| 6 |
and semantics are as close as possible to those of the Perl 5 language. |
and semantics are as close as possible to those of the Perl 5 language. |
| 7 |
|
|
| 8 |
Written by Philip Hazel |
Written by Philip Hazel |
| 9 |
Copyright (c) 1997-2008 University of Cambridge |
Copyright (c) 1997-2009 University of Cambridge |
| 10 |
|
|
| 11 |
----------------------------------------------------------------------------- |
----------------------------------------------------------------------------- |
| 12 |
Redistribution and use in source and binary forms, with or without |
Redistribution and use in source and binary forms, with or without |
| 322 |
|
|
| 323 |
/* Function arguments that may change */ |
/* Function arguments that may change */ |
| 324 |
|
|
| 325 |
const uschar *Xeptr; |
USPTR Xeptr; |
| 326 |
const uschar *Xecode; |
const uschar *Xecode; |
| 327 |
const uschar *Xmstart; |
USPTR Xmstart; |
| 328 |
int Xoffset_top; |
int Xoffset_top; |
| 329 |
long int Xims; |
long int Xims; |
| 330 |
eptrblock *Xeptrb; |
eptrblock *Xeptrb; |
| 333 |
|
|
| 334 |
/* Function local variables */ |
/* Function local variables */ |
| 335 |
|
|
| 336 |
const uschar *Xcallpat; |
USPTR Xcallpat; |
| 337 |
const uschar *Xcharptr; |
#ifdef SUPPORT_UTF8 |
| 338 |
const uschar *Xdata; |
USPTR Xcharptr; |
| 339 |
const uschar *Xnext; |
#endif |
| 340 |
const uschar *Xpp; |
USPTR Xdata; |
| 341 |
const uschar *Xprev; |
USPTR Xnext; |
| 342 |
const uschar *Xsaved_eptr; |
USPTR Xpp; |
| 343 |
|
USPTR Xprev; |
| 344 |
|
USPTR Xsaved_eptr; |
| 345 |
|
|
| 346 |
recursion_info Xnew_recursive; |
recursion_info Xnew_recursive; |
| 347 |
|
|
| 362 |
uschar Xocchars[8]; |
uschar Xocchars[8]; |
| 363 |
#endif |
#endif |
| 364 |
|
|
| 365 |
|
int Xcodelink; |
| 366 |
int Xctype; |
int Xctype; |
| 367 |
unsigned int Xfc; |
unsigned int Xfc; |
| 368 |
int Xfi; |
int Xfi; |
| 398 |
|
|
| 399 |
/* This function is called recursively in many circumstances. Whenever it |
/* This function is called recursively in many circumstances. Whenever it |
| 400 |
returns a negative (error) response, the outer incarnation must also return the |
returns a negative (error) response, the outer incarnation must also return the |
| 401 |
same response. |
same response. */ |
| 402 |
|
|
| 403 |
|
/* These macros pack up tests that are used for partial matching, and which |
| 404 |
|
appears several times in the code. We set the "hit end" flag if the pointer is |
| 405 |
|
at the end of the subject and also past the start of the subject (i.e. |
| 406 |
|
something has been matched). For hard partial matching, we then return |
| 407 |
|
immediately. The second one is used when we already know we are past the end of |
| 408 |
|
the subject. */ |
| 409 |
|
|
| 410 |
|
#define CHECK_PARTIAL()\ |
| 411 |
|
if (md->partial != 0 && eptr >= md->end_subject && eptr > mstart)\ |
| 412 |
|
{\ |
| 413 |
|
md->hitend = TRUE;\ |
| 414 |
|
if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);\ |
| 415 |
|
} |
| 416 |
|
|
| 417 |
|
#define SCHECK_PARTIAL()\ |
| 418 |
|
if (md->partial != 0 && eptr > mstart)\ |
| 419 |
|
{\ |
| 420 |
|
md->hitend = TRUE;\ |
| 421 |
|
if (md->partial > 1) RRETURN(PCRE_ERROR_PARTIAL);\ |
| 422 |
|
} |
| 423 |
|
|
| 424 |
|
|
| 425 |
Performance note: It might be tempting to extract commonly used fields from the |
/* Performance note: It might be tempting to extract commonly used fields from |
| 426 |
md structure (e.g. utf8, end_subject) into individual variables to improve |
the md structure (e.g. utf8, end_subject) into individual variables to improve |
| 427 |
performance. Tests using gcc on a SPARC disproved this; in the first case, it |
performance. Tests using gcc on a SPARC disproved this; in the first case, it |
| 428 |
made performance worse. |
made performance worse. |
| 429 |
|
|
| 450 |
*/ |
*/ |
| 451 |
|
|
| 452 |
static int |
static int |
| 453 |
match(REGISTER USPTR eptr, REGISTER const uschar *ecode, const uschar *mstart, |
match(REGISTER USPTR eptr, REGISTER const uschar *ecode, USPTR mstart, |
| 454 |
int offset_top, match_data *md, unsigned long int ims, eptrblock *eptrb, |
int offset_top, match_data *md, unsigned long int ims, eptrblock *eptrb, |
| 455 |
int flags, unsigned int rdepth) |
int flags, unsigned int rdepth) |
| 456 |
{ |
{ |
| 464 |
register BOOL utf8; /* Local copy of UTF-8 flag for speed */ |
register BOOL utf8; /* Local copy of UTF-8 flag for speed */ |
| 465 |
|
|
| 466 |
BOOL minimize, possessive; /* Quantifier options */ |
BOOL minimize, possessive; /* Quantifier options */ |
| 467 |
|
int condcode; |
| 468 |
|
|
| 469 |
/* When recursion is not being used, all "local" variables that have to be |
/* When recursion is not being used, all "local" variables that have to be |
| 470 |
preserved over calls to RMATCH() are part of a "frame" which is obtained from |
preserved over calls to RMATCH() are part of a "frame" which is obtained from |
| 507 |
#define charptr frame->Xcharptr |
#define charptr frame->Xcharptr |
| 508 |
#endif |
#endif |
| 509 |
#define callpat frame->Xcallpat |
#define callpat frame->Xcallpat |
| 510 |
|
#define codelink frame->Xcodelink |
| 511 |
#define data frame->Xdata |
#define data frame->Xdata |
| 512 |
#define next frame->Xnext |
#define next frame->Xnext |
| 513 |
#define pp frame->Xpp |
#define pp frame->Xpp |
| 588 |
uschar occhars[8]; |
uschar occhars[8]; |
| 589 |
#endif |
#endif |
| 590 |
|
|
| 591 |
|
int codelink; |
| 592 |
int ctype; |
int ctype; |
| 593 |
int length; |
int length; |
| 594 |
int max; |
int max; |
| 664 |
minimize = possessive = FALSE; |
minimize = possessive = FALSE; |
| 665 |
op = *ecode; |
op = *ecode; |
| 666 |
|
|
|
/* For partial matching, remember if we ever hit the end of the subject after |
|
|
matching at least one subject character. */ |
|
|
|
|
|
if (md->partial && |
|
|
eptr >= md->end_subject && |
|
|
eptr > mstart) |
|
|
md->hitend = TRUE; |
|
|
|
|
| 667 |
switch(op) |
switch(op) |
| 668 |
{ |
{ |
| 669 |
case OP_FAIL: |
case OP_FAIL: |
| 807 |
|
|
| 808 |
case OP_COND: |
case OP_COND: |
| 809 |
case OP_SCOND: |
case OP_SCOND: |
| 810 |
if (ecode[LINK_SIZE+1] == OP_RREF) /* Recursion test */ |
codelink= GET(ecode, 1); |
| 811 |
|
|
| 812 |
|
/* Because of the way auto-callout works during compile, a callout item is |
| 813 |
|
inserted between OP_COND and an assertion condition. */ |
| 814 |
|
|
| 815 |
|
if (ecode[LINK_SIZE+1] == OP_CALLOUT) |
| 816 |
{ |
{ |
| 817 |
offset = GET2(ecode, LINK_SIZE + 2); /* Recursion group number*/ |
if (pcre_callout != NULL) |
| 818 |
condition = md->recursive != NULL && |
{ |
| 819 |
(offset == RREF_ANY || offset == md->recursive->group_num); |
pcre_callout_block cb; |
| 820 |
ecode += condition? 3 : GET(ecode, 1); |
cb.version = 1; /* Version 1 of the callout block */ |
| 821 |
|
cb.callout_number = ecode[LINK_SIZE+2]; |
| 822 |
|
cb.offset_vector = md->offset_vector; |
| 823 |
|
cb.subject = (PCRE_SPTR)md->start_subject; |
| 824 |
|
cb.subject_length = md->end_subject - md->start_subject; |
| 825 |
|
cb.start_match = mstart - md->start_subject; |
| 826 |
|
cb.current_position = eptr - md->start_subject; |
| 827 |
|
cb.pattern_position = GET(ecode, LINK_SIZE + 3); |
| 828 |
|
cb.next_item_length = GET(ecode, 3 + 2*LINK_SIZE); |
| 829 |
|
cb.capture_top = offset_top/2; |
| 830 |
|
cb.capture_last = md->capture_last; |
| 831 |
|
cb.callout_data = md->callout_data; |
| 832 |
|
if ((rrc = (*pcre_callout)(&cb)) > 0) RRETURN(MATCH_NOMATCH); |
| 833 |
|
if (rrc < 0) RRETURN(rrc); |
| 834 |
|
} |
| 835 |
|
ecode += _pcre_OP_lengths[OP_CALLOUT]; |
| 836 |
|
} |
| 837 |
|
|
| 838 |
|
condcode = ecode[LINK_SIZE+1]; |
| 839 |
|
|
| 840 |
|
/* Now see what the actual condition is */ |
| 841 |
|
|
| 842 |
|
if (condcode == OP_RREF || condcode == OP_NRREF) /* Recursion test */ |
| 843 |
|
{ |
| 844 |
|
if (md->recursive == NULL) /* Not recursing => FALSE */ |
| 845 |
|
{ |
| 846 |
|
condition = FALSE; |
| 847 |
|
ecode += GET(ecode, 1); |
| 848 |
|
} |
| 849 |
|
else |
| 850 |
|
{ |
| 851 |
|
int recno = GET2(ecode, LINK_SIZE + 2); /* Recursion group number*/ |
| 852 |
|
condition = (recno == RREF_ANY || recno == md->recursive->group_num); |
| 853 |
|
|
| 854 |
|
/* If the test is for recursion into a specific subpattern, and it is |
| 855 |
|
false, but the test was set up by name, scan the table to see if the |
| 856 |
|
name refers to any other numbers, and test them. The condition is true |
| 857 |
|
if any one is set. */ |
| 858 |
|
|
| 859 |
|
if (!condition && condcode == OP_NRREF && recno != RREF_ANY) |
| 860 |
|
{ |
| 861 |
|
uschar *slotA = md->name_table; |
| 862 |
|
for (i = 0; i < md->name_count; i++) |
| 863 |
|
{ |
| 864 |
|
if (GET2(slotA, 0) == recno) break; |
| 865 |
|
slotA += md->name_entry_size; |
| 866 |
|
} |
| 867 |
|
|
| 868 |
|
/* Found a name for the number - there can be only one; duplicate |
| 869 |
|
names for different numbers are allowed, but not vice versa. First |
| 870 |
|
scan down for duplicates. */ |
| 871 |
|
|
| 872 |
|
if (i < md->name_count) |
| 873 |
|
{ |
| 874 |
|
uschar *slotB = slotA; |
| 875 |
|
while (slotB > md->name_table) |
| 876 |
|
{ |
| 877 |
|
slotB -= md->name_entry_size; |
| 878 |
|
if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0) |
| 879 |
|
{ |
| 880 |
|
condition = GET2(slotB, 0) == md->recursive->group_num; |
| 881 |
|
if (condition) break; |
| 882 |
|
} |
| 883 |
|
else break; |
| 884 |
|
} |
| 885 |
|
|
| 886 |
|
/* Scan up for duplicates */ |
| 887 |
|
|
| 888 |
|
if (!condition) |
| 889 |
|
{ |
| 890 |
|
slotB = slotA; |
| 891 |
|
for (i++; i < md->name_count; i++) |
| 892 |
|
{ |
| 893 |
|
slotB += md->name_entry_size; |
| 894 |
|
if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0) |
| 895 |
|
{ |
| 896 |
|
condition = GET2(slotB, 0) == md->recursive->group_num; |
| 897 |
|
if (condition) break; |
| 898 |
|
} |
| 899 |
|
else break; |
| 900 |
|
} |
| 901 |
|
} |
| 902 |
|
} |
| 903 |
|
} |
| 904 |
|
|
| 905 |
|
/* Chose branch according to the condition */ |
| 906 |
|
|
| 907 |
|
ecode += condition? 3 : GET(ecode, 1); |
| 908 |
|
} |
| 909 |
} |
} |
| 910 |
|
|
| 911 |
else if (ecode[LINK_SIZE+1] == OP_CREF) /* Group used test */ |
else if (condcode == OP_CREF || condcode == OP_NCREF) /* Group used test */ |
| 912 |
{ |
{ |
| 913 |
offset = GET2(ecode, LINK_SIZE+2) << 1; /* Doubled ref number */ |
offset = GET2(ecode, LINK_SIZE+2) << 1; /* Doubled ref number */ |
| 914 |
condition = offset < offset_top && md->offset_vector[offset] >= 0; |
condition = offset < offset_top && md->offset_vector[offset] >= 0; |
| 915 |
|
|
| 916 |
|
/* If the numbered capture is unset, but the reference was by name, |
| 917 |
|
scan the table to see if the name refers to any other numbers, and test |
| 918 |
|
them. The condition is true if any one is set. This is tediously similar |
| 919 |
|
to the code above, but not close enough to try to amalgamate. */ |
| 920 |
|
|
| 921 |
|
if (!condition && condcode == OP_NCREF) |
| 922 |
|
{ |
| 923 |
|
int refno = offset >> 1; |
| 924 |
|
uschar *slotA = md->name_table; |
| 925 |
|
|
| 926 |
|
for (i = 0; i < md->name_count; i++) |
| 927 |
|
{ |
| 928 |
|
if (GET2(slotA, 0) == refno) break; |
| 929 |
|
slotA += md->name_entry_size; |
| 930 |
|
} |
| 931 |
|
|
| 932 |
|
/* Found a name for the number - there can be only one; duplicate names |
| 933 |
|
for different numbers are allowed, but not vice versa. First scan down |
| 934 |
|
for duplicates. */ |
| 935 |
|
|
| 936 |
|
if (i < md->name_count) |
| 937 |
|
{ |
| 938 |
|
uschar *slotB = slotA; |
| 939 |
|
while (slotB > md->name_table) |
| 940 |
|
{ |
| 941 |
|
slotB -= md->name_entry_size; |
| 942 |
|
if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0) |
| 943 |
|
{ |
| 944 |
|
offset = GET2(slotB, 0) << 1; |
| 945 |
|
condition = offset < offset_top && |
| 946 |
|
md->offset_vector[offset] >= 0; |
| 947 |
|
if (condition) break; |
| 948 |
|
} |
| 949 |
|
else break; |
| 950 |
|
} |
| 951 |
|
|
| 952 |
|
/* Scan up for duplicates */ |
| 953 |
|
|
| 954 |
|
if (!condition) |
| 955 |
|
{ |
| 956 |
|
slotB = slotA; |
| 957 |
|
for (i++; i < md->name_count; i++) |
| 958 |
|
{ |
| 959 |
|
slotB += md->name_entry_size; |
| 960 |
|
if (strcmp((char *)slotA + 2, (char *)slotB + 2) == 0) |
| 961 |
|
{ |
| 962 |
|
offset = GET2(slotB, 0) << 1; |
| 963 |
|
condition = offset < offset_top && |
| 964 |
|
md->offset_vector[offset] >= 0; |
| 965 |
|
if (condition) break; |
| 966 |
|
} |
| 967 |
|
else break; |
| 968 |
|
} |
| 969 |
|
} |
| 970 |
|
} |
| 971 |
|
} |
| 972 |
|
|
| 973 |
|
/* Chose branch according to the condition */ |
| 974 |
|
|
| 975 |
ecode += condition? 3 : GET(ecode, 1); |
ecode += condition? 3 : GET(ecode, 1); |
| 976 |
} |
} |
| 977 |
|
|
| 978 |
else if (ecode[LINK_SIZE+1] == OP_DEF) /* DEFINE - always false */ |
else if (condcode == OP_DEF) /* DEFINE - always false */ |
| 979 |
{ |
{ |
| 980 |
condition = FALSE; |
condition = FALSE; |
| 981 |
ecode += GET(ecode, 1); |
ecode += GET(ecode, 1); |
| 1002 |
else |
else |
| 1003 |
{ |
{ |
| 1004 |
condition = FALSE; |
condition = FALSE; |
| 1005 |
ecode += GET(ecode, 1); |
ecode += codelink; |
| 1006 |
} |
} |
| 1007 |
} |
} |
| 1008 |
|
|
| 1025 |
goto TAIL_RECURSE; |
goto TAIL_RECURSE; |
| 1026 |
} |
} |
| 1027 |
} |
} |
| 1028 |
else /* Condition false & no 2nd alternative */ |
else /* Condition false & no alternative */ |
| 1029 |
{ |
{ |
| 1030 |
ecode += 1 + LINK_SIZE; |
ecode += 1 + LINK_SIZE; |
| 1031 |
} |
} |
| 1032 |
break; |
break; |
| 1033 |
|
|
| 1034 |
|
|
| 1035 |
|
/* Before OP_ACCEPT there may be any number of OP_CLOSE opcodes, |
| 1036 |
|
to close any currently open capturing brackets. */ |
| 1037 |
|
|
| 1038 |
|
case OP_CLOSE: |
| 1039 |
|
number = GET2(ecode, 1); |
| 1040 |
|
offset = number << 1; |
| 1041 |
|
|
| 1042 |
|
#ifdef DEBUG |
