run_burl_normalize: 26| 1.65k| size_t in_len) { 27| 1.65k| int qs; 28| 1.65k| buffer_copy_string_len(psrc, in, in_len); 29| 1.65k| qs = burl_normalize(psrc, ptmp, flags); 30| 1.65k|} LLVMFuzzerTestOneInput: 32| 1.66k|int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) { 33| 1.66k| if (size <= 4) { ------------------ | Branch (33:9): [True: 4, False: 1.65k] ------------------ 34| 4| return 0; 35| 4| } 36| 1.65k| int flags = ((int*)data)[0]; 37| 1.65k| data += 4; 38| 1.65k| size -= 4; 39| 1.65k| char *new_str = (char *)malloc(size+1); 40| 1.65k| if (new_str == NULL){ ------------------ | Branch (40:9): [True: 0, False: 1.65k] ------------------ 41| 0| return 0; 42| 0| } 43| 1.65k| memcpy(new_str, data, size); 44| 1.65k| new_str[size] = '\0'; 45| | 46| | /* main fuzzer entrypoint for library */ 47| 1.65k| buffer *psrc = buffer_init(); 48| 1.65k| buffer *ptmp = buffer_init(); 49| 1.65k| run_burl_normalize(psrc, ptmp, flags, __LINE__, new_str, size); 50| 1.65k| buffer_urldecode_path(psrc); 51| | 52| 1.65k| buffer_free(psrc); 53| 1.65k| buffer_free(ptmp); 54| 1.65k| free(new_str); 55| 1.65k| return 0; 56| 1.65k|} buffer_init: 14| 3.31k|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| 3.31k| buffer * const b = calloc(1, sizeof(*b)); 20| 3.31k| force_assert(b); ------------------ | | 448| 3.31k|#define force_assert(x) ck_assert(x) | | ------------------ | | | | 115| 3.31k| do { if (!(x)) ck_assert_failed(__FILE__, __LINE__, #x); } while (0) | | | | ------------------ | | | | | Branch (115:18): [True: 0, False: 3.31k] | | | | | Branch (115:75): [Folded, False: 3.31k] | | | | ------------------ | | ------------------ ------------------ 21| 3.31k| return b; 22| 3.31k| #endif 23| 3.31k|} buffer_free: 25| 3.31k|void buffer_free(buffer *b) { 26| 3.31k| if (NULL == b) return; ------------------ | Branch (26:6): [True: 0, False: 3.31k] ------------------ 27| 3.31k| free(b->ptr); 28| 3.31k| free(b); 29| 3.31k|} buffer_string_prepare_copy: 82| 1.25k|char* buffer_string_prepare_copy(buffer * const b, const size_t size) { 83| 1.25k| 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| 1.25k| return (size < b->size) ------------------ | Branch (87:12): [True: 0, False: 1.25k] ------------------ 88| 1.25k| ? b->ptr 89| 1.25k| : buffer_alloc_replace(b, size); 90| 1.25k|} buffer_extend: 127| 152|{ 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| 152| const uint32_t len = b->used ? b->used-1 : 0; ------------------ | Branch (136:26): [True: 152, False: 0] ------------------ 137| 152| char * const s = (b->size - len >= x + 1) ------------------ | Branch (137:22): [True: 152, False: 0] ------------------ 138| 152| ? b->ptr + len 139| 152| : buffer_string_prepare_append_resize(b, x); 140| 152| b->used = len+x+1; 141| 152| #endif 142| 152| s[x] = '\0'; 143| 152| return s; 144| 152|} buffer_copy_string_len: 172| 3.06k|void buffer_copy_string_len(buffer * const restrict b, const char * const restrict s, const size_t len) { 173| 3.06k| b->used = len + 1; 174| 3.06k| char * const restrict d = (len < b->size) ------------------ | Branch (174:31): [True: 935, False: 2.12k] ------------------ 175| 3.06k| ? b->ptr 176| 3.06k| : buffer_alloc_replace(b, len); 177| 3.06k| d[len] = '\0'; 178| 3.06k| memcpy(d, s, len); 179| 3.06k|} buffer_append_string_len: 198| 152|void buffer_append_string_len(buffer * const restrict b, const char * const restrict s, const size_t len) { 199| 152| memcpy(buffer_extend(b, len), s, len); 200| 152|} hex2int: 386| 97.5M|char hex2int(unsigned char hex) { 387| 97.5M| unsigned char n; 388| 97.5M| return li_cton(hex,n) ? (char)n : 0xFF; ------------------ | | 381| 97.5M| (((n) = (c) - '0') <= 9 || (((n) = ((c)&0xdf) - 'A') <= 5 ? ((n) += 10) : 0)) | | ------------------ | | | Branch (381:4): [True: 86.6M, False: 10.8M] | | | Branch (381:30): [True: 10.8M, False: 0] | | | Branch (381:31): [True: 10.8M, False: 0] | | ------------------ ------------------ 389| 97.5M|} buffer_urldecode_path: 792| 1.65k|void buffer_urldecode_path(buffer * const b) { 793| 1.65k| const size_t len = buffer_clen(b); 794| 1.65k| char *src = len ? memchr(b->ptr, '%', len) : NULL; ------------------ | Branch (794:17): [True: 1.65k, False: 6] ------------------ 795| 1.65k| if (NULL == src) return; ------------------ | Branch (795:9): [True: 359, False: 1.29k] ------------------ 796| | 797| 1.29k| char *dst = src; 798| 48.7M| do { 799| | /* *src == '%' */ 800| 48.7M| unsigned char high = ((unsigned char *)src)[1]; 801| 48.7M| unsigned char low = high ? hex2int(((unsigned char *)src)[2]) : 0xFF; ------------------ | Branch (801:29): [True: 48.7M, False: 0] ------------------ 802| 48.7M| if (0xFF != (high = hex2int(high)) && 0xFF != low) { ------------------ | Branch (802:13): [True: 48.7M, False: 0] | Branch (802:47): [True: 48.7M, False: 0] ------------------ 803| 48.7M| high = (high << 4) | low; /* map ctrls to '_' */ 804| 48.7M| *dst = (high >= 32 && high != 127) ? high : '_'; ------------------ | Branch (804:21): [True: 8.97M, False: 39.7M] | Branch (804:35): [True: 7.94M, False: 1.02M] ------------------ 805| 48.7M| src += 2; 806| 48.7M| } /* else ignore this '%'; leave as-is and move on */ 807| | 808| 51.8M| while ((*++dst = *++src) != '%' && *src) ; ------------------ | Branch (808:16): [True: 3.08M, False: 48.7M] | Branch (808:44): [True: 3.08M, False: 1.29k] ------------------ 809| 48.7M| } while (*src); ------------------ | Branch (809:14): [True: 48.7M, False: 1.29k] ------------------ 810| 1.29k| b->used = (dst - b->ptr) + 1; 811| 1.29k|} buffer_path_simplify: 864| 388|{ 865| 388| char *out = b->ptr; 866| 388| char * const end = b->ptr + b->used - 1; 867| | 868| 388| if (__builtin_expect( (buffer_is_blank(b)), 0)) { ------------------ | Branch (868:9): [True: 0, False: 388] ------------------ 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| 388| *end = '/'; /*(end of path modified to avoid need to check '\0')*/ 881| | 882| 388| char *walk = out; 883| 388| if (__builtin_expect( (*walk == '/'), 1)) { ------------------ | Branch (883:9): [True: 135, False: 253] ------------------ 884| | /* scan to detect (potential) need for path simplification 885| | * (repeated '/' or "/.") */ 886| 494| do { 887| 494| if (*++walk == '.' || *walk == '/') ------------------ | Branch (887:17): [True: 77, False: 417] | Branch (887:35): [True: 58, False: 359] ------------------ 888| 135| break; 889| 1.05M| do { ++walk; } while (*walk != '/'); ------------------ | Branch (889:35): [True: 1.05M, False: 359] ------------------ 890| 359| } while (walk != end); ------------------ | Branch (890:18): [True: 359, False: 0] ------------------ 891| 135| if (__builtin_expect( (walk == end), 1)) { ------------------ | Branch (891:13): [True: 0, False: 135] ------------------ 892| | /* common case: no repeated '/' or "/." */ 893| 0| *end = '\0'; /* overwrite extra '/' added to end of path */ 894| 0| return; 895| 0| } 896| 135| out = walk-1; 897| 135| } 898| 253| else { 899| 253| if (walk[0] == '.' && walk[1] == '/') ------------------ | Branch (899:13): [True: 140, False: 113] | Branch (899:31): [True: 74, False: 66] ------------------ 900| 74| *out = *++walk; 901| 179| else if (walk[0] == '.' && walk[1] == '.' && walk[2] == '/') ------------------ | Branch (901:18): [True: 66, False: 113] | Branch (901:36): [True: 19, False: 47] | Branch (901:54): [True: 3, False: 16] ------------------ 902| 3| *out = *(walk += 2); 903| 176| else { 904| 12.8M| while (*++walk != '/') ; ------------------ | Branch (904:20): [True: 12.8M, False: 176] ------------------ 905| 176| out = walk; 906| 176| } 907| 253| ++walk; 908| 253| } 909| | 910| 5.59k| while (walk <= end) { ------------------ | Branch (910:12): [True: 5.36k, False: 227] ------------------ 911| | /* previous char is '/' at this point (or start of string w/o '/') */ 912| 5.36k| if (__builtin_expect( (walk[0] == '/'), 0)) { ------------------ | Branch (912:13): [True: 1.12k, False: 4.24k] ------------------ 913| | /* skip repeated '/' (e.g. "///" -> "/") */ 914| 1.12k| if (++walk < end) ------------------ | Branch (914:17): [True: 1.06k, False: 60] ------------------ 915| 1.06k| continue; 916| 60| else { 917| 60| ++out; 918| 60| break; 919| 60| } 920| 1.12k| } 921| 4.24k| else if (__builtin_expect( (walk[0] == '.'), 0)) { ------------------ | Branch (921:18): [True: 3.27k, False: 969] ------------------ 922| | /* handle "./" and "../" */ 923| 3.27k| if (walk[1] == '.' && walk[2] == '/') { ------------------ | Branch (923:17): [True: 2.11k, False: 1.15k] | Branch (923:35): [True: 659, False: 1.46k] ------------------ 924| | /* handle "../" */ 925| 6.72M| while (out > b->ptr && *--out != '/') ; ------------------ | Branch (925:24): [True: 6.72M, False: 543] | Branch (925:40): [True: 6.72M, False: 116] ------------------ 926| 659| *out = '/'; /*(in case path had not started with '/')*/ 927| 659| if ((walk += 3) >= end) { ------------------ | Branch (927:21): [True: 47, False: 612] ------------------ 928| 47| ++out; 929| 47| break; 930| 47| } 931| 612| else 932| 612| continue; 933| 659| } 934| 2.61k| else if (walk[1] == '/') { ------------------ | Branch (934:22): [True: 619, False: 1.99k] ------------------ 935| | /* handle "./" */ 936| 619| if ((walk += 2) >= end) { ------------------ | Branch (936:21): [True: 54, False: 565] ------------------ 937| 54| ++out; 938| 54| break; 939| 54| } 940| 565| continue; 