != '\0'; ++p) { 348| | if (*p == '\\') *p = '/'; 349| | } 350| | } 351| | #endif 352| | 353| 1.73k| qs = (flags & HTTP_PARSEOPT_URL_NORMALIZE_REQUIRED) ------------------ | Branch (353:10): [True: 906, False: 831] ------------------ 354| 1.73k| ? burl_normalize_basic_required(b, t) 355| 1.73k| : burl_normalize_basic_unreserved(b, t); 356| 1.73k| if (-2 == qs) { ------------------ | Branch (356:9): [True: 551, False: 1.18k] ------------------ 357| 551| if (flags & HTTP_PARSEOPT_URL_NORMALIZE_INVALID_UTF8_REJECT) return -2; ------------------ | Branch (357:13): [True: 108, False: 443] ------------------ 358| 443| qs = burl_scan_qmark(b); 359| 443| } 360| | 361| 1.62k| if (flags & HTTP_PARSEOPT_URL_NORMALIZE_CTRLS_REJECT) { ------------------ | Branch (361:9): [True: 400, False: 1.22k] ------------------ 362| 400| if (burl_contains_ctrls(b)) return -2; ------------------ | Branch (362:13): [True: 97, False: 303] ------------------ 363| 400| } 364| | 365| 1.53k| if (flags & (HTTP_PARSEOPT_URL_NORMALIZE_PATH_2F_DECODE ------------------ | Branch (365:9): [True: 1.25k, False: 279] ------------------ 366| 1.53k| |HTTP_PARSEOPT_URL_NORMALIZE_PATH_2F_REJECT)) { 367| 1.25k| qs = burl_normalize_2F_to_slash(b, qs, flags); 368| 1.25k| if (-2 == qs) return -2; ------------------ | Branch (368:13): [True: 5, False: 1.24k] ------------------ 369| 1.25k| } 370| | 371| 1.52k| if (flags & (HTTP_PARSEOPT_URL_NORMALIZE_PATH_DOTSEG_REMOVE ------------------ | Branch (371:9): [True: 1.22k, False: 307] ------------------ 372| 1.52k| |HTTP_PARSEOPT_URL_NORMALIZE_PATH_DOTSEG_REJECT)) { 373| 1.22k| qs = burl_normalize_path(b, t, qs, flags); 374| 1.22k| if (-2 == qs) return -2; ------------------ | Branch (374:13): [True: 22, False: 1.19k] ------------------ 375| 1.22k| } 376| | 377| 1.50k| if (flags & HTTP_PARSEOPT_URL_NORMALIZE_QUERY_20_PLUS) { ------------------ | Branch (377:9): [True: 897, False: 608] ------------------ 378| 897| if (qs >= 0) burl_normalize_qs20_to_plus(b, qs); ------------------ | Branch (378:13): [True: 397, False: 500] ------------------ 379| 897| } 380| | 381| 1.50k| return qs; 382| 1.52k|} burl.c:burl_normalize_basic_required: 175| 906|{ 176| 906| const unsigned char * const s = (unsigned char *)b->ptr; 177| 906| const int used = (int)buffer_clen(b); 178| 906| unsigned int n1, n2, x; 179| 906| int qs = -1; 180| 906| int invalid_utf8 = 0; 181| | 182| 1.86M| for (int i = 0; i < used; ++i) { ------------------ | Branch (182:21): [True: 1.86M, False: 181] ------------------ 183| 1.86M| if (!encoded_chars_http_uri_reqd[s[i]]) { ------------------ | Branch (183:13): [True: 1.85M, False: 9.90k] ------------------ 184| 1.85M| if (s[i] == '?') qs = i; ------------------ | Branch (184:17): [True: 1.79k, False: 1.85M] ------------------ 185| 1.85M| } 186| 9.90k| else if (s[i]=='%' && li_cton(s[i+1], n1) && li_cton(s[i+2], n2) ------------------ | | 43| 19.2k| (((n) = (c) - '0') <= 9 || (((n) = ((c)&0xdf) - 'A') <= 5 ? ((n) += 10) : 