| 1043 |
|
printf("end bracket %d at *ACCEPT", number); |
| 1044 |
|
printf("\n"); |
| 1045 |
|
#endif |
| 1046 |
|
|
| 1047 |
|
md->capture_last = number; |
| 1048 |
|
if (offset >= md->offset_max) md->offset_overflow = TRUE; else |
| 1049 |
|
{ |
| 1050 |
|
md->offset_vector[offset] = |
| 1051 |
|
md->offset_vector[md->offset_end - number]; |
| 1052 |
|
md->offset_vector[offset+1] = eptr - md->start_subject; |
| 1053 |
|
if (offset_top <= offset) offset_top = offset + 2; |
| 1054 |
|
} |
| 1055 |
|
ecode += 3; |
| 1056 |
|
break; |
| 1057 |
|
|
| 1058 |
|
|
| 1059 |
/* End of the pattern, either real or forced. If we are in a top-level |
/* End of the pattern, either real or forced. If we are in a top-level |
| 1060 |
recursion, we should restore the offsets appropriately and continue from |
recursion, we should restore the offsets appropriately and continue from |
| 1061 |
after the call. */ |
after the call. */ |
| 1069 |
md->recursive = rec->prevrec; |
md->recursive = rec->prevrec; |
| 1070 |
memmove(md->offset_vector, rec->offset_save, |
memmove(md->offset_vector, rec->offset_save, |
| 1071 |
rec->saved_max * sizeof(int)); |
rec->saved_max * sizeof(int)); |
| 1072 |
|
offset_top = rec->save_offset_top; |
| 1073 |
mstart = rec->save_start; |
mstart = rec->save_start; |
| 1074 |
ims = original_ims; |
ims = original_ims; |
| 1075 |
ecode = rec->after_call; |
ecode = rec->after_call; |
| 1076 |
break; |
break; |
| 1077 |
} |
} |
| 1078 |
|
|
| 1079 |
/* Otherwise, if PCRE_NOTEMPTY is set, fail if we have matched an empty |
/* Otherwise, if we have matched an empty string, fail if PCRE_NOTEMPTY is |
| 1080 |
string - backtracking will then try other alternatives, if any. */ |
set, or if PCRE_NOTEMPTY_ATSTART is set and we have matched at the start of |
| 1081 |
|
the subject. In both cases, backtracking will then try other alternatives, |
| 1082 |
|
if any. */ |
| 1083 |
|
|
| 1084 |
|
if (eptr == mstart && |
| 1085 |
|
(md->notempty || |
| 1086 |
|
(md->notempty_atstart && |
| 1087 |
|
mstart == md->start_subject + md->start_offset))) |
| 1088 |
|
RRETURN(MATCH_NOMATCH); |
| 1089 |
|
|
| 1090 |
|
/* Otherwise, we have a match. */ |
| 1091 |
|
|
|
if (md->notempty && eptr == mstart) RRETURN(MATCH_NOMATCH); |
|
| 1092 |
md->end_match_ptr = eptr; /* Record where we ended */ |
md->end_match_ptr = eptr; /* Record where we ended */ |
| 1093 |
md->end_offset_top = offset_top; /* and how many extracts were taken */ |
md->end_offset_top = offset_top; /* and how many extracts were taken */ |
| 1094 |
md->start_match_ptr = mstart; /* and the start (\K can modify) */ |
md->start_match_ptr = mstart; /* and the start (\K can modify) */ |
| 1179 |
if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH); |
if (eptr < md->start_subject) RRETURN(MATCH_NOMATCH); |
| 1180 |
} |
} |
| 1181 |
|
|
| 1182 |
/* Skip to next op code */ |
/* Save the earliest consulted character, then skip to next op code */ |
| 1183 |
|
|
| 1184 |
|
if (eptr < md->start_used_ptr) md->start_used_ptr = eptr; |
| 1185 |
ecode += 1 + LINK_SIZE; |
ecode += 1 + LINK_SIZE; |
| 1186 |
break; |
break; |
| 1187 |
|
|
| 1261 |
memcpy(new_recursive.offset_save, md->offset_vector, |
memcpy(new_recursive.offset_save, md->offset_vector, |
| 1262 |
new_recursive.saved_max * sizeof(int)); |
new_recursive.saved_max * sizeof(int)); |
| 1263 |
new_recursive.save_start = mstart; |
new_recursive.save_start = mstart; |
| 1264 |
|
new_recursive.save_offset_top = offset_top; |
| 1265 |
mstart = eptr; |
mstart = eptr; |
| 1266 |
|
|
| 1267 |
/* OK, now we can do the recursion. For each top-level alternative we |
/* OK, now we can do the recursion. For each top-level alternative we |
| 1284 |
else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) |
else if (rrc != MATCH_NOMATCH && rrc != MATCH_THEN) |
| 1285 |
{ |
{ |
| 1286 |
DPRINTF(("Recursion gave error %d\n", rrc)); |
DPRINTF(("Recursion gave error %d\n", rrc)); |
| 1287 |
|
if (new_recursive.offset_save != stacksave) |
| 1288 |
|
(pcre_free)(new_recursive.offset_save); |
| 1289 |
RRETURN(rrc); |
RRETURN(rrc); |
| 1290 |
} |
} |
| 1291 |
|
|
| 1486 |
mstart = rec->save_start; |
mstart = rec->save_start; |
| 1487 |
memcpy(md->offset_vector, rec->offset_save, |
memcpy(md->offset_vector, rec->offset_save, |
| 1488 |
rec->saved_max * sizeof(int)); |
rec->saved_max * sizeof(int)); |
| 1489 |
|
offset_top = rec->save_offset_top; |
| 1490 |
ecode = rec->after_call; |
ecode = rec->after_call; |
| 1491 |
ims = original_ims; |
ims = original_ims; |
| 1492 |
break; |
break; |
| 1626 |
|
|
| 1627 |
/* Find out if the previous and current characters are "word" characters. |
/* Find out if the previous and current characters are "word" characters. |
| 1628 |
It takes a bit more work in UTF-8 mode. Characters > 255 are assumed to |
It takes a bit more work in UTF-8 mode. Characters > 255 are assumed to |
| 1629 |
be "non-word" characters. */ |
be "non-word" characters. Remember the earliest consulted character for |
| 1630 |
|
partial matching. */ |
| 1631 |
|
|
| 1632 |
#ifdef SUPPORT_UTF8 |
#ifdef SUPPORT_UTF8 |
| 1633 |
if (utf8) |
if (utf8) |
| 1634 |
{ |
{ |
| 1635 |
if (eptr == md->start_subject) prev_is_word = FALSE; else |
if (eptr == md->start_subject) prev_is_word = FALSE; else |
| 1636 |
{ |
{ |
| 1637 |
const uschar *lastptr = eptr - 1; |
USPTR lastptr = eptr - 1; |
| 1638 |
while((*lastptr & 0xc0) == 0x80) lastptr--; |
while((*lastptr & 0xc0) == 0x80) lastptr--; |
| 1639 |
|
if (lastptr < md->start_used_ptr) md->start_used_ptr = lastptr; |
| 1640 |
GETCHAR(c, lastptr); |
GETCHAR(c, lastptr); |
| 1641 |
prev_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0; |
prev_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0; |
| 1642 |
} |
} |
| 1643 |
if (eptr >= md->end_subject) cur_is_word = FALSE; else |
if (eptr >= md->end_subject) |
| 1644 |
|
{ |
| 1645 |
|
SCHECK_PARTIAL(); |
| 1646 |
|
cur_is_word = FALSE; |
| 1647 |
|
} |
| 1648 |
|
else |
| 1649 |
{ |
{ |
| 1650 |
GETCHAR(c, eptr); |
GETCHAR(c, eptr); |
| 1651 |
cur_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0; |
cur_is_word = c < 256 && (md->ctypes[c] & ctype_word) != 0; |
| 1654 |
else |
else |
| 1655 |
#endif |
#endif |
| 1656 |
|
|
| 1657 |
/* More streamlined when not in UTF-8 mode */ |
/* Not in UTF-8 mode */ |
| 1658 |
|
|
| 1659 |
{ |
{ |
| 1660 |
prev_is_word = (eptr != md->start_subject) && |
if (eptr == md->start_subject) prev_is_word = FALSE; else |
| 1661 |
((md->ctypes[eptr[-1]] & ctype_word) != 0); |
{ |
| 1662 |
cur_is_word = (eptr < md->end_subject) && |
if (eptr <= md->start_used_ptr) md->start_used_ptr = eptr - 1; |
| 1663 |
((md->ctypes[*eptr] & ctype_word) != 0); |
prev_is_word = ((md->ctypes[eptr[-1]] & ctype_word) != 0); |
| 1664 |
|
} |
| 1665 |
|
if (eptr >= md->end_subject) |
| 1666 |
|
{ |
| 1667 |
|
SCHECK_PARTIAL(); |
| 1668 |
|
cur_is_word = FALSE; |
| 1669 |
|
} |
| 1670 |
|
else cur_is_word = ((md->ctypes[*eptr] & ctype_word) != 0); |
| 1671 |
} |
} |
| 1672 |
|
|
| 1673 |
/* Now see if the situation is what we want */ |
/* Now see if the situation is what we want */ |
| 1685 |
/* Fall through */ |
/* Fall through */ |
| 1686 |
|
|
| 1687 |
case OP_ALLANY: |
case OP_ALLANY: |
| 1688 |
if (eptr++ >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr++ >= md->end_subject) |
| 1689 |
|
{ |
| 1690 |
|
SCHECK_PARTIAL(); |
| 1691 |
|
RRETURN(MATCH_NOMATCH); |
| 1692 |
|
} |
| 1693 |
if (utf8) while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++; |
if (utf8) while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++; |
| 1694 |
ecode++; |
ecode++; |
| 1695 |
break; |
break; |
| 1698 |
any byte, even newline, independent of the setting of PCRE_DOTALL. */ |
any byte, even newline, independent of the setting of PCRE_DOTALL. */ |
| 1699 |
|
|
| 1700 |
case OP_ANYBYTE: |
case OP_ANYBYTE: |
| 1701 |
if (eptr++ >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr++ >= md->end_subject) |
| 1702 |
|
{ |
| 1703 |
|
SCHECK_PARTIAL(); |
| 1704 |
|
RRETURN(MATCH_NOMATCH); |
| 1705 |
|
} |
| 1706 |
ecode++; |
ecode++; |
| 1707 |
break; |
break; |
| 1708 |
|
|
| 1709 |
case OP_NOT_DIGIT: |
case OP_NOT_DIGIT: |
| 1710 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 1711 |
|
{ |
| 1712 |
|
SCHECK_PARTIAL(); |
| 1713 |
|
RRETURN(MATCH_NOMATCH); |
| 1714 |
|
} |
| 1715 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 1716 |
if ( |
if ( |
| 1717 |
#ifdef SUPPORT_UTF8 |
#ifdef SUPPORT_UTF8 |
| 1724 |
break; |
break; |
| 1725 |
|
|
| 1726 |
case OP_DIGIT: |
case OP_DIGIT: |
| 1727 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 1728 |
|
{ |
| 1729 |
|
SCHECK_PARTIAL(); |
| 1730 |
|
RRETURN(MATCH_NOMATCH); |
| 1731 |
|
} |
| 1732 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 1733 |
if ( |
if ( |
| 1734 |
#ifdef SUPPORT_UTF8 |
#ifdef SUPPORT_UTF8 |
| 1741 |
break; |
break; |
| 1742 |
|
|
| 1743 |
case OP_NOT_WHITESPACE: |
case OP_NOT_WHITESPACE: |
| 1744 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 1745 |
|
{ |
| 1746 |
|
SCHECK_PARTIAL(); |
| 1747 |
|
RRETURN(MATCH_NOMATCH); |
| 1748 |
|
} |
| 1749 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 1750 |
if ( |
if ( |
| 1751 |
#ifdef SUPPORT_UTF8 |
#ifdef SUPPORT_UTF8 |
| 1758 |
break; |
break; |
| 1759 |
|
|
| 1760 |
case OP_WHITESPACE: |
case OP_WHITESPACE: |
| 1761 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 1762 |
|
{ |
| 1763 |
|
SCHECK_PARTIAL(); |
| 1764 |
|
RRETURN(MATCH_NOMATCH); |
| 1765 |
|
} |
| 1766 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 1767 |
if ( |
if ( |
| 1768 |
#ifdef SUPPORT_UTF8 |
#ifdef SUPPORT_UTF8 |
| 1775 |
break; |
break; |
| 1776 |
|
|
| 1777 |
case OP_NOT_WORDCHAR: |
case OP_NOT_WORDCHAR: |
| 1778 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 1779 |
|
{ |
| 1780 |
|
SCHECK_PARTIAL(); |
| 1781 |
|
RRETURN(MATCH_NOMATCH); |
| 1782 |
|
} |
| 1783 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 1784 |
if ( |
if ( |
| 1785 |
#ifdef SUPPORT_UTF8 |
#ifdef SUPPORT_UTF8 |
| 1792 |
break; |
break; |
| 1793 |
|
|
| 1794 |
case OP_WORDCHAR: |
case OP_WORDCHAR: |
| 1795 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 1796 |
|
{ |
| 1797 |
|
SCHECK_PARTIAL(); |
| 1798 |
|
RRETURN(MATCH_NOMATCH); |
| 1799 |
|
} |
| 1800 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 1801 |
if ( |
if ( |
| 1802 |
#ifdef SUPPORT_UTF8 |
#ifdef SUPPORT_UTF8 |
| 1809 |
break; |
break; |
| 1810 |
|
|
| 1811 |
case OP_ANYNL: |
case OP_ANYNL: |
| 1812 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 1813 |
|
{ |
| 1814 |
|
SCHECK_PARTIAL(); |
| 1815 |
|
RRETURN(MATCH_NOMATCH); |
| 1816 |
|
} |
| 1817 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 1818 |
switch(c) |
switch(c) |
| 1819 |
{ |
{ |
| 1837 |
break; |
break; |
| 1838 |
|
|
| 1839 |
case OP_NOT_HSPACE: |
case OP_NOT_HSPACE: |
| 1840 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 1841 |
|
{ |
| 1842 |
|
SCHECK_PARTIAL(); |
| 1843 |
|
RRETURN(MATCH_NOMATCH); |
| 1844 |
|
} |
| 1845 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 1846 |
switch(c) |
switch(c) |
| 1847 |
{ |
{ |
| 1871 |
break; |
break; |
| 1872 |
|
|
| 1873 |
case OP_HSPACE: |
case OP_HSPACE: |
| 1874 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 1875 |
|
{ |
| 1876 |
|
SCHECK_PARTIAL(); |
| 1877 |
|
RRETURN(MATCH_NOMATCH); |
| 1878 |
|
} |
| 1879 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 1880 |
switch(c) |
switch(c) |
| 1881 |
{ |
{ |
| 1905 |
break; |
break; |
| 1906 |
|
|
| 1907 |
case OP_NOT_VSPACE: |
case OP_NOT_VSPACE: |
| 1908 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 1909 |
|
{ |
| 1910 |
|
SCHECK_PARTIAL(); |
| 1911 |
|
RRETURN(MATCH_NOMATCH); |
| 1912 |
|
} |
| 1913 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 1914 |
switch(c) |
switch(c) |
| 1915 |
{ |
{ |
| 1927 |
break; |
break; |
| 1928 |
|
|
| 1929 |
case OP_VSPACE: |
case OP_VSPACE: |
| 1930 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 1931 |
|
{ |
| 1932 |
|
SCHECK_PARTIAL(); |
| 1933 |
|
RRETURN(MATCH_NOMATCH); |
| 1934 |
|
} |
| 1935 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 1936 |
switch(c) |
switch(c) |
| 1937 |
{ |
{ |
| 1954 |
|
|
| 1955 |
case OP_PROP: |
case OP_PROP: |
| 1956 |
case OP_NOTPROP: |
case OP_NOTPROP: |
| 1957 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 1958 |
|
{ |
| 1959 |
|
SCHECK_PARTIAL(); |
| 1960 |
|
RRETURN(MATCH_NOMATCH); |
| 1961 |
|
} |
| 1962 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 1963 |
{ |
{ |
| 1964 |
const ucd_record * prop = GET_UCD(c); |
const ucd_record *prop = GET_UCD(c); |
| 1965 |
|
|
| 1966 |
switch(ecode[1]) |
switch(ecode[1]) |
| 1967 |
{ |
{ |
| 2003 |
is in the binary; otherwise a compile-time error occurs. */ |
is in the binary; otherwise a compile-time error occurs. */ |
| 2004 |
|
|
| 2005 |
case OP_EXTUNI: |
case OP_EXTUNI: |
| 2006 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 2007 |
|
{ |
| 2008 |
|
SCHECK_PARTIAL(); |
| 2009 |
|
RRETURN(MATCH_NOMATCH); |
| 2010 |
|
} |
| 2011 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 2012 |
{ |
{ |
| 2013 |
int category = UCD_CATEGORY(c); |
int category = UCD_CATEGORY(c); |
| 2087 |
break; |
break; |
| 2088 |
|
|
| 2089 |
default: /* No repeat follows */ |
default: /* No repeat follows */ |
| 2090 |
if (!match_ref(offset, eptr, length, md, ims)) RRETURN(MATCH_NOMATCH); |
if (!match_ref(offset, eptr, length, md, ims)) |
| 2091 |
|
{ |
| 2092 |
|
CHECK_PARTIAL(); |
| 2093 |
|
RRETURN(MATCH_NOMATCH); |
| 2094 |
|
} |
| 2095 |
eptr += length; |
eptr += length; |
| 2096 |
continue; /* With the main loop */ |
continue; /* With the main loop */ |
| 2097 |
} |
} |
| 2107 |
|
|
| 2108 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 2109 |
{ |
{ |
| 2110 |
if (!match_ref(offset, eptr, length, md, ims)) RRETURN(MATCH_NOMATCH); |
if (!match_ref(offset, eptr, length, md, ims)) |