941| 619| } 942| 1.99k| else { 943| | /* accept "." if not part of "../" or "./" */ 944| 1.99k| *++out = '.'; 945| 1.99k| ++walk; 946| 1.99k| } 947| 3.27k| } 948| | 949| 23.8M| while ((*++out = *walk++) != '/') ; ------------------ | Branch (949:16): [True: 23.8M, False: 2.96k] ------------------ 950| 2.96k| } 951| 388| *out = *end = '\0'; /* overwrite extra '/' added to end of path */ 952| 388| b->used = (out - b->ptr) + 1; 953| | /*buffer_truncate(b, out - b->ptr);*/ 954| 388|} buffer.c:buffer_alloc_replace: 71| 3.38k|static char* buffer_alloc_replace(buffer * const restrict b, const size_t size) { 72| | /*(discard old data so realloc() does not copy)*/ 73| 3.38k| if (NULL != b->ptr) { ------------------ | Branch (73:9): [True: 464, False: 2.92k] ------------------ 74| 464| free(b->ptr); 75| 464| b->ptr = NULL; 76| 464| } 77| | /*(note: if size larger than one lshift, use size instead of power-2)*/ 78| 3.38k| const size_t bsize2x = (b->size & ~1uL) << 1; 79| 3.38k| return buffer_realloc(b, bsize2x > size ? bsize2x-1 : size); ------------------ | Branch (79:30): [True: 222, False: 3.16k] ------------------ 80| 3.38k|} buffer.c:buffer_realloc: 49| 3.38k|static char* buffer_realloc(buffer * const restrict b, const size_t len) { 50| 3.38k| #define BUFFER_PIECE_SIZE 64uL /*(must be power-of-2)*/ 51| 3.38k| size_t sz = (len + 1 + BUFFER_PIECE_SIZE-1) & ~(BUFFER_PIECE_SIZE-1); ------------------ | | 50| 3.38k| #define BUFFER_PIECE_SIZE 64uL /*(must be power-of-2)*/ ------------------ size_t sz = (len + 1 + BUFFER_PIECE_SIZE-1) & ~(BUFFER_PIECE_SIZE-1); ------------------ | | 50| 3.38k| #define BUFFER_PIECE_SIZE 64uL /*(must be power-of-2)*/ ------------------ 52| 3.38k| force_assert(sz > len); ------------------ | | 448| 3.38k|#define force_assert(x) ck_assert(x) | | ------------------ | | | | 115| 3.38k| do { if (!(x)) ck_assert_failed(__FILE__, __LINE__, #x); } while (0) | | | | ------------------ | | | | | Branch (115:18): [True: 0, False: 3.38k] | | | | | Branch (115:75): [Folded, False: 3.38k] | | | | ------------------ | | ------------------ ------------------ 53| 3.38k| if ((sz & (sz-1)) && sz < INT_MAX) {/* not power-2; huge val not expected */ ------------------ | Branch (53:9): [True: 1.15k, False: 2.22k] | Branch (53:26): [True: 1.15k, False: 0] ------------------ 54| | /*(optimizer should recognize this and use ffs or clz or equivalent)*/ 55| 1.15k| const size_t psz = sz; 56| 8.02k| for (sz = 256; sz < psz; sz <<= 1) ; ------------------ | Branch (56:24): [True: 6.86k, False: 1.15k] ------------------ 57| 1.15k| } 58| 3.38k| sz |= 1; /*(extra +1 for '\0' when needed buffer size is exact power-2)*/ 59| | 60| 3.38k| b->size = sz; 61| 3.38k| b->ptr = realloc(b->ptr, sz); 62| | 63| 3.38k| force_assert(NULL != b->ptr); ------------------ | | 448| 3.38k|#define force_assert(x) ck_assert(x) | | ------------------ | | | | 115| 3.38k| do { if (!(x)) ck_assert_failed(__FILE__, __LINE__, #x); } while (0) | | | | ------------------ | | | | | Branch (115:18): [True: 0, False: 3.38k] | | | | | Branch (115:75): [Folded, False: 3.38k] | | | | ------------------ | | ------------------ ------------------ 64| 3.38k| return b->ptr; 65| 3.38k|} burl.c:buffer_truncate: 349| 425|static inline void buffer_truncate(buffer *b, uint32_t len) { 350| 425| b->ptr[len] = '\0'; /* b->ptr must exist; use buffer_blank() for trunc 0 */ 351| 425| b->used = len + 1; 352| 425|} burl.c:light_isalnum: 232| 13.0k|static inline int light_isalnum(int c) { 233| 13.0k| return light_isdigit(c) || light_isalpha(c); ------------------ | Branch (233:9): [True: 295, False: 12.7k] | Branch (233:29): [True: 515, False: 12.2k] ------------------ 234| 13.0k|} burl.c:light_isdigit: 214| 13.0k|static inline int light_isdigit(int c) { 215| 13.0k| return ((uint32_t)c-'0' <= '9'-'0'); 216| 13.0k|} burl.c:light_isalpha: 226| 12.7k|static inline int light_isalpha(int c) { 227| 12.7k| return (((uint32_t)c | 0x20)-'a' <= 'z'-'a'); 228| 12.7k|} burl.c:buffer_clen: 323| 6.12k|static inline uint32_t buffer_clen (const buffer *b) { 324| 6.12k| return b->used - (0 != b->used); 325| 6.12k|} buffer.c:buffer_clen: 323| 1.65k|static inline uint32_t buffer_clen (const buffer *b) { 324| 1.65k| return b->used - (0 != b->used); 325| 1.65k|} buffer.c:buffer_is_blank: 315| 388|static inline int buffer_is_blank(const buffer *b) { 316| 388| return b->used < 2; /* buffer_is_blank() || buffer_is_unset() */ 317| 388|} burl_normalize: 338| 1.65k|{ 339| 1.65k| 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| 1.65k| qs = (flags & HTTP_PARSEOPT_URL_NORMALIZE_REQUIRED) ------------------ | Branch (353:10): [True: 889, False: 767] ------------------ 354| 1.65k| ? burl_normalize_basic_required(b, t) 