0)) | | ------------------ | | | Branch (43:4): [True: 6.56k, False: 2.81k] | | | Branch (43:30): [True: 2.74k, False: 65] | | | Branch (43:31): [True: 2.74k, False: 65] | | ------------------ ------------------ else if (s[i]=='%' && li_cton(s[i+1], n1) && li_cton(s[i+2], n2) ------------------ | | 43| 19.2k| (((n) = (c) - '0') <= 9 || (((n) = ((c)&0xdf) - 'A') <= 5 ? ((n) += 10) : 0)) | | ------------------ | | | Branch (43:4): [True: 3.56k, False: 5.74k] | | | Branch (43:30): [True: 5.68k, False: 62] | | | Branch (43:31): [True: 5.68k, False: 62] | | ------------------ ------------------ | Branch (186:18): [True: 9.37k, False: 527] ------------------ 187| 9.90k| && (encoded_chars_http_uri_reqd[(x = (n1 << 4) | n2)] ------------------ | Branch (187:22): [True: 5.47k, False: 3.77k] ------------------ 188| 9.25k| || (qs < 0 ------------------ | Branch (188:25): [True: 3.70k, False: 71] | Branch (188:26): [True: 1.43k, False: 2.33k] ------------------ 189| 3.77k| ? (x == '/' || x == '?') ------------------ | Branch (189:29): [True: 1.00k, False: 436] | Branch (189:41): [True: 400, False: 36] ------------------ 190| 9.18k| : (x == '&' || x == '=' || x == ';' || x == '+')))) { ------------------ | Branch (190:29): [True: 696, False: 1.64k] | Branch (190:41): [True: 196, False: 1.44k] | Branch (190:53): [True: 369, False: 1.07k] | Branch (190:65): [True: 1.04k, False: 35] ------------------ 191| 9.18k| invalid_utf8 |= li_utf8_invalid_byte(x); ------------------ | | 49| 9.18k|#define li_utf8_invalid_byte(b) light_utf8_invalid_byte(b) | | ------------------ | | | | 260| 9.18k| ( __builtin_expect( ((c) >= 0xF5), 0) \ | | | | ------------------ | | | | | Branch (260:7): [True: 1.02k, False: 8.16k] | | | | ------------------ | | | | 261| 9.18k| || __builtin_expect( (((c)|0x1) == 0xC1), 0) ) | | | | ------------------ | | | | | Branch (261:7): [True: 229, False: 7.93k] | | | | ------------------ | | ------------------ ------------------ 192| 9.18k| if (s[i+1] >= 'a') b->ptr[i+1] &= 0xdf; /* uppercase hex */ ------------------ | Branch (192:17): [True: 1.86k, False: 7.31k] ------------------ 193| 9.18k| if (s[i+2] >= 'a') b->ptr[i+2] &= 0xdf; /* uppercase hex */ ------------------ | Branch (193:17): [True: 3.59k, False: 5.58k] ------------------ 194| 9.18k| i+=2; 195| 9.18k| } 196| 725| else if (s[i] == '#') { /* ignore fragment */ ------------------ | Branch (196:18): [True: 1, False: 724] ------------------ 197| 1| buffer_truncate(b, (size_t)i); 198| 1| break; 199| 1| } 200| 724| else { 201| 724| qs = burl_normalize_basic_required_fix(b, t, i, qs); 202| 724| break; 203| 724| } 204| 1.86M| } 205| | 206| 906| return !invalid_utf8 ? qs : -2; ------------------ | Branch (206:12): [True: 882, False: 24] ------------------ 207| 906|} burl.c:burl_normalize_basic_required_fix: 131| 724|{ 132| 724| int j = i; 133| 724| const int used = (int)buffer_clen(b); 134| 724| const unsigned char * const s = (unsigned char *)b->ptr; 135| 724| unsigned char * const