| 2111 |
|
{ |
| 2112 |
|
CHECK_PARTIAL(); |
| 2113 |
|
RRETURN(MATCH_NOMATCH); |
| 2114 |
|
} |
| 2115 |
eptr += length; |
eptr += length; |
| 2116 |
} |
} |
| 2117 |
|
|
| 2128 |
{ |
{ |
| 2129 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM14); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM14); |
| 2130 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 2131 |
if (fi >= max || !match_ref(offset, eptr, length, md, ims)) |
if (fi >= max) RRETURN(MATCH_NOMATCH); |
| 2132 |
|
if (!match_ref(offset, eptr, length, md, ims)) |
| 2133 |
|
{ |
| 2134 |
|
CHECK_PARTIAL(); |
| 2135 |
RRETURN(MATCH_NOMATCH); |
RRETURN(MATCH_NOMATCH); |
| 2136 |
|
} |
| 2137 |
eptr += length; |
eptr += length; |
| 2138 |
} |
} |
| 2139 |
/* Control never gets here */ |
/* Control never gets here */ |
| 2146 |
pp = eptr; |
pp = eptr; |
| 2147 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 2148 |
{ |
{ |
| 2149 |
if (!match_ref(offset, eptr, length, md, ims)) break; |
if (!match_ref(offset, eptr, length, md, ims)) |
| 2150 |
|
{ |
| 2151 |
|
CHECK_PARTIAL(); |
| 2152 |
|
break; |
| 2153 |
|
} |
| 2154 |
eptr += length; |
eptr += length; |
| 2155 |
} |
} |
| 2156 |
while (eptr >= pp) |
while (eptr >= pp) |
| 2164 |
} |
} |
| 2165 |
/* Control never gets here */ |
/* Control never gets here */ |
| 2166 |
|
|
|
|
|
|
|
|
| 2167 |
/* Match a bit-mapped character class, possibly repeatedly. This op code is |
/* Match a bit-mapped character class, possibly repeatedly. This op code is |
| 2168 |
used when all the characters in the class have values in the range 0-255, |
used when all the characters in the class have values in the range 0-255, |
| 2169 |
and either the matching is caseful, or the characters are in the range |
and either the matching is caseful, or the characters are in the range |
| 2218 |
{ |
{ |
| 2219 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 2220 |
{ |
{ |
| 2221 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 2222 |
|
{ |
| 2223 |
|
SCHECK_PARTIAL(); |
| 2224 |
|
RRETURN(MATCH_NOMATCH); |
| 2225 |
|
} |
| 2226 |
GETCHARINC(c, eptr); |
GETCHARINC(c, eptr); |
| 2227 |
if (c > 255) |
if (c > 255) |
| 2228 |
{ |
{ |
| 2240 |
{ |
{ |
| 2241 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 2242 |
{ |
{ |
| 2243 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 2244 |
|
{ |
| 2245 |
|
SCHECK_PARTIAL(); |
| 2246 |
|
RRETURN(MATCH_NOMATCH); |
| 2247 |
|
} |
| 2248 |
c = *eptr++; |
c = *eptr++; |
| 2249 |
if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH); |
if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH); |
| 2250 |
} |
} |
| 2268 |
{ |
{ |
| 2269 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM16); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM16); |
| 2270 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 2271 |
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (fi >= max) RRETURN(MATCH_NOMATCH); |
| 2272 |
|
if (eptr >= md->end_subject) |
| 2273 |
|
{ |
| 2274 |
|
SCHECK_PARTIAL(); |
| 2275 |
|
RRETURN(MATCH_NOMATCH); |
| 2276 |
|
} |
| 2277 |
GETCHARINC(c, eptr); |
GETCHARINC(c, eptr); |
| 2278 |
if (c > 255) |
if (c > 255) |
| 2279 |
{ |
{ |
| 2293 |
{ |
{ |
| 2294 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM17); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM17); |
| 2295 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 2296 |
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (fi >= max) RRETURN(MATCH_NOMATCH); |
| 2297 |
|
if (eptr >= md->end_subject) |
| 2298 |
|
{ |
| 2299 |
|
SCHECK_PARTIAL(); |
| 2300 |
|
RRETURN(MATCH_NOMATCH); |
| 2301 |
|
} |
| 2302 |
c = *eptr++; |
c = *eptr++; |
| 2303 |
if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH); |
if ((data[c/8] & (1 << (c&7))) == 0) RRETURN(MATCH_NOMATCH); |
| 2304 |
} |
} |
| 2319 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 2320 |
{ |
{ |
| 2321 |
int len = 1; |
int len = 1; |
| 2322 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 2323 |
|
{ |
| 2324 |
|
SCHECK_PARTIAL(); |
| 2325 |
|
break; |
| 2326 |
|
} |
| 2327 |
GETCHARLEN(c, eptr, len); |
GETCHARLEN(c, eptr, len); |
| 2328 |
if (c > 255) |
if (c > 255) |
| 2329 |
{ |
{ |
| 2349 |
{ |
{ |
| 2350 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 2351 |
{ |
{ |
| 2352 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 2353 |
|
{ |
| 2354 |
|
SCHECK_PARTIAL(); |
| 2355 |
|
break; |
| 2356 |
|
} |
| 2357 |
c = *eptr; |
c = *eptr; |
| 2358 |
if ((data[c/8] & (1 << (c&7))) == 0) break; |
if ((data[c/8] & (1 << (c&7))) == 0) break; |
| 2359 |
eptr++; |
eptr++; |
| 2373 |
|
|
| 2374 |
|
|
| 2375 |
/* Match an extended character class. This opcode is encountered only |
/* Match an extended character class. This opcode is encountered only |
| 2376 |
in UTF-8 mode, because that's the only time it is compiled. */ |
when UTF-8 mode mode is supported. Nevertheless, we may not be in UTF-8 |
| 2377 |
|
mode, because Unicode properties are supported in non-UTF-8 mode. */ |
| 2378 |
|
|
| 2379 |
#ifdef SUPPORT_UTF8 |
#ifdef SUPPORT_UTF8 |
| 2380 |
case OP_XCLASS: |
case OP_XCLASS: |
| 2415 |
|
|
| 2416 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 2417 |
{ |
{ |
| 2418 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 2419 |
GETCHARINC(c, eptr); |
{ |
| 2420 |
|
SCHECK_PARTIAL(); |
| 2421 |
|
RRETURN(MATCH_NOMATCH); |
| 2422 |
|
} |
| 2423 |
|
GETCHARINCTEST(c, eptr); |
| 2424 |
if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH); |
if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH); |
| 2425 |
} |
} |
| 2426 |
|
|
| 2438 |
{ |
{ |
| 2439 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM20); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM20); |
| 2440 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 2441 |
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (fi >= max) RRETURN(MATCH_NOMATCH); |
| 2442 |
GETCHARINC(c, eptr); |
if (eptr >= md->end_subject) |
| 2443 |
|
{ |
| 2444 |
|
SCHECK_PARTIAL(); |
| 2445 |
|
RRETURN(MATCH_NOMATCH); |
| 2446 |
|
} |
| 2447 |
|
GETCHARINCTEST(c, eptr); |
| 2448 |
if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH); |
if (!_pcre_xclass(c, data)) RRETURN(MATCH_NOMATCH); |
| 2449 |
} |
} |
| 2450 |
/* Control never gets here */ |
/* Control never gets here */ |
| 2458 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 2459 |
{ |
{ |
| 2460 |
int len = 1; |
int len = 1; |
| 2461 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 2462 |
GETCHARLEN(c, eptr, len); |
{ |
| 2463 |
|
SCHECK_PARTIAL(); |
| 2464 |
|
break; |
| 2465 |
|
} |
| 2466 |
|
GETCHARLENTEST(c, eptr, len); |
| 2467 |
if (!_pcre_xclass(c, data)) break; |
if (!_pcre_xclass(c, data)) break; |
| 2468 |
eptr += len; |
eptr += len; |
| 2469 |
} |
} |
| 2490 |
length = 1; |
length = 1; |
| 2491 |
ecode++; |
ecode++; |
| 2492 |
GETCHARLEN(fc, ecode, length); |
GETCHARLEN(fc, ecode, length); |
| 2493 |
if (length > md->end_subject - eptr) RRETURN(MATCH_NOMATCH); |
if (length > md->end_subject - eptr) |
| 2494 |
|
{ |
| 2495 |
|
CHECK_PARTIAL(); /* Not SCHECK_PARTIAL() */ |
| 2496 |
|
RRETURN(MATCH_NOMATCH); |
| 2497 |
|
} |
| 2498 |
while (length-- > 0) if (*ecode++ != *eptr++) RRETURN(MATCH_NOMATCH); |
while (length-- > 0) if (*ecode++ != *eptr++) RRETURN(MATCH_NOMATCH); |
| 2499 |
} |
} |
| 2500 |
else |
else |
| 2502 |
|
|
| 2503 |
/* Non-UTF-8 mode */ |
/* Non-UTF-8 mode */ |
| 2504 |
{ |
{ |
| 2505 |
if (md->end_subject - eptr < 1) RRETURN(MATCH_NOMATCH); |
if (md->end_subject - eptr < 1) |
| 2506 |
|
{ |
| 2507 |
|
SCHECK_PARTIAL(); /* This one can use SCHECK_PARTIAL() */ |
| 2508 |
|
RRETURN(MATCH_NOMATCH); |
| 2509 |
|
} |
| 2510 |
if (ecode[1] != *eptr++) RRETURN(MATCH_NOMATCH); |
if (ecode[1] != *eptr++) RRETURN(MATCH_NOMATCH); |
| 2511 |
ecode += 2; |
ecode += 2; |
| 2512 |
} |
} |
| 2522 |
ecode++; |
ecode++; |
| 2523 |
GETCHARLEN(fc, ecode, length); |
GETCHARLEN(fc, ecode, length); |
| 2524 |
|
|
| 2525 |
if (length > md->end_subject - eptr) RRETURN(MATCH_NOMATCH); |
if (length > md->end_subject - eptr) |
| 2526 |
|
{ |
| 2527 |
|
CHECK_PARTIAL(); /* Not SCHECK_PARTIAL() */ |
| 2528 |
|
RRETURN(MATCH_NOMATCH); |
| 2529 |
|
} |
| 2530 |
|
|
| 2531 |
/* If the pattern character's value is < 128, we have only one byte, and |
/* If the pattern character's value is < 128, we have only one byte, and |
| 2532 |
can use the fast lookup table. */ |
can use the fast lookup table. */ |
| 2561 |
|
|
| 2562 |
/* Non-UTF-8 mode */ |
/* Non-UTF-8 mode */ |
| 2563 |
{ |
{ |
| 2564 |
if (md->end_subject - eptr < 1) RRETURN(MATCH_NOMATCH); |
if (md->end_subject - eptr < 1) |
| 2565 |
|
{ |
| 2566 |
|
SCHECK_PARTIAL(); /* This one can use SCHECK_PARTIAL() */ |
| 2567 |
|
RRETURN(MATCH_NOMATCH); |
| 2568 |
|
} |
| 2569 |
if (md->lcc[ecode[1]] != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH); |
if (md->lcc[ecode[1]] != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH); |
| 2570 |
ecode += 2; |
ecode += 2; |
| 2571 |
} |
} |
| 2619 |
case OP_MINQUERY: |
case OP_MINQUERY: |
| 2620 |
c = *ecode++ - OP_STAR; |
c = *ecode++ - OP_STAR; |
| 2621 |
minimize = (c & 1) != 0; |
minimize = (c & 1) != 0; |
| 2622 |
|
|
| 2623 |
min = rep_min[c]; /* Pick up values from tables; */ |
min = rep_min[c]; /* Pick up values from tables; */ |
| 2624 |
max = rep_max[c]; /* zero for max => infinity */ |
max = rep_max[c]; /* zero for max => infinity */ |
| 2625 |
if (max == 0) max = INT_MAX; |
if (max == 0) max = INT_MAX; |
| 2626 |
|
|
| 2627 |
/* Common code for all repeated single-character matches. We can give |
/* Common code for all repeated single-character matches. */ |
|
up quickly if there are fewer than the minimum number of characters left in |
|
|
the subject. */ |
|
| 2628 |
|
|
| 2629 |
REPEATCHAR: |
REPEATCHAR: |
| 2630 |
#ifdef SUPPORT_UTF8 |
#ifdef SUPPORT_UTF8 |
| 2633 |
length = 1; |
length = 1; |
| 2634 |
charptr = ecode; |
charptr = ecode; |
| 2635 |
GETCHARLEN(fc, ecode, length); |
GETCHARLEN(fc, ecode, length); |
|
if (min * length > md->end_subject - eptr) RRETURN(MATCH_NOMATCH); |
|
| 2636 |
ecode += length; |
ecode += length; |
| 2637 |
|
|
| 2638 |
/* Handle multibyte character matching specially here. There is |
/* Handle multibyte character matching specially here. There is |
| 2650 |
|
|
| 2651 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 2652 |
{ |
{ |
| 2653 |
if (memcmp(eptr, charptr, length) == 0) eptr += length; |
if (eptr <= md->end_subject - length && |
| 2654 |
|
memcmp(eptr, charptr, length) == 0) eptr += length; |
| 2655 |
#ifdef SUPPORT_UCP |
#ifdef SUPPORT_UCP |
| 2656 |
/* Need braces because of following else */ |
else if (oclength > 0 && |
| 2657 |
else if (oclength == 0) { RRETURN(MATCH_NOMATCH); } |
eptr <= md->end_subject - oclength && |
| 2658 |
|
memcmp(eptr, occhars, oclength) == 0) eptr += oclength; |
| 2659 |
|
#endif /* SUPPORT_UCP */ |
| 2660 |
else |
else |
| 2661 |
{ |
{ |
| 2662 |
if (memcmp(eptr, occhars, oclength) != 0) RRETURN(MATCH_NOMATCH); |
CHECK_PARTIAL(); |
| 2663 |
eptr += oclength; |
RRETURN(MATCH_NOMATCH); |
| 2664 |
} |
} |
|
#else /* without SUPPORT_UCP */ |
|
|
else { RRETURN(MATCH_NOMATCH); } |
|
|
#endif /* SUPPORT_UCP */ |
|
| 2665 |
} |
} |
| 2666 |
|
|
| 2667 |
if (min == max) continue; |
if (min == max) continue; |
| 2672 |
{ |
{ |
| 2673 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM22); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM22); |
| 2674 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 2675 |
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (fi >= max) RRETURN(MATCH_NOMATCH); |
| 2676 |
if (memcmp(eptr, charptr, length) == 0) eptr += length; |
if (eptr <= md->end_subject - length && |
| 2677 |
|
memcmp(eptr, charptr, length) == 0) eptr += length; |
| 2678 |
#ifdef SUPPORT_UCP |
#ifdef SUPPORT_UCP |
| 2679 |
/* Need braces because of following else */ |
else if (oclength > 0 && |
| 2680 |
else if (oclength == 0) { RRETURN(MATCH_NOMATCH); } |
eptr <= md->end_subject - oclength && |
| 2681 |
|
memcmp(eptr, occhars, oclength) == 0) eptr += oclength; |
| 2682 |
|
#endif /* SUPPORT_UCP */ |
| 2683 |
else |
else |
| 2684 |
{ |
{ |
| 2685 |
if (memcmp(eptr, occhars, oclength) != 0) RRETURN(MATCH_NOMATCH); |
CHECK_PARTIAL(); |
| 2686 |
eptr += oclength; |
RRETURN(MATCH_NOMATCH); |
| 2687 |
} |
} |
|
#else /* without SUPPORT_UCP */ |
|
|
else { RRETURN (MATCH_NOMATCH); } |
|
|
#endif /* SUPPORT_UCP */ |
|
| 2688 |
} |
} |
| 2689 |
/* Control never gets here */ |
/* Control never gets here */ |
| 2690 |
} |
} |
| 2694 |
pp = eptr; |
pp = eptr; |
| 2695 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 2696 |
{ |
{ |
| 2697 |
if (eptr > md->end_subject - length) break; |
if (eptr <= md->end_subject - length && |
| 2698 |
if (memcmp(eptr, charptr, length) == 0) eptr += length; |
memcmp(eptr, charptr, length) == 0) eptr += length; |
| 2699 |
#ifdef SUPPORT_UCP |
#ifdef SUPPORT_UCP |
| 2700 |
else if (oclength == 0) break; |
else if (oclength > 0 && |
| 2701 |
|
eptr <= md->end_subject - oclength && |
| 2702 |
|
memcmp(eptr, occhars, oclength) == 0) eptr += oclength; |
| 2703 |
|
#endif /* SUPPORT_UCP */ |
| 2704 |
else |
else |
| 2705 |
{ |
{ |
| 2706 |
if (memcmp(eptr, occhars, oclength) != 0) break; |
CHECK_PARTIAL(); |
| 2707 |
eptr += oclength; |
break; |
| 2708 |
} |
} |
|
#else /* without SUPPORT_UCP */ |
|
|
else break; |
|
|
#endif /* SUPPORT_UCP */ |
|
| 2709 |
} |
} |
| 2710 |
|
|
| 2711 |
if (possessive) continue; |
if (possessive) continue; |
| 2712 |
|
|
| 2713 |
for(;;) |
for(;;) |
| 2714 |
{ |
{ |
| 2715 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM23); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM23); |
| 2716 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 2717 |
if (eptr == pp) RRETURN(MATCH_NOMATCH); |
if (eptr == pp) { RRETURN(MATCH_NOMATCH); } |
| 2718 |
#ifdef SUPPORT_UCP |