355| 1.65k| : burl_normalize_basic_unreserved(b, t); 356| 1.65k| if (-2 == qs) { ------------------ | Branch (356:9): [True: 458, False: 1.19k] ------------------ 357| 458| if (flags & HTTP_PARSEOPT_URL_NORMALIZE_INVALID_UTF8_REJECT) return -2; ------------------ | Branch (357:13): [True: 112, False: 346] ------------------ 358| 346| qs = burl_scan_qmark(b); 359| 346| } 360| | 361| 1.54k| if (flags & HTTP_PARSEOPT_URL_NORMALIZE_CTRLS_REJECT) { ------------------ | Branch (361:9): [True: 342, False: 1.20k] ------------------ 362| 342| if (burl_contains_ctrls(b)) return -2; ------------------ | Branch (362:13): [True: 101, False: 241] ------------------ 363| 342| } 364| | 365| 1.44k| if (flags & (HTTP_PARSEOPT_URL_NORMALIZE_PATH_2F_DECODE ------------------ | Branch (365:9): [True: 1.18k, False: 259] ------------------ 366| 1.44k| |HTTP_PARSEOPT_URL_NORMALIZE_PATH_2F_REJECT)) { 367| 1.18k| qs = burl_normalize_2F_to_slash(b, qs, flags); 368| 1.18k| if (-2 == qs) return -2; ------------------ | Branch (368:13): [True: 11, False: 1.17k] ------------------ 369| 1.18k| } 370| | 371| 1.43k| if (flags & (HTTP_PARSEOPT_URL_NORMALIZE_PATH_DOTSEG_REMOVE ------------------ | Branch (371:9): [True: 1.15k, False: 274] ------------------ 372| 1.43k| |HTTP_PARSEOPT_URL_NORMALIZE_PATH_DOTSEG_REJECT)) { 373| 1.15k| qs = burl_normalize_path(b, t, qs, flags); 374| 1.15k| if (-2 == qs) return -2; ------------------ | Branch (374:13): [True: 16, False: 1.14k] ------------------ 375| 1.15k| } 376| | 377| 1.41k| if (flags & HTTP_PARSEOPT_URL_NORMALIZE_QUERY_20_PLUS) { ------------------ | Branch (377:9): [True: 845, False: 571] ------------------ 378| 845| if (qs >= 0) burl_normalize_qs20_to_plus(b, qs); ------------------ | Branch (378:13): [True: 405, False: 440] ------------------ 379| 845| } 380| | 381| 1.41k| return qs; 382| 1.43k|} burl.c:burl_normalize_basic_required: 175| 889|{ 176| 889| const unsigned char * const s = (unsigned char *)b->ptr; 177| 889| const int used = (int)buffer_clen(b); 178| 889| unsigned int n1, n2, x; 179| 889| int qs = -1; 180| 889| int invalid_utf8 = 0; 181| | 182| 847k| for (int i = 0; i < used; ++i) { ------------------ | Branch (182:21): [True: 847k, False: 211] ------------------ 183| 847k| if (!encoded_chars_http_uri_reqd[s[i]]) { ------------------ | Branch (183:13): [True: 780k, False: 66.6k] ------------------ 184| 780k| if (s[i] == '?') qs = i; ------------------ | Branch (184:17): [True: 54.2k, False: 726k] ------------------ 185| 780k| } 186| 66.6k| else if (s[i]=='%' && li_cton(s[i+1], n1) && li_cton(s[i+2], n2) ------------------ | | 43| 132k| (((n) = (c) - '0') <= 9 || (((n) = ((c)&0xdf) - 'A') <= 5 ? ((n) += 10) : 0)) | | ------------------ | | | Branch (43:4): [True: 62.6k, False: 3.54k] | | | Branch (43:30): [True: 3.46k, False: 78] | | | Branch (43:31): [True: 3.46k, False: 78] | | ------------------ ------------------ else if (s[i]=='%' && li_cton(s[i+1], n1) && li_cton(s[i+2], n2) ------------------ | | 43| 132k| (((n) = (c) - '0') <= 9 || (((n) = ((c)&0xdf) - 'A') <= 5 ? ((n) += 10) : 0)) | | ------------------ | | | Branch (43:4): [True: 39.2k, False: 26.9k] | | | Branch (43:30): [True: 26.8k, False: 56] | | | Branch (43:31): [True: 26.8k, False: 56] | | ------------------ ------------------ | Branch (186:18): [True: 66.2k, False: 479] ------------------ 187| 66.0k| && (encoded_chars_http_uri_reqd[(x = (n1 << 4) | n2)] ------------------ | Branch (187:22): [True: 11.0k, False: 54.9k] ------------------ 188| 54.9k| || (qs < 0 ------------------ | Branch (188:25): [True: 54.9k, False: 65] | Branch (188:26): [True: 862, False: 54.1k] ------------------ 189| 54.9k| ? (x == '/' || x == '?') ------------------ | Branch (189:29): [True: 620, False: 242] | Branch (189:41): [True: 212, False: 30] ------------------ 190| 66.0k| : (x == '&' || x == '=' || x == ';' || x == '+')))) { ------------------ | Branch (190:29): [True: 30.0k, False: 24.0k] | Branch (190:41): [True: 14.6k, False: 9.41k] | Branch (190:53): [True: 3.07k, False: 6.34k] | Branch (190:65): [True: 6.30k, False: 35] ------------------ 191| 66.0k| invalid_utf8 |= li_utf8_invalid_byte(x); ------------------ | | 49| 66.0k|#define li_utf8_invalid_byte(b) light_utf8_invalid_byte(b) | | ------------------ | | | | 260| 66.0k| ( __builtin_expect( ((c) >= 0xF5), 0) \ | | | | ------------------ | | | | | Branch (260:7): [True: 324, False: 65.6k] | | | | ------------------ | | | | 261| 66.0k| || __builtin_expect( (((c)|0x1) == 0xC1), 0) ) | | | | ------------------ | | | | | Branch (261:7): [True: 290, False: 65.3k] | | | | ------------------ | | ------------------ ------------------ 192| 66.0k| if (s[i+1] >= 