p = 136| 724| (unsigned char *)buffer_string_prepare_copy(t,i+(used-i)*3+1); 137| 724| unsigned int n1, n2; 138| 724| int invalid_utf8 = 0; 139| 724| memcpy(p, s, (size_t)i); 140| 26.8M| for (; i < used; ++i, ++j) { ------------------ | Branch (140:12): [True: 26.8M, False: 715] ------------------ 141| 26.8M| if (!encoded_chars_http_uri_reqd[s[i]]) { ------------------ | Branch (141:13): [True: 1.81M, False: 24.9M] ------------------ 142| 1.81M| p[j] = s[i]; 143| 1.81M| if (__builtin_expect( (s[i] == '?'), 0)) qs = j; ------------------ | Branch (143:17): [True: 2.38k, False: 1.81M] ------------------ 144| 1.81M| } 145| 24.9M| else if (s[i]=='%' && li_cton(s[i+1], n1) && li_cton(s[i+2], n2)) { ------------------ | | 43| 25.0M| (((n) = (c) - '0') <= 9 || (((n) = ((c)&0xdf) - 'A') <= 5 ? ((n) += 10) : 0)) | | ------------------ | | | Branch (43:4): [True: 49.6k, False: 14.9k] | | | Branch (43:30): [True: 4.41k, False: 10.4k] | | | Branch (43:31): [True: 4.41k, False: 10.4k] | | ------------------ ------------------ else if (s[i]=='%' && li_cton(s[i+1], n1) && li_cton(s[i+2], n2)) { ------------------ | | 43| 54.0k| (((n) = (c) - '0') <= 9 || (((n) = ((c)&0xdf) - 'A') <= 5 ? ((n) += 10) : 0)) | | ------------------ | | | Branch (43:4): [True: 15.9k, False: 38.1k] | | | Branch (43:30): [True: 24.2k, False: 13.8k] | | | Branch (43:31): [True: 24.2k, False: 13.8k] | | ------------------ ------------------ | Branch (145:18): [True: 64.5k, False: 24.9M] ------------------ 146| 40.2k| const unsigned int x = (n1 << 4) | n2; 147| 40.2k| if (!encoded_chars_http_uri_reqd[x] ------------------ | Branch (147:17): [True: 29.6k, False: 10.5k] ------------------ 148| 40.2k| && (qs < 0 ------------------ | Branch (148:20): [True: 24.0k, False: 5.64k] | Branch (148:21): [True: 2.09k, False: 27.5k] ------------------ 149| 29.6k| ? (x != '/' && x != '?') ------------------ | Branch (149:24): [True: 1.59k, False: 499] | Branch (149:36): [True: 1.10k, False: 490] ------------------ 150| 29.6k| : (x != '&' && x != '=' && x != ';' && x != '+'))) { ------------------ | Branch (150:24): [True: 25.7k, False: 1.87k] | Branch (150:36): [True: 25.1k, False: 557] | Branch (150:48): [True: 24.4k, False: 734] | Branch (150:60): [True: 22.9k, False: 1.48k] ------------------ 151| 24.0k| p[j] = x; 152| 24.0k| } 153| 16.1k| else { 154| 16.1k| p[j] = '%'; 155| 16.1k| p[++j] = hex_chars_uc[n1]; /*(s[i+1] & 0xdf)*/ 156| 16.1k| p[++j] = hex_chars_uc[n2]; /*(s[i+2] & 0xdf)*/ 157| 16.1k| invalid_utf8 |= li_utf8_invalid_byte(x); ------------------ | | 49| 16.1k|#define li_utf8_invalid_byte(b) light_utf8_invalid_byte(b) | | ------------------ | | | | 260| 16.1k| ( __builtin_expect( ((c) >= 0xF5), 0) \ | | | | ------------------ | | | | | Branch (260:7): [True: 517, False: 15.6k] | | | | ------------------ | | | | 261| 16.1k| || __builtin_expect( (((c)|0x1) == 0xC1), 0) ) | | | | ------------------ | | | | | Branch (261:7): [True: 729, False: 14.9k] | | | | ------------------ | | ------------------ ------------------ 158| 16.1k| } 159| 40.2k| i+=2; 160| 40.2k| } 161| 24.9M| else if (s[i] == '#') break; /* ignore fragment */ ------------------ | Branch (161:18): [True: 9, False: 24.9M] ------------------ 162| 24.9M| else { 163| 24.9M| p[j] = '%'; 164| 24.9M| p[++j] = hex_chars_uc[(s[i] >> 4) & 0xF]; 165| 24.9M| p[++j] = hex_chars_uc[s[i] & 0xF]; 166| 24.9M| invalid_utf8 |= li_utf8_invalid_byte(s[i]); ------------------ | | 49| 24.9M|#define li_utf8_invalid_byte(b) light_utf8_invalid_byte(b) | | ------------------ | | | | 260| 24.9M| ( __builtin_expect( ((c) >= 0xF5), 0) \ | | | | ------------------ | | | | | Branch (260:7): [True: 1.46M, False: 23.4M] | | | | ------------------ | | | | 261| 24.9M| || __builtin_expect( (((c)|0x1) == 0xC1), 0) ) | | | | ------------------ | | | | | Branch (261:7): [True: 322, False: 23.4M] | | | | ------------------ | | ------------------ ------------------ 167| 24.9M| } 168| 26.8M| } 169| 724| buffer_copy_string_len(b, (char *)p, (size_t)j); 170| 724| return !invalid_utf8 ? qs : -2; ------------------ | Branch (170:12): [True: 458, False: 266] ------------------ 171| 724|} burl.c:burl_normalize_basic_unreserved: 99| 831|{ 100| 831| const unsigned char * const s = (unsigned char *)b->ptr; 101| 831| const int used = (int)buffer_clen(b); 102| 831| unsigned int n1, n2, x; 103| 831| int qs = -1; 104| | 105| 196k| for (int i = 0; i < used; ++i) { ------------------ | Branch (105:21): [True: 196k, False: 188] ------------------ 106| 196k| if (!encoded_chars_http_uri_reqd[s[i]]) { ------------------ | Branch (106:13): [True: 192k, False: 4.17k] ------------------ 107| 192k| if (__builtin_expect( (s[i] == '?'), 0) && -1 == qs) qs = i; ------------------ | Branch (107:17): [True: 372, False: 191k] | Branch (107:56): [True: 81, False: 291] ------------------ 108| 192k| } 109| 4.17k| else if (s[i]=='%' && li_cton(s[i+1], n1) && li_cton(s[i+2], n2) ------------------ | | 43| 7.89k| (((n) = (c) - '0') <= 9 || (((n) = ((c)&0xdf) - 'A') <= 5 ? ((n) += 10) : 0)) | | ------------------ | | | Branch (43:4): [True: 1.49k, False: 2.22k] | | | Branch (43:30): [True: 2.14k, False: 79] | | | Branch (43:31): [True: 2.14k, False: 79] | | ------------------ ------------------ else if (s[i]=='%' && li_cton(s[i+1], n1) && li_cton(s[i+2], n2) ------------------ | | 43| 7.81k| (((n) = (c) - '0') <= 9 || (((n) = ((c)&0xdf) - 'A') <= 5 ? ((n) += 10) : 0)) | | ------------------ | | | Branch (43:4): [True: 1.24k, False: 2.40k] | | | Branch (43:30): [True: 2.34k, False: 55] | | | Branch (43:31): [True: 2.34k, False: 55] | | ------------------ ------------------ | Branch (109:18): [True: 3.71k, False: 457] ------------------ 110| 4.17k| && !burl_is_unreserved((x = (n1 << 4) | n2))) { ------------------ | Branch (110:21): [True: 3.53k, False: 52] ------------------ 111| 3.53k| if (li_utf8_invalid_byte(x)) qs = -2; ------------------ | | 49| 3.53k|#define li_utf8_invalid_byte(b) light_utf8_invalid_byte(b) | | ------------------ | | | | 260| 3.53k| ( __builtin_expect( ((c) >= 0xF5), 0) \ | | | | ------------------ | | | | | Branch (260:7): [True: 861, False: 2.67k] | | | | ------------------ | | | | 261| 3.53k| || __builtin_expect( (((c)|0x1) == 0xC1), 0) ) | | | | ------------------ | | | | | Branch (261:7): [True: 341, False: 2.33k] | | | | ------------------ | | ------------------ ------------------ 112| 3.53k| if (s[i+1] >= 'a') b->ptr[i+1] &= 0xdf; /* uppercase hex */ ------------------ | Branch (112:17): [True: 1.50k, False: 2.02k] ------------------ 113| 3.53k| if (s[i+2] >= 'a') b->ptr[i+2] &= 0xdf; /* uppercase hex */ ------------------ | Branch (113:17): [True: 1.53k, False: 2.00k] ------------------ 114| 3.53k| i+=2; 115| 3.53k| } 116| 643| else if (s[i] == '#') { /* ignore fragment */ ------------------ | Branch (116:18): [True: 1, False: 642] ------------------ 117| 1| buffer_truncate(b, (size_t)i); 118| 1| break; 119| 1| } 120| 642| else { 121| 642| qs = burl_normalize_basic_unreserved_fix(b, t, i, qs); 122| 642| break; 123| 642| } 124| 196k| } 125| | 126| 831| return qs; 127| 831|} burl.c:burl_is_unreserved: 53| 9.68k|{ 54| 9.68k| return (light_isalnum(c) || c == '-' || c == '.' || c == '_' || c == '~'); ------------------ | Branch (54:13): [True: 604, False: 9.08k] | Branch (54:33): [True: 384, False: 8.69k] | Branch (54:45): [True: 219, False: 8.47k] | Branch (54:57): [True: 222, False: 8.25k] | Branch (54:69): [True: 9, False: 8.24k] ------------------ 55| 9.68k|} burl.c:burl_normalize_basic_unreserved_fix: 59| 642|{ 60| 642| int j = i; 61| 642| const int used = (int)buffer_clen(b); 62| 642| const unsigned char * const s = (unsigned char *)b->ptr; 63| 642| unsigned char * const p = 64| 642| (unsigned char *)buffer_string_prepare_copy(t,i+(used-i)*3+1); 65| 642| unsigned int n1, n2; 66| 642| memcpy(p, s, (size_t)i); 67| 27.3M| for (; i < used; ++i, ++j) { ------------------ | Branch (67:12): [True: 27.3M, False: 632] ------------------ 68| 27.3M| if (!encoded_chars_http_uri_reqd[s[i]]) { ------------------ | Branch (68:13): [True: 2.01M, False: 25.3M] ------------------ 69| 2.01M| p[j] = s[i]; 70| 2.01M| if (__builtin_expect( (s[i] == '?'), 0) && -1 == qs) qs = j; ------------------ | Branch (70:17): [True: 862, False: 2.01M] | Branch (70:56): [True: 70, False: 792] ------------------ 71| 2.01M| } 72| 25.3M| else if (s[i]=='%' && li_cton(s[i+1], n1) && li_cton(s[i+2], n2)) { ------------------ | | 43| 25.3M| (((n) = (c) - '0') <= 9 || (((n) = ((c)&0xdf) - 'A') <= 5 ? ((n) += 10) : 0)) | | ------------------ | | | Branch (43:4): [True: 4.65k, False: 9.07k] | | | Branch (43:30): [True: 2.58k, False: 6.49k] | | | Branch (43:31): [True: 2.58k, False: 6.49k] | | ------------------ ------------------ else if (s[i]=='%' && li_cton(s[i+1], n1) && li_cton(s[i+2], n2)) { ------------------ | | 43| 7.23k| (((n) = (c) - '0') <= 9 || (((n) = ((c)&0xdf) - 'A') <= 5 ? ((n) += 10) : 0)) | | ------------------ | | | Branch (43:4): [True: 2.15k, False: 5.08k] | | | Branch (43:30): [True: 3.94k, False: 1.13k] | | | Branch (43:31): [True: 3.94k, False: 1.13k] | | ------------------ ------------------ | Branch (72:18): [True: 13.7k, False: 25.3M] ------------------ 73| 6.10k| const unsigned int x = (n1 << 4) | n2; 74| 6.10k| if (burl_is_unreserved(x)) { ------------------ | Branch (74:17): [True: 1.38k, False: 4.71k] ------------------ 75| 1.38k| p[j] = x; 76| 1.38k| } 77| 4.71k| else { 78| 4.71k| p[j] = '%'; 79| 4.71k| p[++j] = hex_chars_uc[n1]; /*(s[i+1] & 0xdf)*/ 80| 4.71k| p[++j] = hex_chars_uc[n2]; /*(s[i+2] & 0xdf)*/ 81| 4.71k| if (li_utf8_invalid_byte(x)) qs = -2; ------------------ | | 49| 4.71k|#define li_utf8_invalid_byte(b) light_utf8_invalid_byte(b) | | ------------------ | | | | 260| 4.71k| ( __builtin_expect( ((c) >= 0xF5), 0) \ | | | | ------------------ | | | | | Branch (260:7): [True: 738, False: 3.97k] | | | | ------------------ | | | | 261| 4.71k| || __builtin_expect( (((c)|0x1) == 0xC1), 0) ) | | | | ------------------ | | | | | Branch (261:7): [True: 351, False: 3.62k] | | | | ------------------ | | ------------------ ------------------ 82| 4.71k| } 83| 6.10k| i+=2; 84| 6.10k| } 85| 25.3M| else if (s[i] == '#') break; /* ignore fragment */ ------------------ | Branch (85:18): [True: 10, False: 25.3M] ------------------ 86| 25.3M| else { 87| 25.3M| p[j] = '%'; 88| 25.3M| p[++j] = hex_chars_uc[(s[i] >> 4) & 0xF]; 89| 25.3M| p[++j] = hex_chars_uc[s[i] & 0xF]; 90| 25.3M| if (li_utf8_invalid_byte(s[i])) qs = -2; ------------------ | | 49| 25.3M|#define li_utf8_invalid_byte(b) light_utf8_invalid_byte(b) | | ------------------ | | | | 260| 25.3M| ( __builtin_expect( ((c) >= 0xF5), 0) \ | | | | ------------------ | | | | | Branch (260:7): [True: 2.54M, False: 22.8M] | | | | ------------------ | | | | 261| 25.3M| || __builtin_expect( (((c)|0x1) == 0xC1), 0) ) | | | | ------------------ | | | | | Branch (261:7): [True: 1.52k, False: 22.8M] | | | | ------------------ | | ------------------ ------------------ 91| 25.3M| } 92| 27.3M| } 93| 642| buffer_copy_string_len(b, (char *)p, (size_t)j); 94| 642| return qs; 95| 642|} burl.c:burl_scan_qmark: 331| 443|static int burl_scan_qmark (const buffer * const b) { 332| 443| const char * const qmark = strchr(b->ptr, '?'); 333| 443| return qmark ? (int)(qmark - b->ptr) : -1; ------------------ | Branch (333:12): [True: 189, False: 254] ------------------ 334| 443|} burl.c:burl_contains_ctrls: 211| 400|{ 212| 400| const char * const s = b->ptr; 213| 400| const int used = (int)buffer_clen(b); 214| 18.2M| for (int i = 0; i < used; ++i) { ------------------ | Branch (214:21): [True: 18.2M, False: 303] ------------------ 215| 18.2M| if (s[i] == '%' && (s[i+1] < '2' || (s[i+1] == '7' && s[i+2] == 'F'))) ------------------ | Branch (215:13): [True: 5.99M, False: 12.2M] | Branch (215:29): [True: 90, False: 5.99M] | Branch (215:46): [True: 1.14k, False: 5.99M] | Branch (215:63): [True: 7, False: 