#ifdef SUPPORT_UCP |
| 2719 |
eptr--; |
eptr--; |
| 2720 |
BACKCHAR(eptr); |
BACKCHAR(eptr); |
| 2721 |
#else /* without SUPPORT_UCP */ |
#else /* without SUPPORT_UCP */ |
| 2722 |
eptr -= length; |
eptr -= length; |
| 2723 |
#endif /* SUPPORT_UCP */ |
#endif /* SUPPORT_UCP */ |
| 2724 |
} |
} |
| 2725 |
} |
} |
| 2726 |
/* Control never gets here */ |
/* Control never gets here */ |
| 2727 |
} |
} |
| 2734 |
#endif /* SUPPORT_UTF8 */ |
#endif /* SUPPORT_UTF8 */ |
| 2735 |
|
|
| 2736 |
/* When not in UTF-8 mode, load a single-byte character. */ |
/* When not in UTF-8 mode, load a single-byte character. */ |
| 2737 |
{ |
|
| 2738 |
if (min > md->end_subject - eptr) RRETURN(MATCH_NOMATCH); |
fc = *ecode++; |
|
fc = *ecode++; |
|
|
} |
|
| 2739 |
|
|
| 2740 |
/* The value of fc at this point is always less than 256, though we may or |
/* The value of fc at this point is always less than 256, though we may or |
| 2741 |
may not be in UTF-8 mode. The code is duplicated for the caseless and |
may not be in UTF-8 mode. The code is duplicated for the caseless and |
| 2753 |
{ |
{ |
| 2754 |
fc = md->lcc[fc]; |
fc = md->lcc[fc]; |
| 2755 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 2756 |
|
{ |
| 2757 |
|
if (eptr >= md->end_subject) |
| 2758 |
|
{ |
| 2759 |
|
SCHECK_PARTIAL(); |
| 2760 |
|
RRETURN(MATCH_NOMATCH); |
| 2761 |
|
} |
| 2762 |
if (fc != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH); |
if (fc != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH); |
| 2763 |
|
} |
| 2764 |
if (min == max) continue; |
if (min == max) continue; |
| 2765 |
if (minimize) |
if (minimize) |
| 2766 |
{ |
{ |
| 2768 |
{ |
{ |
| 2769 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM24); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM24); |
| 2770 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 2771 |
if (fi >= max || eptr >= md->end_subject || |
if (fi >= max) RRETURN(MATCH_NOMATCH); |
| 2772 |
fc != md->lcc[*eptr++]) |
if (eptr >= md->end_subject) |
| 2773 |
|
{ |
| 2774 |
|
SCHECK_PARTIAL(); |
| 2775 |
RRETURN(MATCH_NOMATCH); |
RRETURN(MATCH_NOMATCH); |
| 2776 |
|
} |
| 2777 |
|
if (fc != md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH); |
| 2778 |
} |
} |
| 2779 |
/* Control never gets here */ |
/* Control never gets here */ |
| 2780 |
} |
} |
| 2783 |
pp = eptr; |
pp = eptr; |
| 2784 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 2785 |
{ |
{ |
| 2786 |
if (eptr >= md->end_subject || fc != md->lcc[*eptr]) break; |
if (eptr >= md->end_subject) |
| 2787 |
|
{ |
| 2788 |
|
SCHECK_PARTIAL(); |
| 2789 |
|
break; |
| 2790 |
|
} |
| 2791 |
|
if (fc != md->lcc[*eptr]) break; |
| 2792 |
eptr++; |
eptr++; |
| 2793 |
} |
} |
| 2794 |
|
|
| 2795 |
if (possessive) continue; |
if (possessive) continue; |
| 2796 |
|
|
| 2797 |
while (eptr >= pp) |
while (eptr >= pp) |
| 2798 |
{ |
{ |
| 2799 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM25); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM25); |
| 2809 |
|
|
| 2810 |
else |
else |
| 2811 |
{ |
{ |
| 2812 |
for (i = 1; i <= min; i++) if (fc != *eptr++) RRETURN(MATCH_NOMATCH); |
for (i = 1; i <= min; i++) |
| 2813 |
|
{ |
| 2814 |
|
if (eptr >= md->end_subject) |
| 2815 |
|
{ |
| 2816 |
|
SCHECK_PARTIAL(); |
| 2817 |
|
RRETURN(MATCH_NOMATCH); |
| 2818 |
|
} |
| 2819 |
|
if (fc != *eptr++) RRETURN(MATCH_NOMATCH); |
| 2820 |
|
} |
| 2821 |
|
|
| 2822 |
if (min == max) continue; |
if (min == max) continue; |
| 2823 |
|
|
| 2824 |
if (minimize) |
if (minimize) |
| 2825 |
{ |
{ |
| 2826 |
for (fi = min;; fi++) |
for (fi = min;; fi++) |
| 2827 |
{ |
{ |
| 2828 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM26); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM26); |
| 2829 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 2830 |
if (fi >= max || eptr >= md->end_subject || fc != *eptr++) |
if (fi >= max) RRETURN(MATCH_NOMATCH); |
| 2831 |
|
if (eptr >= md->end_subject) |
| 2832 |
|
{ |
| 2833 |
|
SCHECK_PARTIAL(); |
| 2834 |
RRETURN(MATCH_NOMATCH); |
RRETURN(MATCH_NOMATCH); |
| 2835 |
|
} |
| 2836 |
|
if (fc != *eptr++) RRETURN(MATCH_NOMATCH); |
| 2837 |
} |
} |
| 2838 |
/* Control never gets here */ |
/* Control never gets here */ |
| 2839 |
} |
} |
| 2842 |
pp = eptr; |
pp = eptr; |
| 2843 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 2844 |
{ |
{ |
| 2845 |
if (eptr >= md->end_subject || fc != *eptr) break; |
if (eptr >= md->end_subject) |
| 2846 |
|
{ |
| 2847 |
|
SCHECK_PARTIAL(); |
| 2848 |
|
break; |
| 2849 |
|
} |
| 2850 |
|
if (fc != *eptr) break; |
| 2851 |
eptr++; |
eptr++; |
| 2852 |
} |
} |
| 2853 |
if (possessive) continue; |
if (possessive) continue; |
| 2854 |
|
|
| 2855 |
while (eptr >= pp) |
while (eptr >= pp) |
| 2856 |
{ |
{ |
| 2857 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM27); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM27); |
| 2867 |
checking can be multibyte. */ |
checking can be multibyte. */ |
| 2868 |
|
|
| 2869 |
case OP_NOT: |
case OP_NOT: |
| 2870 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 2871 |
|
{ |
| 2872 |
|
SCHECK_PARTIAL(); |
| 2873 |
|
RRETURN(MATCH_NOMATCH); |
| 2874 |
|
} |
| 2875 |
ecode++; |
ecode++; |
| 2876 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 2877 |
if ((ims & PCRE_CASELESS) != 0) |
if ((ims & PCRE_CASELESS) != 0) |
| 2948 |
max = rep_max[c]; /* zero for max => infinity */ |
max = rep_max[c]; /* zero for max => infinity */ |
| 2949 |
if (max == 0) max = INT_MAX; |
if (max == 0) max = INT_MAX; |
| 2950 |
|
|
| 2951 |
/* Common code for all repeated single-byte matches. We can give up quickly |
/* Common code for all repeated single-byte matches. */ |
|
if there are fewer than the minimum number of bytes left in the |
|
|
subject. */ |
|
| 2952 |
|
|
| 2953 |
REPEATNOTCHAR: |
REPEATNOTCHAR: |
|
if (min > md->end_subject - eptr) RRETURN(MATCH_NOMATCH); |
|
| 2954 |
fc = *ecode++; |
fc = *ecode++; |
| 2955 |
|
|
| 2956 |
/* The code is duplicated for the caseless and caseful cases, for speed, |
/* The code is duplicated for the caseless and caseful cases, for speed, |
| 2975 |
register unsigned int d; |
register unsigned int d; |
| 2976 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 2977 |
{ |
{ |
| 2978 |
|
if (eptr >= md->end_subject) |
| 2979 |
|
{ |
| 2980 |
|
SCHECK_PARTIAL(); |
| 2981 |
|
RRETURN(MATCH_NOMATCH); |
| 2982 |
|
} |
| 2983 |
GETCHARINC(d, eptr); |
GETCHARINC(d, eptr); |
| 2984 |
if (d < 256) d = md->lcc[d]; |
if (d < 256) d = md->lcc[d]; |
| 2985 |
if (fc == d) RRETURN(MATCH_NOMATCH); |
if (fc == d) RRETURN(MATCH_NOMATCH); |
| 2991 |
/* Not UTF-8 mode */ |
/* Not UTF-8 mode */ |
| 2992 |
{ |
{ |
| 2993 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 2994 |
if (fc == md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH); |
{ |
| 2995 |
} |
if (eptr >= md->end_subject) |
| 2996 |
|
{ |
| 2997 |
|
SCHECK_PARTIAL(); |
| 2998 |
|
RRETURN(MATCH_NOMATCH); |
| 2999 |
|
} |
| 3000 |
|
if (fc == md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH); |
| 3001 |
|
} |
| 3002 |
|
} |
| 3003 |
|
|
| 3004 |
if (min == max) continue; |
if (min == max) continue; |
| 3005 |
|
|
| 3006 |
if (minimize) |
if (minimize) |
| 3014 |
{ |
{ |
| 3015 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM28); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM28); |
| 3016 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 3017 |
|
if (fi >= max) RRETURN(MATCH_NOMATCH); |
| 3018 |
|
if (eptr >= md->end_subject) |
| 3019 |
|
{ |
| 3020 |
|
SCHECK_PARTIAL(); |
| 3021 |
|
RRETURN(MATCH_NOMATCH); |
| 3022 |
|
} |
| 3023 |
GETCHARINC(d, eptr); |
GETCHARINC(d, eptr); |
| 3024 |
if (d < 256) d = md->lcc[d]; |
if (d < 256) d = md->lcc[d]; |
| 3025 |
if (fi >= max || eptr >= md->end_subject || fc == d) |
if (fc == d) RRETURN(MATCH_NOMATCH); |
|
RRETURN(MATCH_NOMATCH); |
|
| 3026 |
} |
} |
| 3027 |
} |
} |
| 3028 |
else |
else |
| 3033 |
{ |
{ |
| 3034 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM29); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM29); |
| 3035 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 3036 |
if (fi >= max || eptr >= md->end_subject || fc == md->lcc[*eptr++]) |
if (fi >= max) RRETURN(MATCH_NOMATCH); |
| 3037 |
|
if (eptr >= md->end_subject) |
| 3038 |
|
{ |
| 3039 |
|
SCHECK_PARTIAL(); |
| 3040 |
RRETURN(MATCH_NOMATCH); |
RRETURN(MATCH_NOMATCH); |
| 3041 |
|
} |
| 3042 |
|
if (fc == md->lcc[*eptr++]) RRETURN(MATCH_NOMATCH); |
| 3043 |
} |
} |
| 3044 |
} |
} |
| 3045 |
/* Control never gets here */ |
/* Control never gets here */ |
| 3059 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 3060 |
{ |
{ |
| 3061 |
int len = 1; |
int len = 1; |
| 3062 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 3063 |
|
{ |
| 3064 |
|
SCHECK_PARTIAL(); |
| 3065 |
|
break; |
| 3066 |
|
} |
| 3067 |
GETCHARLEN(d, eptr, len); |
GETCHARLEN(d, eptr, len); |
| 3068 |
if (d < 256) d = md->lcc[d]; |
if (d < 256) d = md->lcc[d]; |
| 3069 |
if (fc == d) break; |
if (fc == d) break; |
| 3084 |
{ |
{ |
| 3085 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 3086 |
{ |
{ |
| 3087 |
if (eptr >= md->end_subject || fc == md->lcc[*eptr]) break; |
if (eptr >= md->end_subject) |
| 3088 |
|
{ |
| 3089 |
|
SCHECK_PARTIAL(); |
| 3090 |
|
break; |
| 3091 |
|
} |
| 3092 |
|
if (fc == md->lcc[*eptr]) break; |
| 3093 |
eptr++; |
eptr++; |
| 3094 |
} |
} |
| 3095 |
if (possessive) continue; |
if (possessive) continue; |
| 3117 |
register unsigned int d; |
register unsigned int d; |
| 3118 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3119 |
{ |
{ |
| 3120 |
|
if (eptr >= md->end_subject) |
| 3121 |
|
{ |
| 3122 |
|
SCHECK_PARTIAL(); |
| 3123 |
|
RRETURN(MATCH_NOMATCH); |
| 3124 |
|
} |
| 3125 |
GETCHARINC(d, eptr); |
GETCHARINC(d, eptr); |
| 3126 |
if (fc == d) RRETURN(MATCH_NOMATCH); |
if (fc == d) RRETURN(MATCH_NOMATCH); |
| 3127 |
} |
} |
| 3131 |
/* Not UTF-8 mode */ |
/* Not UTF-8 mode */ |
| 3132 |
{ |
{ |
| 3133 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3134 |
|
{ |
| 3135 |
|
if (eptr >= md->end_subject) |
| 3136 |
|
{ |
| 3137 |
|
SCHECK_PARTIAL(); |
| 3138 |
|
RRETURN(MATCH_NOMATCH); |
| 3139 |
|
} |
| 3140 |
if (fc == *eptr++) RRETURN(MATCH_NOMATCH); |
if (fc == *eptr++) RRETURN(MATCH_NOMATCH); |
| 3141 |
|
} |
| 3142 |
} |
} |
| 3143 |
|
|
| 3144 |
if (min == max) continue; |
if (min == max) continue; |
| 3154 |
{ |
{ |
| 3155 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM32); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM32); |
| 3156 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 3157 |
GETCHARINC(d, eptr); |
if (fi >= max) RRETURN(MATCH_NOMATCH); |
| 3158 |
if (fi >= max || eptr >= md->end_subject || fc == d) |
if (eptr >= md->end_subject) |
| 3159 |
|
{ |
| 3160 |
|
SCHECK_PARTIAL(); |
| 3161 |
RRETURN(MATCH_NOMATCH); |
RRETURN(MATCH_NOMATCH); |
| 3162 |
|
} |
| 3163 |
|
GETCHARINC(d, eptr); |
| 3164 |
|
if (fc == d) RRETURN(MATCH_NOMATCH); |
| 3165 |
} |
} |
| 3166 |
} |
} |
| 3167 |
else |
else |
| 3172 |
{ |
{ |
| 3173 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM33); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM33); |
| 3174 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 3175 |
if (fi >= max || eptr >= md->end_subject || fc == *eptr++) |
if (fi >= max) RRETURN(MATCH_NOMATCH); |
| 3176 |
|
if (eptr >= md->end_subject) |
| 3177 |
|
{ |
| 3178 |
|
SCHECK_PARTIAL(); |
| 3179 |
RRETURN(MATCH_NOMATCH); |
RRETURN(MATCH_NOMATCH); |
| 3180 |
|
} |
| 3181 |
|
if (fc == *eptr++) RRETURN(MATCH_NOMATCH); |
| 3182 |
} |
} |
| 3183 |
} |
} |
| 3184 |
/* Control never gets here */ |
/* Control never gets here */ |
| 3198 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 3199 |
{ |
{ |
| 3200 |
int len = 1; |
int len = 1; |
| 3201 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 3202 |
|
{ |
| 3203 |
|
SCHECK_PARTIAL(); |
| 3204 |
|
break; |
| 3205 |
|
} |
| 3206 |
GETCHARLEN(d, eptr, len); |
GETCHARLEN(d, eptr, len); |
| 3207 |
if (fc == d) break; |
if (fc == d) break; |
| 3208 |
eptr += len; |
eptr += len; |
| 3222 |
{ |
{ |
| 3223 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 3224 |
{ |
{ |
| 3225 |
if (eptr >= md->end_subject || fc == *eptr) break; |
if (eptr >= md->end_subject) |
| 3226 |
|
{ |
| 3227 |
|
SCHECK_PARTIAL(); |
| 3228 |
|
break; |
| 3229 |
|
} |
| 3230 |
|
if (fc == *eptr) break; |
| 3231 |
eptr++; |
eptr++; |
| 3232 |
} |
} |
| 3233 |
if (possessive) continue; |
if (possessive) continue; |
| 3321 |
|
|
| 3322 |
/* First, ensure the minimum number of matches are present. Use inline |
/* First, ensure the minimum number of matches are present. Use inline |
| 3323 |
code for maximizing the speed, and do the type test once at the start |
code for maximizing the speed, and do the type test once at the start |
| 3324 |
(i.e. keep it out of the loop). Also we can test that there are at least |
(i.e. keep it out of the loop). Separate the UTF-8 code completely as that |
|
the minimum number of bytes before we start. This isn't as effective in |
|
|
UTF-8 mode, but it does no harm. Separate the UTF-8 code completely as that |
|
| 3325 |
is tidier. Also separate the UCP code, which can be the same for both UTF-8 |
is tidier. Also separate the UCP code, which can be the same for both UTF-8 |
| 3326 |
and single-bytes. */ |
and single-bytes. */ |
| 3327 |
|
|
|
if (min > md->end_subject - eptr) RRETURN(MATCH_NOMATCH); |
|
| 3328 |
if (min > 0) |
if (min > 0) |
| 3329 |
{ |
{ |
| 3330 |
#ifdef SUPPORT_UCP |
#ifdef SUPPORT_UCP |
| 3336 |
if (prop_fail_result) RRETURN(MATCH_NOMATCH); |
if (prop_fail_result) RRETURN(MATCH_NOMATCH); |
| 3337 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3338 |
{ |
{ |
| 3339 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 3340 |
|
{ |
| 3341 |
|
SCHECK_PARTIAL(); |
| 3342 |
|
RRETURN(MATCH_NOMATCH); |
| 3343 |
|
} |