'a') b->ptr[i+1] &= 0xdf; /* uppercase hex */ ------------------ | Branch (192:17): [True: 715, False: 65.2k] ------------------ 193| 66.0k| if (s[i+2] >= 'a') b->ptr[i+2] &= 0xdf; /* uppercase hex */ ------------------ | Branch (193:17): [True: 9.96k, False: 56.0k] ------------------ 194| 66.0k| i+=2; 195| 66.0k| } 196| 678| else if (s[i] == '#') { /* ignore fragment */ ------------------ | Branch (196:18): [True: 1, False: 677] ------------------ 197| 1| buffer_truncate(b, (size_t)i); 198| 1| break; 199| 1| } 200| 677| else { 201| 677| qs = burl_normalize_basic_required_fix(b, t, i, qs); 202| 677| break; 203| 677| } 204| 847k| } 205| | 206| 889| return !invalid_utf8 ? qs : -2; ------------------ | Branch (206:12): [True: 865, False: 24] ------------------ 207| 889|} burl.c:burl_normalize_basic_required_fix: 131| 677|{ 132| 677| int j = i; 133| 677| const int used = (int)buffer_clen(b); 134| 677| const unsigned char * const s = (unsigned char *)b->ptr; 135| 677| unsigned char * const p = 136| 677| (unsigned char *)buffer_string_prepare_copy(t,i+(used-i)*3+1); 137| 677| unsigned int n1, n2; 138| 677| int invalid_utf8 = 0; 139| 677| memcpy(p, s, (size_t)i); 140| 26.6M| for (; i < used; ++i, ++j) { ------------------ | Branch (140:12): [True: 26.6M, False: 665] ------------------ 141| 26.6M| if (!encoded_chars_http_uri_reqd[s[i]]) { ------------------ | Branch (141:13): [True: 573k, False: 26.0M] ------------------ 142| 573k| p[j] = s[i]; 143| 573k| if (__builtin_expect( (s[i] == '?'), 0)) qs = j; ------------------ | Branch (143:17): [True: 9.06k, False: 564k] ------------------ 144| 573k| } 145| 26.0M| else if (s[i]=='%' && li_cton(s[i+1], n1) && li_cton(s[i+2], n2)) { ------------------ | | 43| 26.0M| (((n) = (c) - '0') <= 9 || (((n) = ((c)&0xdf) - 'A') <= 5 ? ((n) += 10) : 0)) | | ------------------ | | | Branch (43:4): [True: 15.6k, False: 19.5k] | | | Branch (43:30): [True: 3.83k, False: 15.7k] | | | Branch (43:31): [True: 3.83k, False: 15.7k] | | ------------------ ------------------ else if (s[i]=='%' && li_cton(s[i+1], n1) && li_cton(s[i+2], n2)) { ------------------ | | 43| 19.5k| (((n) = (c) - '0') <= 9 || (((n) = ((c)&0xdf) - 'A') <= 5 ? ((n) += 10) : 0)) | | ------------------ | | | Branch (43:4): [True: 9.46k, False: 10.0k] | | | Branch (43:30): [True: 8.20k, False: 1.83k] | | | Branch (43:31): [True: 8.20k, False: 1.83k] | | ------------------ ------------------ | Branch (145:18): [True: 35.2k, False: 26.0M] ------------------ 146| 17.6k| const unsigned int x = (n1 << 4) | n2; 147| 17.6k| if (!encoded_chars_http_uri_reqd[x] ------------------ | Branch (147:17): [True: 12.2k, False: 5.44k] ------------------ 148| 12.2k| && (qs < 0 ------------------ | Branch (148:20): [True: 2.04k, False: 10.1k] | Branch (148:21): [True: 1.36k, False: 10.8k] ------------------ 149| 12.2k| ? (x != '/' && x != '?') ------------------ | Branch (149:24): [True: 909, False: 457] | Branch (149:36): [True: 663, False: 246] ------------------ 150| 12.2k| : (x != '&' && x != '=' && x != ';' && x != '+'))) { ------------------ | Branch (150:24): [True: 5.93k, False: 4.92k] | Branch (150:36): [True: 3.49k, False: 2.44k] | Branch (150:48): [True: 2.72k, False: 769] | Branch (150:60): [True: 1.38k, False: 1.33k] ------------------ 151| 2.04k| p[j] = x; 152| 2.04k| } 153| 15.6k| else { 154| 15.6k| p[j] = '%'; 155| 15.6k| p[++j] = hex_chars_uc[n1]; /*(s[i+1] & 0xdf)*/ 156| 15.6k| p[++j] = hex_chars_uc[n2]; /*(s[i+2] & 0xdf)*/ 157| 15.6k| invalid_utf8 |= li_utf8_invalid_byte(x); ------------------ | | 49| 15.6k|#define li_utf8_invalid_byte(b) light_utf8_invalid_byte(b) | | ------------------ | | | | 260| 15.6k| ( __builtin_expect( ((c) >= 0xF5), 0) \ | | | | ------------------ | | | | | Branch (260:7): [True: 1.21k, False: 14.4k] | | | | ------------------ | | | | 261| 15.6k| || __builtin_expect( (((c)|0x1) == 0xC1), 0) ) | | | | ------------------ | | | | | Branch (261:7): [True: 247, False: 14.1k] | | | | ------------------ | | ------------------ ------------------ 158| 15.6k| } 159| 17.6k| i+=2; 160| 17.6k| } 161| 26.0M| else if (s[i] == '#') break; /* ignore fragment */ ------------------ | Branch (161:18): [True: 12, False: 26.0M] ------------------ 162| 26.0M| else { 163| 26.0M| p[j] = '%'; 164| 26.0M| p[++j] = hex_chars_uc[(s[i] >> 4) & 0xF]; 165| 26.0M| p[++j] = hex_chars_uc[s[i] & 0xF]; 166| 26.0M| invalid_utf8 |= li_utf8_invalid_byte(s[i]); ------------------ | | 49| 26.0M|#define li_utf8_invalid_byte(b) light_utf8_invalid_byte(b) | | ------------------ | | | | 260| 26.0M| ( __builtin_expect( ((c) >= 0xF5), 0) \ | | | | ------------------ | | | | | Branch (260:7): [True: 1.87M, False: 24.1M] | | | | ------------------ | | | | 261| 26.0M| || __builtin_expect( (((c)|0x1) == 0xC1), 0) ) | | | | ------------------ | | | | | Branch (261:7): [True: 790, False: 24.1M] | | | | ------------------ | | ------------------ ------------------ 167| 26.0M| } 168| 26.6M| } 169| 677| buffer_copy_string_len(b, (char *)p, (size_t)j); 170| 677| return !invalid_utf8 ? qs : -2; ------------------ | Branch (170:12): [True: 448, False: 229] ------------------ 171| 677|} burl.c:burl_normalize_basic_unreserved: 99| 767|{ 100| 767| const unsigned char * const s = (unsigned char *)b->ptr; 101| 767| const int used = (int)buffer_clen(b); 102| 767| unsigned int n1, n2, x; 103| 767| int qs = -1; 104| | 105| 590k| for (int i = 0; i < used; ++i) { ------------------ | Branch (105:21): [True: 590k, False: 187] ------------------ 106| 590k| if (!encoded_chars_http_uri_reqd[s[i]]) { ------------------ | Branch (106:13): [True: 587k, False: 3.05k] ------------------ 107| 587k| if (__builtin_expect( (s[i] == '?'), 0) && -1 == qs) qs = i; ------------------ | Branch (107:17): [True: 1.12k, False: 586k] | Branch (107:56): [True: 78, False: 1.05k] ------------------ 108| 587k| } 109| 3.05k| else if (s[i]=='%' && li_cton(s[i+1], n1) && li_cton(s[i+2], n2) ------------------ | | 43| 5.69k| (((n) = (c) - '0') <= 9 || (((n) = ((c)&0xdf) - 'A') <= 5 ? ((n) += 10) : 0)) | | ------------------ | | | Branch (43:4): [True: 1.71k, False: 928] | | | Branch (43:30): [True: 865, False: 63] | | | Branch (43:31): [True: 865, False: 63] | | ------------------ ------------------ else if (s[i]=='%' && li_cton(s[i+1], n1) && li_cton(s[i+2], n2) ------------------ | | 43| 5.63k| (((n) = (c) - '0') <= 9 || (((n) = ((c)&0xdf) - 'A') <= 5 ? ((n) += 10) : 0)) | | ------------------ | | | Branch (43:4): [True: 1.34k, False: 1.23k] | | | Branch (43:30): [True: 1.18k, False: 50] | | | Branch (43:31): [True: 1.18k, False: 50] | | ------------------ ------------------ | Branch (109:18): [True: 2.64k, False: 414] ------------------ 110| 2.52k| && !burl_is_unreserved((x = (n1 << 4) | n2))) { ------------------ | Branch (110:21): [True: 2.47k, False: 53] ------------------ 111| 2.47k| if (li_utf8_invalid_byte(x)) qs = -2; ------------------ | | 49| 2.47k|#define li_utf8_invalid_byte(b) light_utf8_invalid_byte(b) | | ------------------ | | | | 260| 2.47k| ( __builtin_expect( ((c) >= 0xF5), 0) \ | | | | ------------------ | | | | | Branch (260:7): [True: 262, False: 2.21k] | | | | ------------------ | | | | 261| 2.47k| || __builtin_expect( (((c)|0x1) == 0xC1), 0) ) | | | | ------------------ | | | | | Branch (261:7): [True: 447, False: 1.76k] | | | | ------------------ | | ------------------ ------------------ 112| 2.47k| if (s[i+1] >= 'a') b->ptr[i+1] &= 0xdf; /* uppercase hex */ ------------------ | Branch (112:17): [True: 739, False: 1.73k] ------------------ 113| 2.47k| if (s[i+2] >= 'a') b->ptr[i+2] &= 0xdf; /* uppercase hex */ ------------------ | Branch (113:17): [True: 717, False: 1.75k] ------------------ 114| 2.47k| i+=2; 115| 2.47k| } 116| 580| else if (s[i] == '#') { /* ignore fragment */ ------------------ | Branch (116:18): [True: 1, False: 579] ------------------ 117| 1| buffer_truncate(b, (size_t)i); 118| 1| break; 119| 1| } 120| 579| else { 121| 579| qs = burl_normalize_basic_unreserved_fix(b, t, i, qs); 122| 579| break; 123| 579| } 124| 590k| } 125| | 126| 767| return qs; 127| 767|} burl.c:burl_is_unreserved: 53| 13.0k|{ 54| 13.0k| return (light_isalnum(c) || c == '-' || c == '.' || c == '_' || c == '~'); ------------------ | Branch (54:13): [True: 810, False: 12.2k] | Branch (54:33): [True: 302, False: 11.9k] | Branch (54:45): [True: 657, False: 11.3k] | Branch (54:57): [True: 249, False: 11.0k] | Branch (54:69): [True: 211, False: 10.8k] ------------------ 55| 13.0k|} burl.c:burl_normalize_basic_unreserved_fix: 59| 579|{ 60| 579| int j = i; 61| 579| const int used = (int)buffer_clen(b); 62| 579| const unsigned char * const s = (unsigned char *)b->ptr; 63| 579| unsigned char * const p = 64| 579| (unsigned char *)buffer_string_prepare_copy(t,i+(used-i)*3+1); 65| 579| unsigned int n1, n2; 66| 579| memcpy(p, s, (size_t)i); 67| 26.2M| for (; i < used; ++i, ++j) { ------------------ | Branch (67:12): [True: 26.2M, False: 573] ------------------ 68| 26.2M| if (!encoded_chars_http_uri_reqd[s[i]]) { ------------------ | Branch (68:13): [True: 1.33M, False: 24.8M] ------------------ 69| 1.33M| p[j] = s[i]; 70| 1.33M| if (__builtin_expect( (s[i] == '?'), 0) && -1 == qs) qs = j; ------------------ | Branch (70:17): [True: 1.98k, False: 1.33M] | Branch (70:56): [True: 75, False: 1.91k] ------------------ 71| 1.33M| } 72| 24.8M| else if (s[i]=='%' && li_cton(s[i+1], n1) && li_cton(s[i+2], n2)) { ------------------ | | 43| 24.8M| (((n) = (c) - '0') <= 9 || (((n) = ((c)&0xdf) - 'A') <= 5 ? ((n) += 10) : 0)) | | ------------------ | | | Branch (43:4): [True: 6.65k, False: 8.62k] | | | Branch (43:30): [True: 5.70k, False: 2.92k] | | | Branch (43:31): [True: 5.70k, False: 2.92k] | | ------------------ ------------------ else if (s[i]=='%' && li_cton(s[i+1], n1) && li_cton(s[i+2], n2)) { ------------------ | | 43| 12.3k| (((n) = (c) - '0') <= 9 || (((n) = ((c)&0xdf) - 'A') <= 5 ? ((n) += 10) : 0)) | | ------------------ | | | Branch (43:4): [True: 5.14k, False: 7.20k] | | | Branch (43:30): [True: 5.40k, False: 1.80k] | | | Branch (43:31): [True: 5.40k, False: 1.80k] | | ------------------ ------------------ | Branch (72:18): [True: 15.2k, False: 24.8M] ------------------ 73| 10.5k| const unsigned int x = (n1 << 4) | n2; 74| 10.5k| if (burl_is_unreserved(x)) { ------------------ | Branch (74:17): [True: 2.17k, False: 8.37k] ------------------ 75| 2.17k| p[j] = x; 76| 2.17k| } 77| 8.37k| else { 78| 8.37k| p[j] = '%'; 79| 8.37k| p[++j] = hex_chars_uc[n1]; /*(s[i+1] & 0xdf)*/ 80| 8.37k| p[++j] = hex_chars_uc[n2]; /*(s[i+2] & 0xdf)*/ 81| 8.37k| if (li_utf8_invalid_byte(x)) qs = -2; ------------------ | | 49| 8.37k|#define li_utf8_invalid_byte(b) light_utf8_invalid_byte(b) | | ------------------ | | | | 260| 8.37k| ( __builtin_expect( ((c) >= 0xF5), 0) \ | | | | ------------------ | | | | | Branch (260:7): [True: 3.08k, False: 5.29k] | | | | ------------------ | | | | 261| 8.37k| || __builtin_expect( (((c)|0x1) == 0xC1), 0) ) | | | | ------------------ | | | | | Branch (261:7): [True: 231, False: 5.06k] | | | | ------------------ | | ------------------ ------------------ 82| 8.37k| } 83| 10.5k| i+=2; 84| 10.5k| } 85| 24.8M| else if (s[i] == '#') break; /* ignore fragment */ ------------------ | Branch (85:18): [True: 6, False: 24.8M] ------------------ 86| 24.8M| else { 87| 24.8M| p[j] = '%'; 88| 24.8M| p[++j] = hex_chars_uc[(s[i] >> 4) & 0xF]; 89| 24.8M| p[++j] = hex_chars_uc[s[i] & 0xF]; 90| 24.8M| if (li_utf8_invalid_byte(s[i])) qs = -2; ------------------ | | 49| 24.8M|#define li_utf8_invalid_byte(b) light_utf8_invalid_byte(b) | | ------------------ | | | | 260| 24.8M| ( __builtin_expect( ((c) >= 0xF5), 0) \ | | | | ------------------ | | | | | Branch (260:7): [True: 59.9k, False: 24.8M] | | | | ------------------ | | | | 261| 24.8M| || __builtin_expect( (((c)|0x1) == 0xC1), 0) ) | | | | ------------------ | | | | | Branch (261:7): [True: 1.34k, False: 24.8M] | | | | ------------------ | | ------------------ ------------------ 91| 24.8M| } 92| 26.2M| } 93| 579| buffer_copy_string_len(b, (char *)p, (size_t)j); 94| 579| return qs; 95| 579|} burl.c:burl_scan_qmark: 331| 346|static int burl_scan_qmark (const buffer * const b) { 332| 346| const char * const qmark = strchr(b->ptr, '?'); 333| 346| return qmark ? (int)(qmark - b->ptr) : -1; ------------------ | Branch (333:12): [True: 170, False: 176] ------------------ 334| 346|} burl.c:burl_contains_ctrls: 211| 342|{ 212| 342| const char * const s = b->ptr; 213| 342| const int used = (int)buffer_clen(b); 214| 7.48M| for (int i = 0; i < used; ++i) { ------------------ | Branch (214:21): [True: 7.48M, False: 241] ------------------ 215| 7.48M| if (s[i] == '%' && (s[i+1] < '2' || (s[i+1] == '7' && s[i+2] == 'F'))) ------------------ | Branch (215:13): [True: 2.16M, False: 5.32M] | Branch (215:29): [True: 95, False: 2.16M] | Branch (215:46): [True: 707, False: 2.16M] | Branch (215:63): [True: 6, False: 701] ------------------ 216| 101| return 1; 217| 7.48M| } 218| 241| return 0; 219| 342|} burl.c:burl_normalize_2F_to_slash: 276| 1.18k|{ 277| | /*("%2F" must already have been uppercased during normalization)*/ 278| 1.18k| const char * const s = b->ptr; 279| 1.18k| const int used = qs < 0 ? (int)buffer_clen(b) : qs; ------------------ | Branch (279:22): [True: 737, False: 447] ------------------ 280| 78.4M| for (int i = 0; i < used; ++i) { ------------------ | Branch (280:21): [True: 78.4M, False: 1.00k] ------------------ 281| 78.4M| if (s[i] == '%' && s[i+1] == '2' && s[i+2] == 'F') { ------------------ | Branch (281:13): [True: 25.8M, False: 52.5M] | Branch (281:28): [True: 41.4k, False: 25.8M] | Branch (281:45): [True: 181, False: 41.2k] ------------------ 282| 181| return (flags & HTTP_PARSEOPT_URL_NORMALIZE_PATH_2F_DECODE) ------------------ | Branch (282:20): [True: 170, False: 11] ------------------ 283| 181| ? burl_normalize_2F_to_slash_fix(b, qs, i) 284| 181| : -2; /*(flags & HTTP_PARSEOPT_URL_NORMALIZE_PATH_2F_REJECT)*/ 285| 181| } 286| 78.4M| } 287| 1.00k| return qs; 288| 1.18k|} burl.c:burl_normalize_2F_to_slash_fix: 252| 170|{ 253| 170| char * const s = b->ptr; 254| 170| const int blen = (int)buffer_clen(b); 255| 170| const int used = qs < 0 ? blen : qs; ------------------ | Branch (255:22): [True: 108, False: 62] ------------------ 256| 170| int j = i; 257| 11.9M| for (; i < used; ++i, ++j) { ------------------ | Branch (257:12): [True: 11.9M, False: 170] ------------------ 258| 11.9M| s[j] = s[i]; 259| 11.9M| if (s[i] == '%' && s[i+1] == '2' && s[i+2] == 'F') { ------------------ | Branch (259:13): [True: 3.96M, False: 7.94M] | Branch (259:28): [True: 18.0k, False: 3.94M] | Branch (259:45): [True: 1.02k, False: 17.0k] ------------------ 260| 1.02k| s[j] = '/'; 261| 1.02k| i+=2; 262| 1.02k| } 263| 11.9M| } 264| 170| if (qs >= 0) { ------------------ | Branch (264:9): [True: 62, False: 108] ------------------ 265| 62| const int qslen = blen - qs; 266| 62| memmove(s+j, s+qs, (size_t)qslen); 267| 62| qs = j; 268| 62| j += qslen; 269| 62| } 270| 170| buffer_truncate(b, j); 271| 170| return qs; 272| 170|} burl.c:burl_normalize_path: 292| 1.15k|{ 293| 1.15k| const unsigned char * const s = (unsigned char *)b->ptr; 294| 1.15k| const int used = (int)buffer_clen(b); 295| 1.15k| int path_simplify = 0; 296| 4.48k| for (int i = 0, len = qs < 0 ? used : qs; i < len; ++i) { ------------------ | Branch (296:27): [True: 720, False: 438] | Branch (296:47): [True: 3.72k, False: 754] ------------------ 297| 3.72k| if (s[i] == '.' && (s[i+1] != '.' || ++i) ------------------ | Branch (297:13): [True: 1.49k, False: 2.23k] | Branch (297:29): [True: 849, False: 641] | Branch (297:46): [True: 641, False: 0] ------------------ 298| 1.49k| && (s[i+1] == '/' || s[i+1] == '?' || s[i+1] == '\0')) { ------------------ | Branch (298:17): [True: 87, False: 1.40k] | Branch (298:34): [True: 86, False: 1.31k] | Branch (298:51): [True: 80, False: 1.23k] ------------------ 299| 253| path_simplify = 1; 300| 253| break; 301| 253| } 302| 73.5M| while (i < len && s[i] != '/') ++i; ------------------ | Branch (302:16): [True: 73.5M, False: 622] | Branch (302:27): [True: 73.5M, False: 2.85k] ------------------ 303| 3.47k| if (s[i] == '/' && s[i+1] == '/') { /*(s[len] != '/')*/ ------------------ | Branch (303:13): [True: 2.85k, False: 622] | Branch (303:28): [True: 151, False: 2.70k] ------------------ 304| 151| path_simplify = 1; 305| 151| break; 306| 151| } 307| 3.47k| } 308| | 309| 1.15k| if (path_simplify) { ------------------ | Branch (309:9): [True: 404, False: 754] ------------------ 310| 404| if (flags & HTTP_PARSEOPT_URL_NORMALIZE_PATH_DOTSEG_REJECT) return -2; ------------------ | Branch (310:13): [True: 16, False: 388] ------------------ 311| 388| if (qs >= 0) { ------------------ | Branch (311:13): [True: 152, False: 236] ------------------ 312| 152| buffer_copy_string_len(t, b->ptr+qs, used - qs); 313| 152| buffer_truncate(b, qs); 314| 152| } 315| | 316| 388| buffer_path_simplify(b); 317| | 318| 388| if (qs >= 0) { ------------------ | Branch (318:13): [True: 152, False: 236] ------------------ 319| 152| qs = (int)buffer_clen(b); 320| 152| buffer_append_string_len(b, BUF_PTR_LEN(t)); ------------------ | | 297| 152|#define BUF_PTR_LEN(x) (x)->ptr, buffer_clen(x) ------------------ 321| 152| } 322| 388| } 323| | 324| 1.14k| return qs; 325| 1.15k|} burl.c:burl_normalize_qs20_to_plus: 239| 405|{ 240| 405| const char * const s = b->ptr; 241| 405| const int used = qs < 0 ? 0 : (int)buffer_clen(b); ------------------ | Branch (241:22): [True: 0, False: 405] ------------------ 242| 405| int i; 243| 405| if (qs < 0) return; ------------------ | Branch (243:9): [True: 0, False: 405] ------------------ 244| 5.87M| for (i = qs+1; i < used; ++i) { ------------------ | Branch (244:20): [True: 5.87M, False: 304] ------------------ 245| 5.87M| if (s[i] == '%' && s[i+1] == '2' && s[i+2] == '0') break; ------------------ | Branch (245:13): [True: 1.94M, False: 3.92M] | Branch (245:28): [True: 1.85k, False: 1.94M] | Branch (245:45): [True: 101, False: 1.75k] ------------------ 246| 5.87M| } 247| 405| if (i != used) burl_normalize_qs20_to_plus_fix(b, i); ------------------ | Branch (247:9): [True: 101, False: 304] ------------------ 248| 405|} 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| 4.10M| for (; i < used; ++i, ++j) { ------------------ | Branch (227:12): [True: 4.10M, False: 101] ------------------ 228| 4.10M| s[j] = s[i]; 229| 4.10M| if (s[i] == '%' && s[i+1] == '2' && s[i+2] == '0') { ------------------ | Branch (229:13): [True: 1.35M, False: 2.74M] | Branch (229:28): [True: 1.82k, False: 1.35M] | Branch (229:45): [True: 1.01k, False: 810] ------------------ 230| 1.01k| s[j] = '+'; 231| 1.01k| i+=2; 232| 1.01k| } 233| 4.10M| } 234| 101| buffer_truncate(b, j); 235| 101|}