1.14k] ------------------ 216| 97| return 1; 217| 18.2M| } 218| 303| return 0; 219| 400|} burl.c:burl_normalize_2F_to_slash: 276| 1.25k|{ 277| | /*("%2F" must already have been uppercased during normalization)*/ 278| 1.25k| const char * const s = b->ptr; 279| 1.25k| const int used = qs < 0 ? (int)buffer_clen(b) : qs; ------------------ | Branch (279:22): [True: 797, False: 456] ------------------ 280| 68.5M| for (int i = 0; i < used; ++i) { ------------------ | Branch (280:21): [True: 68.5M, False: 1.07k] ------------------ 281| 68.5M| if (s[i] == '%' && s[i+1] == '2' && s[i+2] == 'F') { ------------------ | Branch (281:13): [True: 22.2M, False: 46.3M] | Branch (281:28): [True: 789k, False: 21.4M] | Branch (281:45): [True: 180, False: 789k] ------------------ 282| 180| return (flags & HTTP_PARSEOPT_URL_NORMALIZE_PATH_2F_DECODE) ------------------ | Branch (282:20): [True: 175, False: 5] ------------------ 283| 180| ? burl_normalize_2F_to_slash_fix(b, qs, i) 284| 180| : -2; /*(flags & HTTP_PARSEOPT_URL_NORMALIZE_PATH_2F_REJECT)*/ 285| 180| } 286| 68.5M| } 287| 1.07k| return qs; 288| 1.25k|} burl.c:burl_normalize_2F_to_slash_fix: 252| 175|{ 253| 175| char * const s = b->ptr; 254| 175| const int blen = (int)buffer_clen(b); 255| 175| const int used = qs < 0 ? blen : qs; ------------------ | Branch (255:22): [True: 125, False: 50] ------------------ 256| 175| int j = i; 257| 20.4M| for (; i < used; ++i, ++j) { ------------------ | Branch (257:12): [True: 20.4M, False: 175] ------------------ 258| 20.4M| s[j] = s[i]; 259| 20.4M| if (s[i] == '%' && s[i+1] == '2' && s[i+2] == 'F') { ------------------ | Branch (259:13): [True: 6.74M, False: 13.7M] | Branch (259:28): [True: 6.84k, False: 6.73M] | Branch (259:45): [True: 2.11k, False: 4.72k] ------------------ 260| 2.11k| s[j] = '/'; 261| 2.11k| i+=2; 262| 2.11k| } 263| 20.4M| } 264| 175| if (qs >= 0) { ------------------ | Branch (264:9): [True: 50, False: 125] ------------------ 265| 50| const int qslen = blen - qs; 266| 50| memmove(s+j, s+qs, (size_t)qslen); 267| 50| qs = j; 268| 50| j += qslen; 269| 50| } 270| 175| buffer_truncate(b, j); 271| 175| return qs; 272| 175|} burl.c:burl_normalize_path: 292| 1.22k|{ 293| 1.22k| const unsigned char * const s = (unsigned char *)b->ptr; 294| 1.22k| const int used = (int)buffer_clen(b); 295| 1.22k| int path_simplify = 0; 296| 7.27k| for (int i = 0, len = qs < 0 ? used : qs; i < len; ++i) { ------------------ | Branch (296:27): [True: 785, False: 435] | Branch (296:47): [True: 6.45k, False: 820] ------------------ 297| 6.45k| if (s[i] == '.' && (s[i+1] != '.' || ++i) ------------------ | Branch (297:13): [True: 1.69k, False: 4.75k] | Branch (297:29): [True: 836, False: 858] | Branch (297:46): [True: 858, False: 0] ------------------ 298| 6.45k| && (s[i+1] == '/' || s[i+1] == '?' || s[i+1] == '\0')) { ------------------ | Branch (298:17): [True: 89, False: 1.60k] | Branch (298:34): [True: 86, False: 1.51k] | Branch (298:51): [True: 76, False: 