| 3344 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 3345 |
} |
} |
| 3346 |
break; |
break; |
| 3348 |
case PT_LAMP: |
case PT_LAMP: |
| 3349 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3350 |
{ |
{ |
| 3351 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 3352 |
|
{ |
| 3353 |
|
SCHECK_PARTIAL(); |
| 3354 |
|
RRETURN(MATCH_NOMATCH); |
| 3355 |
|
} |
| 3356 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 3357 |
prop_chartype = UCD_CHARTYPE(c); |
prop_chartype = UCD_CHARTYPE(c); |
| 3358 |
if ((prop_chartype == ucp_Lu || |
if ((prop_chartype == ucp_Lu || |
| 3365 |
case PT_GC: |
case PT_GC: |
| 3366 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3367 |
{ |
{ |
| 3368 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 3369 |
|
{ |
| 3370 |
|
SCHECK_PARTIAL(); |
| 3371 |
|
RRETURN(MATCH_NOMATCH); |
| 3372 |
|
} |
| 3373 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 3374 |
prop_category = UCD_CATEGORY(c); |
prop_category = UCD_CATEGORY(c); |
| 3375 |
if ((prop_category == prop_value) == prop_fail_result) |
if ((prop_category == prop_value) == prop_fail_result) |
| 3380 |
case PT_PC: |
case PT_PC: |
| 3381 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3382 |
{ |
{ |
| 3383 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 3384 |
|
{ |
| 3385 |
|
SCHECK_PARTIAL(); |
| 3386 |
|
RRETURN(MATCH_NOMATCH); |
| 3387 |
|
} |
| 3388 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 3389 |
prop_chartype = UCD_CHARTYPE(c); |
prop_chartype = UCD_CHARTYPE(c); |
| 3390 |
if ((prop_chartype == prop_value) == prop_fail_result) |
if ((prop_chartype == prop_value) == prop_fail_result) |
| 3395 |
case PT_SC: |
case PT_SC: |
| 3396 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3397 |
{ |
{ |
| 3398 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 3399 |
|
{ |
| 3400 |
|
SCHECK_PARTIAL(); |
| 3401 |
|
RRETURN(MATCH_NOMATCH); |
| 3402 |
|
} |
| 3403 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 3404 |
prop_script = UCD_SCRIPT(c); |
prop_script = UCD_SCRIPT(c); |
| 3405 |
if ((prop_script == prop_value) == prop_fail_result) |
if ((prop_script == prop_value) == prop_fail_result) |
| 3419 |
{ |
{ |
| 3420 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3421 |
{ |
{ |
| 3422 |
|
if (eptr >= md->end_subject) |
| 3423 |
|
{ |
| 3424 |
|
SCHECK_PARTIAL(); |
| 3425 |
|
RRETURN(MATCH_NOMATCH); |
| 3426 |
|
} |
| 3427 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 3428 |
prop_category = UCD_CATEGORY(c); |
prop_category = UCD_CATEGORY(c); |
| 3429 |
if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH); |
if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH); |
| 3430 |
while (eptr < md->end_subject) |
while (eptr < md->end_subject) |
| 3431 |
{ |
{ |
| 3432 |
int len = 1; |
int len = 1; |
| 3433 |
if (!utf8) c = *eptr; else |
if (!utf8) c = *eptr; |
| 3434 |
{ |
else { GETCHARLEN(c, eptr, len); } |
|
GETCHARLEN(c, eptr, len); |
|
|
} |
|
| 3435 |
prop_category = UCD_CATEGORY(c); |
prop_category = UCD_CATEGORY(c); |
| 3436 |
if (prop_category != ucp_M) break; |
if (prop_category != ucp_M) break; |
| 3437 |
eptr += len; |
eptr += len; |
| 3450 |
case OP_ANY: |
case OP_ANY: |
| 3451 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3452 |
{ |
{ |
| 3453 |
if (eptr >= md->end_subject || IS_NEWLINE(eptr)) |
if (eptr >= md->end_subject) |
| 3454 |
|
{ |
| 3455 |
|
SCHECK_PARTIAL(); |
| 3456 |
RRETURN(MATCH_NOMATCH); |
RRETURN(MATCH_NOMATCH); |
| 3457 |
|
} |
| 3458 |
|
if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH); |
| 3459 |
eptr++; |
eptr++; |
| 3460 |
while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++; |
while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++; |
| 3461 |
} |
} |
| 3464 |
case OP_ALLANY: |
case OP_ALLANY: |
| 3465 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3466 |
{ |
{ |
| 3467 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 3468 |
|
{ |
| 3469 |
|
SCHECK_PARTIAL(); |
| 3470 |
|
RRETURN(MATCH_NOMATCH); |
| 3471 |
|
} |
| 3472 |
eptr++; |
eptr++; |
| 3473 |
while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++; |
while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++; |
| 3474 |
} |
} |
| 3475 |
break; |
break; |
| 3476 |
|
|
| 3477 |
case OP_ANYBYTE: |
case OP_ANYBYTE: |
| 3478 |
|
if (eptr > md->end_subject - min) RRETURN(MATCH_NOMATCH); |
| 3479 |
eptr += min; |
eptr += min; |
| 3480 |
break; |
break; |
| 3481 |
|
|
| 3482 |
case OP_ANYNL: |
case OP_ANYNL: |
| 3483 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3484 |
{ |
{ |
| 3485 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 3486 |
|
{ |
| 3487 |
|
SCHECK_PARTIAL(); |
| 3488 |
|
RRETURN(MATCH_NOMATCH); |
| 3489 |
|
} |
| 3490 |
GETCHARINC(c, eptr); |
GETCHARINC(c, eptr); |
| 3491 |
switch(c) |
switch(c) |
| 3492 |
{ |
{ |
| 3512 |
case OP_NOT_HSPACE: |
case OP_NOT_HSPACE: |
| 3513 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3514 |
{ |
{ |
| 3515 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 3516 |
|
{ |
| 3517 |
|
SCHECK_PARTIAL(); |
| 3518 |
|
RRETURN(MATCH_NOMATCH); |
| 3519 |
|
} |
| 3520 |
GETCHARINC(c, eptr); |
GETCHARINC(c, eptr); |
| 3521 |
switch(c) |
switch(c) |
| 3522 |
{ |
{ |
| 3548 |
case OP_HSPACE: |
case OP_HSPACE: |
| 3549 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3550 |
{ |
{ |
| 3551 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 3552 |
|
{ |
| 3553 |
|
SCHECK_PARTIAL(); |
| 3554 |
|
RRETURN(MATCH_NOMATCH); |
| 3555 |
|
} |
| 3556 |
GETCHARINC(c, eptr); |
GETCHARINC(c, eptr); |
| 3557 |
switch(c) |
switch(c) |
| 3558 |
{ |
{ |
| 3584 |
case OP_NOT_VSPACE: |
case OP_NOT_VSPACE: |
| 3585 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3586 |
{ |
{ |
| 3587 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 3588 |
|
{ |
| 3589 |
|
SCHECK_PARTIAL(); |
| 3590 |
|
RRETURN(MATCH_NOMATCH); |
| 3591 |
|
} |
| 3592 |
GETCHARINC(c, eptr); |
GETCHARINC(c, eptr); |
| 3593 |
switch(c) |
switch(c) |
| 3594 |
{ |
{ |
| 3608 |
case OP_VSPACE: |
case OP_VSPACE: |
| 3609 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3610 |
{ |
{ |
| 3611 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 3612 |
|
{ |
| 3613 |
|
SCHECK_PARTIAL(); |
| 3614 |
|
RRETURN(MATCH_NOMATCH); |
| 3615 |
|
} |
| 3616 |
GETCHARINC(c, eptr); |
GETCHARINC(c, eptr); |
| 3617 |
switch(c) |
switch(c) |
| 3618 |
{ |
{ |
| 3632 |
case OP_NOT_DIGIT: |
case OP_NOT_DIGIT: |
| 3633 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3634 |
{ |
{ |
| 3635 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 3636 |
|
{ |
| 3637 |
|
SCHECK_PARTIAL(); |
| 3638 |
|
RRETURN(MATCH_NOMATCH); |
| 3639 |
|
} |
| 3640 |
GETCHARINC(c, eptr); |
GETCHARINC(c, eptr); |
| 3641 |
if (c < 128 && (md->ctypes[c] & ctype_digit) != 0) |
if (c < 128 && (md->ctypes[c] & ctype_digit) != 0) |
| 3642 |
RRETURN(MATCH_NOMATCH); |
RRETURN(MATCH_NOMATCH); |
| 3646 |
case OP_DIGIT: |
case OP_DIGIT: |
| 3647 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3648 |
{ |
{ |
| 3649 |
if (eptr >= md->end_subject || |
if (eptr >= md->end_subject) |
| 3650 |
*eptr >= 128 || (md->ctypes[*eptr++] & ctype_digit) == 0) |
{ |
| 3651 |
|
SCHECK_PARTIAL(); |
| 3652 |
|
RRETURN(MATCH_NOMATCH); |
| 3653 |
|
} |
| 3654 |
|
if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_digit) == 0) |
| 3655 |
RRETURN(MATCH_NOMATCH); |
RRETURN(MATCH_NOMATCH); |
| 3656 |
/* No need to skip more bytes - we know it's a 1-byte character */ |
/* No need to skip more bytes - we know it's a 1-byte character */ |
| 3657 |
} |
} |
| 3660 |
case OP_NOT_WHITESPACE: |
case OP_NOT_WHITESPACE: |
| 3661 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3662 |
{ |
{ |
| 3663 |
if (eptr >= md->end_subject || |
if (eptr >= md->end_subject) |
| 3664 |
(*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0)) |
{ |
| 3665 |
|
SCHECK_PARTIAL(); |
| 3666 |
|
RRETURN(MATCH_NOMATCH); |
| 3667 |
|
} |
| 3668 |
|
if (*eptr < 128 && (md->ctypes[*eptr] & ctype_space) != 0) |
| 3669 |
RRETURN(MATCH_NOMATCH); |
RRETURN(MATCH_NOMATCH); |
| 3670 |
while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80); |
while (++eptr < md->end_subject && (*eptr & 0xc0) == 0x80); |
| 3671 |
} |
} |
| 3674 |
case OP_WHITESPACE: |
case OP_WHITESPACE: |
| 3675 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3676 |
{ |
{ |
| 3677 |
if (eptr >= md->end_subject || |
if (eptr >= md->end_subject) |
| 3678 |
*eptr >= 128 || (md->ctypes[*eptr++] & ctype_space) == 0) |
{ |
| 3679 |
|
SCHECK_PARTIAL(); |
| 3680 |
|
RRETURN(MATCH_NOMATCH); |
| 3681 |
|
} |
| 3682 |
|
if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_space) == 0) |
| 3683 |
RRETURN(MATCH_NOMATCH); |
RRETURN(MATCH_NOMATCH); |
| 3684 |
/* No need to skip more bytes - we know it's a 1-byte character */ |
/* No need to skip more bytes - we know it's a 1-byte character */ |
| 3685 |
} |
} |
| 3698 |
case OP_WORDCHAR: |
case OP_WORDCHAR: |
| 3699 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3700 |
{ |
{ |
| 3701 |
if (eptr >= md->end_subject || |
if (eptr >= md->end_subject) |
| 3702 |
*eptr >= 128 || (md->ctypes[*eptr++] & ctype_word) == 0) |
{ |
| 3703 |
|
SCHECK_PARTIAL(); |
| 3704 |
|
RRETURN(MATCH_NOMATCH); |
| 3705 |
|
} |
| 3706 |
|
if (*eptr >= 128 || (md->ctypes[*eptr++] & ctype_word) == 0) |
| 3707 |
RRETURN(MATCH_NOMATCH); |
RRETURN(MATCH_NOMATCH); |
| 3708 |
/* No need to skip more bytes - we know it's a 1-byte character */ |
/* No need to skip more bytes - we know it's a 1-byte character */ |
| 3709 |
} |
} |
| 3717 |
#endif /* SUPPORT_UTF8 */ |
#endif /* SUPPORT_UTF8 */ |
| 3718 |
|
|
| 3719 |
/* Code for the non-UTF-8 case for minimum matching of operators other |
/* Code for the non-UTF-8 case for minimum matching of operators other |
| 3720 |
than OP_PROP and OP_NOTPROP. We can assume that there are the minimum |
than OP_PROP and OP_NOTPROP. */ |
|
number of bytes present, as this was tested above. */ |
|
| 3721 |
|
|
| 3722 |
switch(ctype) |
switch(ctype) |
| 3723 |
{ |
{ |
| 3724 |
case OP_ANY: |
case OP_ANY: |
| 3725 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3726 |
{ |
{ |
| 3727 |
|
if (eptr >= md->end_subject) |
| 3728 |
|
{ |
| 3729 |
|
SCHECK_PARTIAL(); |
| 3730 |
|
RRETURN(MATCH_NOMATCH); |
| 3731 |
|
} |
| 3732 |
if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH); |
if (IS_NEWLINE(eptr)) RRETURN(MATCH_NOMATCH); |
| 3733 |
eptr++; |
eptr++; |
| 3734 |
} |
} |
| 3735 |
break; |
break; |
| 3736 |
|
|
| 3737 |
case OP_ALLANY: |
case OP_ALLANY: |
| 3738 |
|
if (eptr > md->end_subject - min) |
| 3739 |
|
{ |
| 3740 |
|
SCHECK_PARTIAL(); |
| 3741 |
|
RRETURN(MATCH_NOMATCH); |
| 3742 |
|
} |
| 3743 |
eptr += min; |
eptr += min; |
| 3744 |
break; |
break; |
| 3745 |
|
|
| 3746 |
case OP_ANYBYTE: |
case OP_ANYBYTE: |
| 3747 |
|
if (eptr > md->end_subject - min) |
| 3748 |
|
{ |
| 3749 |
|
SCHECK_PARTIAL(); |
| 3750 |
|
RRETURN(MATCH_NOMATCH); |
| 3751 |
|
} |
| 3752 |
eptr += min; |
eptr += min; |
| 3753 |
break; |
break; |
| 3754 |
|
|
|
/* Because of the CRLF case, we can't assume the minimum number of |
|
|
bytes are present in this case. */ |
|
|
|
|
| 3755 |
case OP_ANYNL: |
case OP_ANYNL: |
| 3756 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3757 |
{ |
{ |
| 3758 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 3759 |
|
{ |
| 3760 |
|
SCHECK_PARTIAL(); |
| 3761 |
|
RRETURN(MATCH_NOMATCH); |
| 3762 |
|
} |
| 3763 |
switch(*eptr++) |
switch(*eptr++) |
| 3764 |
{ |
{ |
| 3765 |
default: RRETURN(MATCH_NOMATCH); |
default: RRETURN(MATCH_NOMATCH); |
| 3781 |
case OP_NOT_HSPACE: |
case OP_NOT_HSPACE: |
| 3782 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3783 |
{ |
{ |
| 3784 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 3785 |
|
{ |
| 3786 |
|
SCHECK_PARTIAL(); |
| 3787 |
|
RRETURN(MATCH_NOMATCH); |
| 3788 |
|
} |
| 3789 |
switch(*eptr++) |
switch(*eptr++) |
| 3790 |
{ |
{ |
| 3791 |
default: break; |
default: break; |
| 3800 |
case OP_HSPACE: |
case OP_HSPACE: |
| 3801 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3802 |
{ |
{ |
| 3803 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 3804 |
|
{ |
| 3805 |
|
SCHECK_PARTIAL(); |
| 3806 |
|
RRETURN(MATCH_NOMATCH); |
| 3807 |
|
} |
| 3808 |
switch(*eptr++) |
switch(*eptr++) |
| 3809 |
{ |
{ |
| 3810 |
default: RRETURN(MATCH_NOMATCH); |
default: RRETURN(MATCH_NOMATCH); |
| 3819 |
case OP_NOT_VSPACE: |
case OP_NOT_VSPACE: |
| 3820 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3821 |
{ |
{ |
| 3822 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 3823 |
|
{ |
| 3824 |
|
SCHECK_PARTIAL(); |
| 3825 |
|
RRETURN(MATCH_NOMATCH); |
| 3826 |
|
} |
| 3827 |
switch(*eptr++) |
switch(*eptr++) |
| 3828 |
{ |
{ |
| 3829 |
default: break; |
default: break; |
| 3840 |
case OP_VSPACE: |
case OP_VSPACE: |
| 3841 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3842 |
{ |
{ |
| 3843 |
if (eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (eptr >= md->end_subject) |
| 3844 |
|
{ |
| 3845 |
|
SCHECK_PARTIAL(); |
| 3846 |
|
RRETURN(MATCH_NOMATCH); |
| 3847 |
|
} |
| 3848 |
switch(*eptr++) |
switch(*eptr++) |
| 3849 |
{ |
{ |
| 3850 |
default: RRETURN(MATCH_NOMATCH); |
default: RRETURN(MATCH_NOMATCH); |
| 3860 |
|
|
| 3861 |
case OP_NOT_DIGIT: |
case OP_NOT_DIGIT: |
| 3862 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3863 |
|
{ |
| 3864 |
|
if (eptr >= md->end_subject) |
| 3865 |
|
{ |
| 3866 |
|
SCHECK_PARTIAL(); |
| 3867 |
|
RRETURN(MATCH_NOMATCH); |
| 3868 |
|
} |
| 3869 |
if ((md->ctypes[*eptr++] & ctype_digit) != 0) RRETURN(MATCH_NOMATCH); |
if ((md->ctypes[*eptr++] & ctype_digit) != 0) RRETURN(MATCH_NOMATCH); |
| 3870 |
|
} |
| 3871 |
break; |
break; |
| 3872 |
|
|
| 3873 |
case OP_DIGIT: |
case OP_DIGIT: |
| 3874 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3875 |
|
{ |
| 3876 |
|
if (eptr >= md->end_subject) |
| 3877 |
|
{ |
| 3878 |
|
SCHECK_PARTIAL(); |