1.44k] ------------------ 299| 251| path_simplify = 1; 300| 251| break; 301| 251| } 302| 68.3M| while (i < len && s[i] != '/') ++i; ------------------ | Branch (302:16): [True: 68.3M, False: 669] | Branch (302:27): [True: 68.3M, False: 5.53k] ------------------ 303| 6.20k| if (s[i] == '/' && s[i+1] == '/') { /*(s[len] != '/')*/ ------------------ | Branch (303:13): [True: 5.53k, False: 669] | Branch (303:28): [True: 149, False: 5.38k] ------------------ 304| 149| path_simplify = 1; 305| 149| break; 306| 149| } 307| 6.20k| } 308| | 309| 1.22k| if (path_simplify) { ------------------ | Branch (309:9): [True: 400, False: 820] ------------------ 310| 400| if (flags & HTTP_PARSEOPT_URL_NORMALIZE_PATH_DOTSEG_REJECT) return -2; ------------------ | Branch (310:13): [True: 22, False: 378] ------------------ 311| 378| if (qs >= 0) { ------------------ | Branch (311:13): [True: 135, False: 243] ------------------ 312| 135| buffer_copy_string_len(t, b->ptr+qs, used - qs); 313| 135| buffer_truncate(b, qs); 314| 135| } 315| | 316| 378| buffer_path_simplify(b); 317| | 318| 378| if (qs >= 0) { ------------------ | Branch (318:13): [True: 135, False: 243] ------------------ 319| 135| qs = (int)buffer_clen(b); 320| 135| buffer_append_string_len(b, BUF_PTR_LEN(t)); ------------------ | | 297| 135|#define BUF_PTR_LEN(x) (x)->ptr, buffer_clen(x) ------------------ 321| 135| } 322| 378| } 323| | 324| 1.19k| return qs; 325| 1.22k|} burl.c:burl_normalize_qs20_to_plus: 239| 397|{ 240| 397| const char * const s = b->ptr; 241| 397| const int used = qs < 0 ? 0 : (int)buffer_clen(b); ------------------ | Branch (241:22): [True: 0, False: 397] ------------------ 242| 397| int i; 243| 397| if (qs < 0) return; ------------------ | Branch (243:9): [True: 0, False: 397] ------------------ 244| 11.7M| for (i = qs+1; i < used; ++i) { ------------------ | Branch (244:20): [True: 11.7M, False: 296] ------------------ 245| 11.7M| if (s[i] == '%' && s[i+1] == '2' && s[i+2] == '0') break; ------------------ | Branch (245:13): [True: 3.91M, False: 7.82M] | Branch (245:28): [True: 2.42k, False: 3.90M] | Branch (245:45): [True: 101, False: 2.32k] ------------------ 246| 11.7M| } 247| 397| if (i != used) burl_normalize_qs20_to_plus_fix(b, i); ------------------ | Branch (247:9): [True: 101, False: 296] ------------------ 248| 397|} burl.c:burl_normalize_qs20_to_plus_fix: 223| 101|{ 224| 101| char * const s = b->ptr; 225| 101| const int used = (int)buffer_clen(b); 226| 101| int j = i; 227| 10.2M| for (; i < used; ++i, ++j) { ------------------ | Branch (227:12): [True: 10.2M, False: 101] ------------------ 228| 10.2M| s[j] = s[i]; 229| 10.2M| if (s[i] == '%' && s[i+1] == '2' && s[i+2] == '0') { ------------------ | Branch (229:13): [True: 3.16M, False: 7.07M] | Branch (229:28): [True: 63.5k, False: 3.10M] | Branch (229:45): [True: 42.5k, False: 21.0k] ------------------ 230| 42.5k| s[j] = '+'; 231| 42.5k| i+=2; 232| 42.5k| } 233| 10.2M| } 234| 101| buffer_truncate(b, j); 235| 101|}