| 3879 |
|
RRETURN(MATCH_NOMATCH); |
| 3880 |
|
} |
| 3881 |
if ((md->ctypes[*eptr++] & ctype_digit) == 0) RRETURN(MATCH_NOMATCH); |
if ((md->ctypes[*eptr++] & ctype_digit) == 0) RRETURN(MATCH_NOMATCH); |
| 3882 |
|
} |
| 3883 |
break; |
break; |
| 3884 |
|
|
| 3885 |
case OP_NOT_WHITESPACE: |
case OP_NOT_WHITESPACE: |
| 3886 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3887 |
|
{ |
| 3888 |
|
if (eptr >= md->end_subject) |
| 3889 |
|
{ |
| 3890 |
|
SCHECK_PARTIAL(); |
| 3891 |
|
RRETURN(MATCH_NOMATCH); |
| 3892 |
|
} |
| 3893 |
if ((md->ctypes[*eptr++] & ctype_space) != 0) RRETURN(MATCH_NOMATCH); |
if ((md->ctypes[*eptr++] & ctype_space) != 0) RRETURN(MATCH_NOMATCH); |
| 3894 |
|
} |
| 3895 |
break; |
break; |
| 3896 |
|
|
| 3897 |
case OP_WHITESPACE: |
case OP_WHITESPACE: |
| 3898 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3899 |
|
{ |
| 3900 |
|
if (eptr >= md->end_subject) |
| 3901 |
|
{ |
| 3902 |
|
SCHECK_PARTIAL(); |
| 3903 |
|
RRETURN(MATCH_NOMATCH); |
| 3904 |
|
} |
| 3905 |
if ((md->ctypes[*eptr++] & ctype_space) == 0) RRETURN(MATCH_NOMATCH); |
if ((md->ctypes[*eptr++] & ctype_space) == 0) RRETURN(MATCH_NOMATCH); |
| 3906 |
|
} |
| 3907 |
break; |
break; |
| 3908 |
|
|
| 3909 |
case OP_NOT_WORDCHAR: |
case OP_NOT_WORDCHAR: |
| 3910 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3911 |
|
{ |
| 3912 |
|
if (eptr >= md->end_subject) |
| 3913 |
|
{ |
| 3914 |
|
SCHECK_PARTIAL(); |
| 3915 |
|
RRETURN(MATCH_NOMATCH); |
| 3916 |
|
} |
| 3917 |
if ((md->ctypes[*eptr++] & ctype_word) != 0) |
if ((md->ctypes[*eptr++] & ctype_word) != 0) |
| 3918 |
RRETURN(MATCH_NOMATCH); |
RRETURN(MATCH_NOMATCH); |
| 3919 |
|
} |
| 3920 |
break; |
break; |
| 3921 |
|
|
| 3922 |
case OP_WORDCHAR: |
case OP_WORDCHAR: |
| 3923 |
for (i = 1; i <= min; i++) |
for (i = 1; i <= min; i++) |
| 3924 |
|
{ |
| 3925 |
|
if (eptr >= md->end_subject) |
| 3926 |
|
{ |
| 3927 |
|
SCHECK_PARTIAL(); |
| 3928 |
|
RRETURN(MATCH_NOMATCH); |
| 3929 |
|
} |
| 3930 |
if ((md->ctypes[*eptr++] & ctype_word) == 0) |
if ((md->ctypes[*eptr++] & ctype_word) == 0) |
| 3931 |
RRETURN(MATCH_NOMATCH); |
RRETURN(MATCH_NOMATCH); |
| 3932 |
|
} |
| 3933 |
break; |
break; |
| 3934 |
|
|
| 3935 |
default: |
default: |
| 3957 |
{ |
{ |
| 3958 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM36); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM36); |
| 3959 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 3960 |
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (fi >= max) RRETURN(MATCH_NOMATCH); |
| 3961 |
|
if (eptr >= md->end_subject) |
| 3962 |
|
{ |
| 3963 |
|
SCHECK_PARTIAL(); |
| 3964 |
|
RRETURN(MATCH_NOMATCH); |
| 3965 |
|
} |
| 3966 |
GETCHARINC(c, eptr); |
GETCHARINC(c, eptr); |
| 3967 |
if (prop_fail_result) RRETURN(MATCH_NOMATCH); |
if (prop_fail_result) RRETURN(MATCH_NOMATCH); |
| 3968 |
} |
} |
| 3973 |
{ |
{ |
| 3974 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM37); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM37); |
| 3975 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 3976 |
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (fi >= max) RRETURN(MATCH_NOMATCH); |
| 3977 |
|
if (eptr >= md->end_subject) |
| 3978 |
|
{ |
| 3979 |
|
SCHECK_PARTIAL(); |
| 3980 |
|
RRETURN(MATCH_NOMATCH); |
| 3981 |
|
} |
| 3982 |
GETCHARINC(c, eptr); |
GETCHARINC(c, eptr); |
| 3983 |
prop_chartype = UCD_CHARTYPE(c); |
prop_chartype = UCD_CHARTYPE(c); |
| 3984 |
if ((prop_chartype == ucp_Lu || |
if ((prop_chartype == ucp_Lu || |
| 3993 |
{ |
{ |
| 3994 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM38); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM38); |
| 3995 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 3996 |
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (fi >= max) RRETURN(MATCH_NOMATCH); |
| 3997 |
|
if (eptr >= md->end_subject) |
| 3998 |
|
{ |
| 3999 |
|
SCHECK_PARTIAL(); |
| 4000 |
|
RRETURN(MATCH_NOMATCH); |
| 4001 |
|
} |
| 4002 |
GETCHARINC(c, eptr); |
GETCHARINC(c, eptr); |
| 4003 |
prop_category = UCD_CATEGORY(c); |
prop_category = UCD_CATEGORY(c); |
| 4004 |
if ((prop_category == prop_value) == prop_fail_result) |
if ((prop_category == prop_value) == prop_fail_result) |
| 4011 |
{ |
{ |
| 4012 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM39); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM39); |
| 4013 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 4014 |
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (fi >= max) RRETURN(MATCH_NOMATCH); |
| 4015 |
|
if (eptr >= md->end_subject) |
| 4016 |
|
{ |
| 4017 |
|
SCHECK_PARTIAL(); |
| 4018 |
|
RRETURN(MATCH_NOMATCH); |
| 4019 |
|
} |
| 4020 |
GETCHARINC(c, eptr); |
GETCHARINC(c, eptr); |
| 4021 |
prop_chartype = UCD_CHARTYPE(c); |
prop_chartype = UCD_CHARTYPE(c); |
| 4022 |
if ((prop_chartype == prop_value) == prop_fail_result) |
if ((prop_chartype == prop_value) == prop_fail_result) |
| 4029 |
{ |
{ |
| 4030 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM40); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM40); |
| 4031 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 4032 |
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (fi >= max) RRETURN(MATCH_NOMATCH); |
| 4033 |
|
if (eptr >= md->end_subject) |
| 4034 |
|
{ |
| 4035 |
|
SCHECK_PARTIAL(); |
| 4036 |
|
RRETURN(MATCH_NOMATCH); |
| 4037 |
|
} |
| 4038 |
GETCHARINC(c, eptr); |
GETCHARINC(c, eptr); |
| 4039 |
prop_script = UCD_SCRIPT(c); |
prop_script = UCD_SCRIPT(c); |
| 4040 |
if ((prop_script == prop_value) == prop_fail_result) |
if ((prop_script == prop_value) == prop_fail_result) |
| 4056 |
{ |
{ |
| 4057 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM41); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM41); |
| 4058 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 4059 |
if (fi >= max || eptr >= md->end_subject) RRETURN(MATCH_NOMATCH); |
if (fi >= max) RRETURN(MATCH_NOMATCH); |
| 4060 |
|
if (eptr >= md->end_subject) |
| 4061 |
|
{ |
| 4062 |
|
SCHECK_PARTIAL(); |
| 4063 |
|
RRETURN(MATCH_NOMATCH); |
| 4064 |
|
} |
| 4065 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 4066 |
prop_category = UCD_CATEGORY(c); |
prop_category = UCD_CATEGORY(c); |
| 4067 |
if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH); |
if (prop_category == ucp_M) RRETURN(MATCH_NOMATCH); |
| 4068 |
while (eptr < md->end_subject) |
while (eptr < md->end_subject) |
| 4069 |
{ |
{ |
| 4070 |
int len = 1; |
int len = 1; |
| 4071 |
if (!utf8) c = *eptr; else |
if (!utf8) c = *eptr; |
| 4072 |
{ |
else { GETCHARLEN(c, eptr, len); } |
|
GETCHARLEN(c, eptr, len); |
|
|
} |
|
| 4073 |
prop_category = UCD_CATEGORY(c); |
prop_category = UCD_CATEGORY(c); |
| 4074 |
if (prop_category != ucp_M) break; |
if (prop_category != ucp_M) break; |
| 4075 |
eptr += len; |
eptr += len; |
| 4088 |
{ |
{ |
| 4089 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM42); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM42); |
| 4090 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 4091 |
if (fi >= max || eptr >= md->end_subject || |
if (fi >= max) RRETURN(MATCH_NOMATCH); |
| 4092 |
(ctype == OP_ANY && IS_NEWLINE(eptr))) |
if (eptr >= md->end_subject) |
| 4093 |
|
{ |
| 4094 |
|
SCHECK_PARTIAL(); |
| 4095 |
|
RRETURN(MATCH_NOMATCH); |
| 4096 |
|
} |
| 4097 |
|
if (ctype == OP_ANY && IS_NEWLINE(eptr)) |
| 4098 |
RRETURN(MATCH_NOMATCH); |
RRETURN(MATCH_NOMATCH); |
|
|
|
| 4099 |
GETCHARINC(c, eptr); |
GETCHARINC(c, eptr); |
| 4100 |
switch(ctype) |
switch(ctype) |
| 4101 |
{ |
{ |
| 4251 |
{ |
{ |
| 4252 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM43); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM43); |
| 4253 |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
if (rrc != MATCH_NOMATCH) RRETURN(rrc); |
| 4254 |
if (fi >= max || eptr >= md->end_subject || |
if (fi >= max) RRETURN(MATCH_NOMATCH); |
| 4255 |
(ctype == OP_ANY && IS_NEWLINE(eptr))) |
if (eptr >= md->end_subject) |
| 4256 |
|
{ |
| 4257 |
|
SCHECK_PARTIAL(); |
| 4258 |
|
RRETURN(MATCH_NOMATCH); |
| 4259 |
|
} |
| 4260 |
|
if (ctype == OP_ANY && IS_NEWLINE(eptr)) |
| 4261 |
RRETURN(MATCH_NOMATCH); |
RRETURN(MATCH_NOMATCH); |
|
|
|
| 4262 |
c = *eptr++; |
c = *eptr++; |
| 4263 |
switch(ctype) |
switch(ctype) |
| 4264 |
{ |
{ |
| 4383 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4384 |
{ |
{ |
| 4385 |
int len = 1; |
int len = 1; |
| 4386 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4387 |
|
{ |
| 4388 |
|
SCHECK_PARTIAL(); |
| 4389 |
|
break; |
| 4390 |
|
} |
| 4391 |
GETCHARLEN(c, eptr, len); |
GETCHARLEN(c, eptr, len); |
| 4392 |
if (prop_fail_result) break; |
if (prop_fail_result) break; |
| 4393 |
eptr+= len; |
eptr+= len; |
| 4398 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4399 |
{ |
{ |
| 4400 |
int len = 1; |
int len = 1; |
| 4401 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4402 |
|
{ |
| 4403 |
|
SCHECK_PARTIAL(); |
| 4404 |
|
break; |
| 4405 |
|
} |
| 4406 |
GETCHARLEN(c, eptr, len); |
GETCHARLEN(c, eptr, len); |
| 4407 |
prop_chartype = UCD_CHARTYPE(c); |
prop_chartype = UCD_CHARTYPE(c); |
| 4408 |
if ((prop_chartype == ucp_Lu || |
if ((prop_chartype == ucp_Lu || |
| 4417 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4418 |
{ |
{ |
| 4419 |
int len = 1; |
int len = 1; |
| 4420 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4421 |
|
{ |
| 4422 |
|
SCHECK_PARTIAL(); |
| 4423 |
|
break; |
| 4424 |
|
} |
| 4425 |
GETCHARLEN(c, eptr, len); |
GETCHARLEN(c, eptr, len); |
| 4426 |
prop_category = UCD_CATEGORY(c); |
prop_category = UCD_CATEGORY(c); |
| 4427 |
if ((prop_category == prop_value) == prop_fail_result) |
if ((prop_category == prop_value) == prop_fail_result) |
| 4434 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4435 |
{ |
{ |
| 4436 |
int len = 1; |
int len = 1; |
| 4437 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4438 |
|
{ |
| 4439 |
|
SCHECK_PARTIAL(); |
| 4440 |
|
break; |
| 4441 |
|
} |
| 4442 |
GETCHARLEN(c, eptr, len); |
GETCHARLEN(c, eptr, len); |
| 4443 |
prop_chartype = UCD_CHARTYPE(c); |
prop_chartype = UCD_CHARTYPE(c); |
| 4444 |
if ((prop_chartype == prop_value) == prop_fail_result) |
if ((prop_chartype == prop_value) == prop_fail_result) |
| 4451 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4452 |
{ |
{ |
| 4453 |
int len = 1; |
int len = 1; |
| 4454 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4455 |
|
{ |
| 4456 |
|
SCHECK_PARTIAL(); |
| 4457 |
|
break; |
| 4458 |
|
} |
| 4459 |
GETCHARLEN(c, eptr, len); |
GETCHARLEN(c, eptr, len); |
| 4460 |
prop_script = UCD_SCRIPT(c); |
prop_script = UCD_SCRIPT(c); |
| 4461 |
if ((prop_script == prop_value) == prop_fail_result) |
if ((prop_script == prop_value) == prop_fail_result) |
| 4484 |
{ |
{ |
| 4485 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4486 |
{ |
{ |
| 4487 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4488 |
|
{ |
| 4489 |
|
SCHECK_PARTIAL(); |
| 4490 |
|
break; |
| 4491 |
|
} |
| 4492 |
GETCHARINCTEST(c, eptr); |
GETCHARINCTEST(c, eptr); |
| 4493 |
prop_category = UCD_CATEGORY(c); |
prop_category = UCD_CATEGORY(c); |
| 4494 |
if (prop_category == ucp_M) break; |
if (prop_category == ucp_M) break; |
| 4508 |
/* eptr is now past the end of the maximum run */ |
/* eptr is now past the end of the maximum run */ |
| 4509 |
|
|
| 4510 |
if (possessive) continue; |
if (possessive) continue; |
| 4511 |
|
|
| 4512 |
for(;;) |
for(;;) |
| 4513 |
{ |
{ |
| 4514 |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM45); |
RMATCH(eptr, ecode, offset_top, md, ims, eptrb, 0, RM45); |
| 4544 |
{ |
{ |
| 4545 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4546 |
{ |
{ |
| 4547 |
if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break; |
if (eptr >= md->end_subject) |
| 4548 |
|
{ |
| 4549 |
|
SCHECK_PARTIAL(); |
| 4550 |
|
break; |
| 4551 |
|
} |
| 4552 |
|
if (IS_NEWLINE(eptr)) break; |
| 4553 |
eptr++; |
eptr++; |
| 4554 |
while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++; |
while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++; |
| 4555 |
} |
} |
| 4561 |
{ |
{ |
| 4562 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4563 |
{ |
{ |
| 4564 |
if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break; |
if (eptr >= md->end_subject) |
| 4565 |
|
{ |
| 4566 |
|
SCHECK_PARTIAL(); |
| 4567 |
|
break; |
| 4568 |
|
} |
| 4569 |
|
if (IS_NEWLINE(eptr)) break; |
| 4570 |
eptr++; |
eptr++; |
| 4571 |
while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++; |
while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++; |
| 4572 |
} |
} |
| 4578 |
{ |
{ |
| 4579 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4580 |
{ |
{ |
| 4581 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4582 |
|
{ |
| 4583 |
|
SCHECK_PARTIAL(); |
| 4584 |
|
break; |
| 4585 |
|
} |
| 4586 |
eptr++; |
eptr++; |
| 4587 |
while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++; |
while (eptr < md->end_subject && (*eptr & 0xc0) == 0x80) eptr++; |
| 4588 |
} |
} |
| 4595 |
case OP_ANYBYTE: |
case OP_ANYBYTE: |
| 4596 |
c = max - min; |
c = max - min; |
| 4597 |
if (c > (unsigned int)(md->end_subject - eptr)) |
if (c > (unsigned int)(md->end_subject - eptr)) |
| 4598 |
c = md->end_subject - eptr; |
{ |
| 4599 |
eptr += c; |
eptr = md->end_subject; |
| 4600 |
|
SCHECK_PARTIAL(); |
| 4601 |
|
} |
| 4602 |
|
else eptr += c; |
| 4603 |
break; |
break; |
| 4604 |
|
|
| 4605 |
case OP_ANYNL: |
case OP_ANYNL: |
| 4606 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4607 |
{ |
{ |
| 4608 |
int len = 1; |
int len = 1; |
| 4609 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4610 |
|
{ |
| 4611 |
|
SCHECK_PARTIAL(); |
| 4612 |
|
break; |
| 4613 |
|
} |
| 4614 |
GETCHARLEN(c, eptr, len); |
GETCHARLEN(c, eptr, len); |
| 4615 |
if (c == 0x000d) |
if (c == 0x000d) |
| 4616 |
{ |
{ |
| 4635 |
{ |
{ |
| 4636 |
BOOL gotspace; |
BOOL gotspace; |
| 4637 |
int len = 1; |
int len = 1; |
| 4638 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4639 |
|
{ |
| 4640 |
|
SCHECK_PARTIAL(); |
| 4641 |
|
break; |
| 4642 |
|
} |
| 4643 |
GETCHARLEN(c, eptr, len); |
GETCHARLEN(c, eptr, len); |
| 4644 |
switch(c) |
switch(c) |
| 4645 |
{ |
{ |
| 4677 |
{ |
{ |
| 4678 |
BOOL gotspace; |
BOOL gotspace; |
| 4679 |
int len = 1; |
int len = 1; |
| 4680 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4681 |
|
{ |
| 4682 |
|
SCHECK_PARTIAL(); |
| 4683 |
|
break; |
| 4684 |
|
} |
| 4685 |
GETCHARLEN(c, eptr, len); |
GETCHARLEN(c, eptr, len); |
| 4686 |
switch(c) |
switch(c) |
| 4687 |
{ |
{ |
| 4705 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4706 |
{ |
{ |
| 4707 |
int len = 1; |
int len = 1; |
| 4708 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4709 |
|
{ |
| 4710 |
|
SCHECK_PARTIAL(); |
| 4711 |
|
break; |
| 4712 |
|
} |
| 4713 |
GETCHARLEN(c, eptr, len); |
GETCHARLEN(c, eptr, len); |
| 4714 |
if (c < 256 && (md->ctypes[c] & ctype_digit) != 0) break; |
if (c < 256 && (md->ctypes[c] & ctype_digit) != 0) break; |
| 4715 |
eptr+= len; |
eptr+= len; |
| 4720 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4721 |
{ |
{ |
| 4722 |
int len = 1; |
int len = 1; |
| 4723 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4724 |
|
{ |
| 4725 |
|
SCHECK_PARTIAL(); |
| 4726 |
|
break; |
| 4727 |
|
} |
| 4728 |
GETCHARLEN(c, eptr, len); |
GETCHARLEN(c, eptr, len); |
| 4729 |
if (c >= 256 ||(md->ctypes[c] & ctype_digit) == 0) break; |
if (c >= 256 ||(md->ctypes[c] & ctype_digit) == 0) break; |
| 4730 |
eptr+= len; |
eptr+= len; |
| 4735 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4736 |
{ |
{ |
| 4737 |
int len = 1; |
int len = 1; |
| 4738 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4739 |
|
{ |
| 4740 |
|
SCHECK_PARTIAL(); |
| 4741 |
|
break; |
| 4742 |
|
} |
| 4743 |
GETCHARLEN(c, eptr, len); |
GETCHARLEN(c, eptr, len); |
| 4744 |
if (c < 256 && (md->ctypes[c] & ctype_space) != 0) break; |
if (c < 256 && (md->ctypes[c] & ctype_space) != 0) break; |
| 4745 |
eptr+= len; |
eptr+= len; |
| 4750 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4751 |
{ |
{ |
| 4752 |
int len = 1; |
int len = 1; |
| 4753 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4754 |
|
{ |
| 4755 |
|
SCHECK_PARTIAL(); |
| 4756 |
|
break; |
| 4757 |
|
} |
| 4758 |
GETCHARLEN(c, eptr, len); |
GETCHARLEN(c, eptr, len); |
| 4759 |
if (c >= 256 ||(md->ctypes[c] & ctype_space) == 0) break; |
if (c >= 256 ||(md->ctypes[c] & ctype_space) == 0) break; |
| 4760 |
eptr+= len; |
eptr+= len; |
| 4765 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4766 |
{ |
{ |
| 4767 |
int len = 1; |
int len = 1; |
| 4768 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4769 |
|
{ |
| 4770 |
|
SCHECK_PARTIAL(); |
| 4771 |
|
break; |
| 4772 |
|
} |
| 4773 |
GETCHARLEN(c, eptr, len); |
GETCHARLEN(c, eptr, len); |
| 4774 |
if (c < 256 && (md->ctypes[c] & ctype_word) != 0) break; |
if (c < 256 && (md->ctypes[c] & ctype_word) != 0) break; |
| 4775 |
eptr+= len; |
eptr+= len; |
| 4780 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4781 |
{ |
{ |
| 4782 |
int len = 1; |
int len = 1; |
| 4783 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4784 |
|
{ |
| 4785 |
|
SCHECK_PARTIAL(); |
| 4786 |
|
break; |
| 4787 |
|
} |
| 4788 |
GETCHARLEN(c, eptr, len); |
GETCHARLEN(c, eptr, len); |
| 4789 |
if (c >= 256 || (md->ctypes[c] & ctype_word) == 0) break; |
if (c >= 256 || (md->ctypes[c] & ctype_word) == 0) break; |
| 4790 |
eptr+= len; |
eptr+= len; |
| 4816 |
case OP_ANY: |
case OP_ANY: |
| 4817 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4818 |
{ |
{ |
| 4819 |
if (eptr >= md->end_subject || IS_NEWLINE(eptr)) break; |
if (eptr >= md->end_subject) |
| 4820 |
|
{ |
| 4821 |
|
SCHECK_PARTIAL(); |
| 4822 |
|
break; |
| 4823 |
|
} |
| 4824 |
|
if (IS_NEWLINE(eptr)) break; |
| 4825 |
eptr++; |
eptr++; |
| 4826 |
} |
} |
| 4827 |
break; |
break; |
| 4830 |
case OP_ANYBYTE: |
case OP_ANYBYTE: |
| 4831 |
c = max - min; |
c = max - min; |
| 4832 |
if (c > (unsigned int)(md->end_subject - eptr)) |
if (c > (unsigned int)(md->end_subject - eptr)) |
| 4833 |
c = md->end_subject - eptr; |
{ |
| 4834 |
eptr += c; |
eptr = md->end_subject; |
| 4835 |
|
SCHECK_PARTIAL(); |
| 4836 |
|
} |
| 4837 |
|
else eptr += c; |
| 4838 |
break; |
break; |
| 4839 |
|
|
| 4840 |
case OP_ANYNL: |
case OP_ANYNL: |
| 4841 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4842 |
{ |
{ |
| 4843 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4844 |
|
{ |
| 4845 |
|
SCHECK_PARTIAL(); |
| 4846 |
|
break; |
| 4847 |
|
} |
| 4848 |
c = *eptr; |
c = *eptr; |
| 4849 |
if (c == 0x000d) |
if (c == 0x000d) |
| 4850 |
{ |
{ |
| 4865 |
case OP_NOT_HSPACE: |
case OP_NOT_HSPACE: |
| 4866 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4867 |
{ |
{ |
| 4868 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4869 |
|
{ |
| 4870 |
|
SCHECK_PARTIAL(); |
| 4871 |
|
break; |
| 4872 |
|
} |
| 4873 |
c = *eptr; |
c = *eptr; |
| 4874 |
if (c == 0x09 || c == 0x20 || c == 0xa0) break; |
if (c == 0x09 || c == 0x20 || c == 0xa0) break; |
| 4875 |
eptr++; |
eptr++; |
| 4879 |
case OP_HSPACE: |
case OP_HSPACE: |
| 4880 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4881 |
{ |
{ |
| 4882 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4883 |
|
{ |
| 4884 |
|
SCHECK_PARTIAL(); |
| 4885 |
|
break; |
| 4886 |
|
} |
| 4887 |
c = *eptr; |
c = *eptr; |
| 4888 |
if (c != 0x09 && c != 0x20 && c != 0xa0) break; |
if (c != 0x09 && c != 0x20 && c != 0xa0) break; |
| 4889 |
eptr++; |
eptr++; |
| 4893 |
case OP_NOT_VSPACE: |
case OP_NOT_VSPACE: |
| 4894 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4895 |
{ |
{ |
| 4896 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4897 |
|
{ |
| 4898 |
|
SCHECK_PARTIAL(); |
| 4899 |
|
break; |
| 4900 |
|
} |
| 4901 |
c = *eptr; |
c = *eptr; |
| 4902 |
if (c == 0x0a || c == 0x0b || c == 0x0c || c == 0x0d || c == 0x85) |
if (c == 0x0a || c == 0x0b || c == 0x0c || c == 0x0d || c == 0x85) |
| 4903 |
break; |
break; |
| 4908 |
case OP_VSPACE: |
case OP_VSPACE: |
| 4909 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4910 |
{ |
{ |
| 4911 |
if (eptr >= md->end_subject) break; |
if (eptr >= md->end_subject) |
| 4912 |
|
{ |
| 4913 |
|
SCHECK_PARTIAL(); |
| 4914 |
|
break; |
| 4915 |
|
} |
| 4916 |
c = *eptr; |
c = *eptr; |
| 4917 |
if (c != 0x0a && c != 0x0b && c != 0x0c && c != 0x0d && c != 0x85) |
if (c != 0x0a && c != 0x0b && c != 0x0c && c != 0x0d && c != 0x85) |
| 4918 |
break; |
break; |
| 4923 |
case OP_NOT_DIGIT: |
case OP_NOT_DIGIT: |
| 4924 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4925 |
{ |
{ |
| 4926 |
if (eptr >= md->end_subject || (md->ctypes[*eptr] & ctype_digit) != 0) |
if (eptr >= md->end_subject) |
| 4927 |
|
{ |
| 4928 |
|
SCHECK_PARTIAL(); |
| 4929 |
break; |
break; |
| 4930 |
|
} |
| 4931 |
|
if ((md->ctypes[*eptr] & ctype_digit) != 0) break; |
| 4932 |
eptr++; |
eptr++; |
| 4933 |
} |
} |
| 4934 |
break; |
break; |
| 4936 |
case OP_DIGIT: |
case OP_DIGIT: |
| 4937 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4938 |
{ |
{ |
| 4939 |
if (eptr >= md->end_subject || (md->ctypes[*eptr] & ctype_digit) == 0) |
if (eptr >= md->end_subject) |
| 4940 |
|
{ |
| 4941 |
|
SCHECK_PARTIAL(); |
| 4942 |
break; |
break; |
| 4943 |
|
} |
| 4944 |
|
if ((md->ctypes[*eptr] & ctype_digit) == 0) break; |
| 4945 |
eptr++; |
eptr++; |
| 4946 |
} |
} |
| 4947 |
break; |
break; |
| 4949 |
case OP_NOT_WHITESPACE: |
case OP_NOT_WHITESPACE: |
| 4950 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4951 |
{ |
{ |
| 4952 |
if (eptr >= md->end_subject || (md->ctypes[*eptr] & ctype_space) != 0) |
if (eptr >= md->end_subject) |
| 4953 |
|
{ |
| 4954 |
|
SCHECK_PARTIAL(); |
| 4955 |
break; |
break; |
| 4956 |
|
} |
| 4957 |
|
if ((md->ctypes[*eptr] & ctype_space) != 0) break; |
| 4958 |
eptr++; |
eptr++; |
| 4959 |
} |
} |
| 4960 |
break; |
break; |
| 4962 |
case OP_WHITESPACE: |
case OP_WHITESPACE: |
| 4963 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4964 |
{ |
{ |
| 4965 |
if (eptr >= md->end_subject || (md->ctypes[*eptr] & ctype_space) == 0) |
if (eptr >= md->end_subject) |
| 4966 |
|
{ |
| 4967 |
|
SCHECK_PARTIAL(); |
| 4968 |
break; |
break; |
| 4969 |
|
} |
| 4970 |
|
if ((md->ctypes[*eptr] & ctype_space) == 0) break; |
| 4971 |
eptr++; |
eptr++; |
| 4972 |
} |
} |
| 4973 |
break; |
break; |
| 4975 |
case OP_NOT_WORDCHAR: |
case OP_NOT_WORDCHAR: |
| 4976 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4977 |
{ |
{ |
| 4978 |
if (eptr >= md->end_subject || (md->ctypes[*eptr] & ctype_word) != 0) |
if (eptr >= md->end_subject) |
| 4979 |
|
{ |
| 4980 |
|
SCHECK_PARTIAL(); |
| 4981 |
break; |
break; |
| 4982 |
|
} |
| 4983 |
|
if ((md->ctypes[*eptr] & ctype_word) != 0) break; |
| 4984 |
eptr++; |
eptr++; |
| 4985 |
} |
} |
| 4986 |
break; |
break; |
| 4988 |
case OP_WORDCHAR: |
case OP_WORDCHAR: |
| 4989 |
for (i = min; i < max; i++) |
for (i = min; i < max; i++) |
| 4990 |
{ |
{ |
| 4991 |
if (eptr >= md->end_subject || (md->ctypes[*eptr] & ctype_word) == 0) |
if (eptr >= md->end_subject) |
| 4992 |
|
{ |
| 4993 |
|
SCHECK_PARTIAL(); |
| 4994 |
break; |
break; |
| 4995 |
|
} |
| 4996 |
|
if ((md->ctypes[*eptr] & ctype_word) == 0) break; |
| 4997 |
eptr++; |
eptr++; |
| 4998 |
} |
} |
| 4999 |
break; |
break; |
| 5171 |
const uschar *start_bits = NULL; |
const uschar *start_bits = NULL; |
| 5172 |
USPTR start_match = (USPTR)subject + start_offset; |
USPTR start_match = (USPTR)subject + start_offset; |
| 5173 |
USPTR end_subject; |
USPTR end_subject; |
| 5174 |
|
USPTR start_partial = NULL; |
| 5175 |
USPTR req_byte_ptr = start_match - 1; |
USPTR req_byte_ptr = start_match - 1; |
| 5176 |
|
|
| 5177 |
pcre_study_data internal_study; |
pcre_study_data internal_study; |
| 5188 |
(offsets == NULL && offsetcount > 0)) return PCRE_ERROR_NULL; |
(offsets == NULL && offsetcount > 0)) return PCRE_ERROR_NULL; |
| 5189 |
if (offsetcount < 0) return PCRE_ERROR_BADCOUNT; |
if (offsetcount < 0) return PCRE_ERROR_BADCOUNT; |
| 5190 |
|
|
| 5191 |
|
/* This information is for finding all the numbers associated with a given |
| 5192 |
|
name, for condition testing. */ |
| 5193 |
|
|
| 5194 |
|
md->name_table = (uschar *)re + re->name_table_offset; |
| 5195 |
|
md->name_count = re->name_count; |
| 5196 |
|
md->name_entry_size = re->name_entry_size; |
| 5197 |
|
|
| 5198 |
/* Fish out the optional data from the extra_data structure, first setting |
/* Fish out the optional data from the extra_data structure, first setting |
| 5199 |
the default values. */ |
the default values. */ |
| 5200 |
|
|
| 5262 |
md->notbol = (options & PCRE_NOTBOL) != 0; |
md->notbol = (options & PCRE_NOTBOL) != 0; |
| 5263 |
md->noteol = (options & PCRE_NOTEOL) != 0; |
md->noteol = (options & PCRE_NOTEOL) != 0; |
| 5264 |
md->notempty = (options & PCRE_NOTEMPTY) != 0; |
md->notempty = (options & PCRE_NOTEMPTY) != 0; |
| 5265 |
md->partial = (options & PCRE_PARTIAL) != 0; |
md->notempty_atstart = (options & PCRE_NOTEMPTY_ATSTART) != 0; |
| 5266 |
|
md->partial = ((options & PCRE_PARTIAL_HARD) != 0)? 2 : |
| 5267 |
|
((options & PCRE_PARTIAL_SOFT) != 0)? 1 : 0; |
| 5268 |
md->hitend = FALSE; |
md->hitend = FALSE; |
| 5269 |
|
|
| 5270 |
md->recursive = NULL; /* No recursion at top level */ |
md->recursive = NULL; /* No recursion at top level */ |
| 5305 |
(pcre_uint32)options) & PCRE_NEWLINE_BITS) |
(pcre_uint32)options) & PCRE_NEWLINE_BITS) |
| 5306 |
{ |
{ |
| 5307 |
case 0: newline = NEWLINE; break; /* Compile-time default */ |
case 0: newline = NEWLINE; break; /* Compile-time default */ |
| 5308 |
case PCRE_NEWLINE_CR: newline = '\r'; break; |
case PCRE_NEWLINE_CR: newline = CHAR_CR; break; |
| 5309 |
case PCRE_NEWLINE_LF: newline = '\n'; break; |
case PCRE_NEWLINE_LF: newline = CHAR_NL; break; |
| 5310 |
case PCRE_NEWLINE_CR+ |
case PCRE_NEWLINE_CR+ |
| 5311 |
PCRE_NEWLINE_LF: newline = ('\r' << 8) | '\n'; break; |
PCRE_NEWLINE_LF: newline = (CHAR_CR << 8) | CHAR_NL; break; |
| 5312 |
case PCRE_NEWLINE_ANY: newline = -1; break; |
case PCRE_NEWLINE_ANY: newline = -1; break; |
| 5313 |
case PCRE_NEWLINE_ANYCRLF: newline = -2; break; |
case PCRE_NEWLINE_ANYCRLF: newline = -2; break; |
| 5314 |
default: return PCRE_ERROR_BADNEWLINE; |
default: return PCRE_ERROR_BADNEWLINE; |
| 5338 |
} |
} |
| 5339 |
} |
} |
| 5340 |
|
|
| 5341 |
/* Partial matching is supported only for a restricted set of regexes at the |
/* Partial matching was originally supported only for a restricted set of |
| 5342 |
moment. */ |
regexes; from release 8.00 there are no restrictions, but the bits are still |
| 5343 |
|
defined (though never set). So there's no harm in leaving this code. */ |
| 5344 |
|
|
| 5345 |
if (md->partial && (re->flags & PCRE_NOPARTIAL) != 0) |
if (md->partial && (re->flags & PCRE_NOPARTIAL) != 0) |
| 5346 |
return PCRE_ERROR_BADPARTIAL; |
return PCRE_ERROR_BADPARTIAL; |
| 5351 |
#ifdef SUPPORT_UTF8 |
#ifdef SUPPORT_UTF8 |
| 5352 |
if (utf8 && (options & PCRE_NO_UTF8_CHECK) == 0) |
if (utf8 && (options & PCRE_NO_UTF8_CHECK) == 0) |
| 5353 |
{ |
{ |
| 5354 |
if (_pcre_valid_utf8((uschar *)subject, length) >= 0) |
if (_pcre_valid_utf8((USPTR)subject, length) >= 0) |
| 5355 |
return PCRE_ERROR_BADUTF8; |
return PCRE_ERROR_BADUTF8; |
| 5356 |
if (start_offset > 0 && start_offset < length) |
if (start_offset > 0 && start_offset < length) |
| 5357 |
{ |
{ |
| 5358 |
int tb = ((uschar *)subject)[start_offset]; |
int tb = ((USPTR)subject)[start_offset]; |
| 5359 |
if (tb > 127) |
if (tb > 127) |
| 5360 |
{ |
{ |
| 5361 |
tb &= 0xc0; |
tb &= 0xc0; |
| 5427 |
} |
} |
| 5428 |
else |
else |
| 5429 |
if (!startline && study != NULL && |
if (!startline && study != NULL && |
| 5430 |
(study->options & PCRE_STUDY_MAPPED) != 0) |
(study->flags & PCRE_STUDY_MAPPED) != 0) |
| 5431 |
start_bits = study->start_bits; |
start_bits = study->start_bits; |
| 5432 |
} |
} |
| 5433 |
|
|
| 5461 |
while (iptr < iend) *iptr++ = -1; |
while (iptr < iend) *iptr++ = -1; |
| 5462 |
} |
} |
| 5463 |
|
|
| 5464 |
/* Advance to a unique first char if possible. If firstline is TRUE, the |
/* If firstline is TRUE, the start of the match is constrained to the first |
| 5465 |
start of the match is constrained to the first line of a multiline string. |
line of a multiline string. That is, the match must be before or at the first |
| 5466 |
That is, the match must be before or at the first newline. Implement this by |
newline. Implement this by temporarily adjusting end_subject so that we stop |
| 5467 |
temporarily adjusting end_subject so that we stop scanning at a newline. If |
scanning at a newline. If the match fails at the newline, later code breaks |
| 5468 |
the match fails at the newline, later code breaks this loop. */ |
this loop. */ |
| 5469 |
|
|
| 5470 |
if (firstline) |
if (firstline) |
| 5471 |
{ |
{ |
| 5472 |
USPTR t = start_match; |
USPTR t = start_match; |
| 5473 |
#ifdef SUPPORT_UTF8 |
#ifdef SUPPORT_UTF8 |
| 5474 |
if (utf8) |
if (utf8) |
| 5475 |
{ |
{ |
| 5476 |
while (t < md->end_subject && !IS_NEWLINE(t)) |
while (t < md->end_subject && !IS_NEWLINE(t)) |
| 5477 |
{ |
{ |
| 5478 |
t++; |
t++; |
| 5479 |
while (t < end_subject && (*t & 0xc0) == 0x80) t++; |
while (t < end_subject && (*t & 0xc0) == 0x80) t++; |
| 5480 |
} |
} |
| 5481 |
} |
} |
| 5482 |
else |
else |
| 5483 |
#endif |
#endif |
| 5484 |
while (t < md->end_subject && !IS_NEWLINE(t)) t++; |
while (t < md->end_subject && !IS_NEWLINE(t)) t++; |
| 5485 |
end_subject = t; |
end_subject = t; |
| 5486 |
} |
} |
| 5487 |
|
|
| 5488 |
/* Now advance to a unique first byte if there is one. */ |
/* There are some optimizations that avoid running the match if a known |
| 5489 |
|
starting point is not found, or if a known later character is not present. |
| 5490 |
|
However, there is an option that disables these, for testing and for ensuring |
| 5491 |
|
that all callouts do actually occur. */ |
| 5492 |
|
|
| 5493 |
if (first_byte >= 0) |
if ((options & PCRE_NO_START_OPTIMIZE) == 0) |
| 5494 |
{ |
{ |
| 5495 |
if (first_byte_caseless) |
/* Advance to a unique first byte if there is one. */ |
|
while (start_match < end_subject && md->lcc[*start_match] != first_byte) |
|
|
start_match++; |
|
|
else |
|
|
while (start_match < end_subject && *start_match != first_byte) |
|
|
start_match++; |
|
|
} |
|
| 5496 |
|
|
| 5497 |
/* Or to just after a linebreak for a multiline match */ |
if (first_byte >= 0) |
| 5498 |
|
{ |
| 5499 |
|
if (first_byte_caseless) |
| 5500 |
|
while (start_match < end_subject && md->lcc[*start_match] != first_byte) |
| 5501 |
|
start_match++; |
| 5502 |
|
else |
| 5503 |
|
while (start_match < end_subject && *start_match != first_byte) |
| 5504 |
|
start_match++; |
| 5505 |
|
} |
| 5506 |
|
|
| 5507 |
else if (startline) |
/* Or to just after a linebreak for a multiline match */ |
| 5508 |
{ |
|
| 5509 |
if (start_match > md->start_subject + start_offset) |
else if (startline) |
| 5510 |
{ |
{ |
| 5511 |
#ifdef SUPPORT_UTF8 |
if (start_match > md->start_subject + start_offset) |
|
if (utf8) |
|
| 5512 |
{ |
{ |
| 5513 |
while (start_match < end_subject && !WAS_NEWLINE(start_match)) |
#ifdef SUPPORT_UTF8 |
| 5514 |
|
if (utf8) |
| 5515 |
{ |
{ |
| 5516 |
start_match++; |
while (start_match < end_subject && !WAS_NEWLINE(start_match)) |
| 5517 |
while(start_match < end_subject && (*start_match & 0xc0) == 0x80) |
{ |
| 5518 |
start_match++; |
start_match++; |
| 5519 |
} |
while(start_match < end_subject && (*start_match & 0xc0) == 0x80) |
| 5520 |
|
start_match++; |
| 5521 |
|
} |
| 5522 |
|
} |
| 5523 |
|
else |
| 5524 |
|
#endif |
| 5525 |
|
while (start_match < end_subject && !WAS_NEWLINE(start_match)) |
| 5526 |
|
start_match++; |
| 5527 |
|
|
| 5528 |
|
/* If we have just passed a CR and the newline option is ANY or ANYCRLF, |
| 5529 |
|
and we are now at a LF, advance the match position by one more character. |
| 5530 |
|
*/ |
| 5531 |
|
|
| 5532 |
|
if (start_match[-1] == CHAR_CR && |
| 5533 |
|
(md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) && |
| 5534 |
|
start_match < end_subject && |
| 5535 |
|
*start_match == CHAR_NL) |
| 5536 |
|
start_match++; |
| 5537 |
} |
} |
|
else |
|
|
#endif |
|
|
while (start_match < end_subject && !WAS_NEWLINE(start_match)) |
|
|
start_match++; |
|
|
|
|
|
/* If we have just passed a CR and the newline option is ANY or ANYCRLF, |
|
|
and we are now at a LF, advance the match position by one more character. |
|
|
*/ |
|
|
|
|
|
if (start_match[-1] == '\r' && |
|
|
(md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) && |
|
|
start_match < end_subject && |
|
|
*start_match == '\n') |
|
|
start_match++; |
|
| 5538 |
} |
} |
|
} |
|
| 5539 |
|
|
| 5540 |
/* Or to a non-unique first byte after study */ |
/* Or to a non-unique first byte after study */ |
| 5541 |
|
|
| 5542 |
else if (start_bits != NULL) |
else if (start_bits != NULL) |
|
{ |
|
|
while (start_match < end_subject) |
|
| 5543 |
{ |
{ |
| 5544 |
register unsigned int c = *start_match; |
while (start_match < end_subject) |
| 5545 |
if ((start_bits[c/8] & (1 << (c&7))) == 0) start_match++; |
{ |
| 5546 |
else break; |
register unsigned int c = *start_match; |
| 5547 |
|
if ((start_bits[c/8] & (1 << (c&7))) == 0) start_match++; |
| 5548 |
|
else break; |
| 5549 |
|
} |
| 5550 |
} |
} |
| 5551 |
} |
} /* Starting optimizations */ |
| 5552 |
|
|
| 5553 |
/* Restore fudged end_subject */ |
/* Restore fudged end_subject */ |
| 5554 |
|
|
| 5555 |
end_subject = save_end_subject; |
end_subject = save_end_subject; |
| 5556 |
|
|
| 5557 |
#ifdef DEBUG /* Sigh. Some compilers never learn. */ |
/* The following two optimizations are disabled for partial matching or if |
| 5558 |
printf(">>>> Match against: "); |
disabling is explicitly requested. */ |
|
pchars(start_match, end_subject - start_match, TRUE, md); |
|
|
printf("\n"); |
|
|
#endif |
|
| 5559 |
|
|
| 5560 |
/* If req_byte is set, we know that that character must appear in the subject |
if ((options & PCRE_NO_START_OPTIMIZE) == 0 && !md->partial) |
|
for the match to succeed. If the first character is set, req_byte must be |
|
|
later in the subject; otherwise the test starts at the match point. This |
|
|
optimization can save a huge amount of backtracking in patterns with nested |
|
|
unlimited repeats that aren't going to match. Writing separate code for |
|
|
cased/caseless versions makes it go faster, as does using an autoincrement |
|
|
and backing off on a match. |
|
|
|
|
|
HOWEVER: when the subject string is very, very long, searching to its end can |
|
|
take a long time, and give bad performance on quite ordinary patterns. This |
|
|
showed up when somebody was matching something like /^\d+C/ on a 32-megabyte |
|
|
string... so we don't do this when the string is sufficiently long. |
|
|
|
|
|
ALSO: this processing is disabled when partial matching is requested. |
|
|
*/ |
|
|
|
|
|
if (req_byte >= 0 && |
|
|
end_subject - start_match < REQ_BYTE_MAX && |
|
|
!md->partial) |
|
| 5561 |
{ |
{ |
| 5562 |
register USPTR p = start_match + ((first_byte >= 0)? 1 : 0); |
/* If the pattern was studied, a minimum subject length may be set. This is |
| 5563 |
|
a lower bound; no actual string of that length may actually match the |
| 5564 |
|
pattern. Although the value is, strictly, in characters, we treat it as |
| 5565 |
|
bytes to avoid spending too much time in this optimization. */ |
| 5566 |
|
|
| 5567 |
/* We don't need to repeat the search if we haven't yet reached the |
if (study != NULL && (study->flags & PCRE_STUDY_MINLEN) != 0 && |
| 5568 |
place we found it at last time. */ |
end_subject - start_match < study->minlength) |
| 5569 |
|
{ |
| 5570 |
|
rc = MATCH_NOMATCH; |
| 5571 |
|
break; |
| 5572 |
|
} |
| 5573 |
|
|
| 5574 |
|
/* If req_byte is set, we know that that character must appear in the |
| 5575 |
|
subject for the match to succeed. If the first character is set, req_byte |
| 5576 |
|
must be later in the subject; otherwise the test starts at the match point. |
| 5577 |
|
This optimization can save a huge amount of backtracking in patterns with |
| 5578 |
|
nested unlimited repeats that aren't going to match. Writing separate code |
| 5579 |
|
for cased/caseless versions makes it go faster, as does using an |
| 5580 |
|
autoincrement and backing off on a match. |
| 5581 |
|
|
| 5582 |
if (p > req_byte_ptr) |
HOWEVER: when the subject string is very, very long, searching to its end |
| 5583 |
|
can take a long time, and give bad performance on quite ordinary patterns. |
| 5584 |
|
This showed up when somebody was matching something like /^\d+C/ on a |
| 5585 |
|
32-megabyte string... so we don't do this when the string is sufficiently |
| 5586 |
|
long. */ |
| 5587 |
|
|
| 5588 |
|
if (req_byte >= 0 && end_subject - start_match < REQ_BYTE_MAX) |
| 5589 |
{ |
{ |
| 5590 |
if (req_byte_caseless) |
register USPTR p = start_match + ((first_byte >= 0)? 1 : 0); |
| 5591 |
|
|
| 5592 |
|
/* We don't need to repeat the search if we haven't yet reached the |
| 5593 |
|
place we found it at last time. */ |
| 5594 |
|
|
| 5595 |
|
if (p > req_byte_ptr) |
| 5596 |
{ |
{ |
| 5597 |
while (p < end_subject) |
if (req_byte_caseless) |
| 5598 |
{ |
{ |
| 5599 |
register int pp = *p++; |
while (p < end_subject) |
| 5600 |
if (pp == req_byte || pp == req_byte2) { p--; break; } |
{ |
| 5601 |
|
register int pp = *p++; |
| 5602 |
|
if (pp == req_byte || pp == req_byte2) { p--; break; } |
| 5603 |
|
} |
| 5604 |
} |
} |
| 5605 |
} |
else |
|
else |
|
|
{ |
|
|
while (p < end_subject) |
|
| 5606 |
{ |
{ |
| 5607 |
if (*p++ == req_byte) { p--; break; } |
while (p < end_subject) |
| 5608 |
|
{ |
| 5609 |
|
if (*p++ == req_byte) { p--; break; } |
| 5610 |
|
} |
| 5611 |
} |
} |
|
} |
|
| 5612 |
|
|
| 5613 |
/* If we can't find the required character, break the matching loop, |
/* If we can't find the required character, break the matching loop, |
| 5614 |
forcing a match failure. */ |
forcing a match failure. */ |
| 5615 |
|
|
| 5616 |
if (p >= end_subject) |
if (p >= end_subject) |
| 5617 |
{ |
{ |
| 5618 |
rc = MATCH_NOMATCH; |
rc = MATCH_NOMATCH; |
| 5619 |
break; |
break; |
| 5620 |
} |
} |
| 5621 |
|
|
| 5622 |
/* If we have found the required character, save the point where we |
/* If we have found the required character, save the point where we |
| 5623 |
found it, so that we don't search again next time round the loop if |
found it, so that we don't search again next time round the loop if |
| 5624 |
the start hasn't passed this character yet. */ |
the start hasn't passed this character yet. */ |
| 5625 |
|
|
| 5626 |
req_byte_ptr = p; |
req_byte_ptr = p; |
| 5627 |
|
} |
| 5628 |
} |
} |
| 5629 |
} |
} |
| 5630 |
|
|
| 5631 |
/* OK, we can now run the match. */ |
#ifdef DEBUG /* Sigh. Some compilers never learn. */ |
| 5632 |
|
printf(">>>> Match against: "); |
| 5633 |
|
pchars(start_match, end_subject - start_match, TRUE, md); |
| 5634 |
|
printf("\n"); |
| 5635 |
|
#endif |
| 5636 |
|
|
| 5637 |
|
/* OK, we can now run the match. If "hitend" is set afterwards, remember the |
| 5638 |
|
first starting point for which a partial match was found. */ |
| 5639 |
|
|
| 5640 |
md->start_match_ptr = start_match; |
md->start_match_ptr = start_match; |
| 5641 |
|
md->start_used_ptr = start_match; |
| 5642 |
md->match_call_count = 0; |
md->match_call_count = 0; |
| 5643 |
rc = match(start_match, md->start_code, start_match, 2, md, ims, NULL, 0, 0); |
rc = match(start_match, md->start_code, start_match, 2, md, ims, NULL, 0, 0); |
| 5644 |
|
if (md->hitend && start_partial == NULL) start_partial = md->start_used_ptr; |
| 5645 |
|
|
| 5646 |
switch(rc) |
switch(rc) |
| 5647 |
{ |
{ |
| 5671 |
rc = MATCH_NOMATCH; |
rc = MATCH_NOMATCH; |
| 5672 |
goto ENDLOOP; |
goto ENDLOOP; |
| 5673 |
|
|
| 5674 |
/* Any other return is some kind of error. */ |
/* Any other return is either a match, or some kind of error. */ |
| 5675 |
|
|
| 5676 |
default: |
default: |
| 5677 |
goto ENDLOOP; |
goto ENDLOOP; |
| 5701 |
not contain any explicit matches for \r or \n, and the newline option is CRLF |
not contain any explicit matches for \r or \n, and the newline option is CRLF |
| 5702 |
or ANY or ANYCRLF, advance the match position by one more character. */ |
or ANY or ANYCRLF, advance the match position by one more character. */ |
| 5703 |
|
|
| 5704 |
if (start_match[-1] == '\r' && |
if (start_match[-1] == CHAR_CR && |
| 5705 |
start_match < end_subject && |
start_match < end_subject && |
| 5706 |
*start_match == '\n' && |
*start_match == CHAR_NL && |
| 5707 |
(re->flags & PCRE_HASCRORLF) == 0 && |
(re->flags & PCRE_HASCRORLF) == 0 && |
| 5708 |
(md->nltype == NLTYPE_ANY || |
(md->nltype == NLTYPE_ANY || |
| 5709 |
md->nltype == NLTYPE_ANYCRLF || |
md->nltype == NLTYPE_ANYCRLF || |
| 5777 |
(pcre_free)(md->offset_vector); |
(pcre_free)(md->offset_vector); |
| 5778 |
} |
} |
| 5779 |
|
|
| 5780 |
if (rc != MATCH_NOMATCH) |
if (rc != MATCH_NOMATCH && rc != PCRE_ERROR_PARTIAL) |
| 5781 |
{ |
{ |
| 5782 |
DPRINTF((">>>> error: returning %d\n", rc)); |
DPRINTF((">>>> error: returning %d\n", rc)); |
| 5783 |
return rc; |
return rc; |
| 5784 |
} |
} |
| 5785 |
else if (md->partial && md->hitend) |
else if (start_partial != NULL) |
| 5786 |
{ |
{ |
| 5787 |
DPRINTF((">>>> returning PCRE_ERROR_PARTIAL\n")); |
DPRINTF((">>>> returning PCRE_ERROR_PARTIAL\n")); |
| 5788 |
|
if (offsetcount > 1) |
| 5789 |
|
{ |
| 5790 |
|
offsets[0] = start_partial - (USPTR)subject; |
| 5791 |
|
offsets[1] = end_subject - (USPTR)subject; |
| 5792 |
|
} |
| 5793 |
return PCRE_ERROR_PARTIAL; |
return PCRE_ERROR_PARTIAL; |
| 5794 |
} |
} |
| 5795 |
else |
else |