| 7 |
below for why this module is different). |
below for why this module is different). |
| 8 |
|
|
| 9 |
Written by Philip Hazel |
Written by Philip Hazel |
| 10 |
Copyright (c) 1997-2009 University of Cambridge |
Copyright (c) 1997-2010 University of Cambridge |
| 11 |
|
|
| 12 |
----------------------------------------------------------------------------- |
----------------------------------------------------------------------------- |
| 13 |
Redistribution and use in source and binary forms, with or without |
Redistribution and use in source and binary forms, with or without |
| 106 |
|
|
| 107 |
|
|
| 108 |
/* This table identifies those opcodes that are followed immediately by a |
/* This table identifies those opcodes that are followed immediately by a |
| 109 |
character that is to be tested in some way. This makes is possible to |
character that is to be tested in some way. This makes it possible to |
| 110 |
centralize the loading of these characters. In the case of Type * etc, the |
centralize the loading of these characters. In the case of Type * etc, the |
| 111 |
"character" is the opcode for \D, \d, \S, \s, \W, or \w, which will always be a |
"character" is the opcode for \D, \d, \S, \s, \W, or \w, which will always be a |
| 112 |
small value. Non-zero values in the table are the offsets from the opcode where |
small value. Non-zero values in the table are the offsets from the opcode where |
| 118 |
0, 0, 0, 0, 0, /* \A, \G, \K, \B, \b */ |
0, 0, 0, 0, 0, /* \A, \G, \K, \B, \b */ |
| 119 |
0, 0, 0, 0, 0, 0, /* \D, \d, \S, \s, \W, \w */ |
0, 0, 0, 0, 0, 0, /* \D, \d, \S, \s, \W, \w */ |
| 120 |
0, 0, 0, /* Any, AllAny, Anybyte */ |
0, 0, 0, /* Any, AllAny, Anybyte */ |
| 121 |
0, 0, 0, /* NOTPROP, PROP, EXTUNI */ |
0, 0, /* \P, \p */ |
| 122 |
0, 0, 0, 0, 0, /* \R, \H, \h, \V, \v */ |
0, 0, 0, 0, 0, /* \R, \H, \h, \V, \v */ |
| 123 |
|
0, /* \X */ |
| 124 |
0, 0, 0, 0, 0, /* \Z, \z, Opt, ^, $ */ |
0, 0, 0, 0, 0, /* \Z, \z, Opt, ^, $ */ |
| 125 |
1, /* Char */ |
1, /* Char */ |
| 126 |
1, /* Charnc */ |
1, /* Charnc */ |
| 157 |
0, /* Reverse */ |
0, /* Reverse */ |
| 158 |
0, 0, 0, 0, /* ONCE, BRA, CBRA, COND */ |
0, 0, 0, 0, /* ONCE, BRA, CBRA, COND */ |
| 159 |
0, 0, 0, /* SBRA, SCBRA, SCOND */ |
0, 0, 0, /* SBRA, SCBRA, SCOND */ |
| 160 |
0, /* CREF */ |
0, 0, /* CREF, NCREF */ |
| 161 |
0, /* RREF */ |
0, 0, /* RREF, NRREF */ |
| 162 |
0, /* DEF */ |
0, /* DEF */ |
| 163 |
0, 0, /* BRAZERO, BRAMINZERO */ |
0, 0, /* BRAZERO, BRAMINZERO */ |
| 164 |
0, 0, 0, 0, /* PRUNE, SKIP, THEN, COMMIT */ |
0, 0, 0, /* MARK, PRUNE, PRUNE_ARG, */ |
| 165 |
0, 0, 0, 0 /* FAIL, ACCEPT, CLOSE, SKIPZERO */ |
0, 0, 0, 0, /* SKIP, SKIP_ARG, THEN, THEN_ARG, */ |
| 166 |
|
0, 0, 0, 0, 0 /* COMMIT, FAIL, ACCEPT, CLOSE, SKIPZERO */ |
| 167 |
}; |
}; |
| 168 |
|
|
| 169 |
/* This table identifies those opcodes that inspect a character. It is used to |
/* This table identifies those opcodes that inspect a character. It is used to |
| 176 |
0, 0, 0, 1, 1, /* \A, \G, \K, \B, \b */ |
0, 0, 0, 1, 1, /* \A, \G, \K, \B, \b */ |
| 177 |
1, 1, 1, 1, 1, 1, /* \D, \d, \S, \s, \W, \w */ |
1, 1, 1, 1, 1, 1, /* \D, \d, \S, \s, \W, \w */ |
| 178 |
1, 1, 1, /* Any, AllAny, Anybyte */ |
1, 1, 1, /* Any, AllAny, Anybyte */ |
| 179 |
1, 1, 1, /* NOTPROP, PROP, EXTUNI */ |
1, 1, /* \P, \p */ |
| 180 |
1, 1, 1, 1, 1, /* \R, \H, \h, \V, \v */ |
1, 1, 1, 1, 1, /* \R, \H, \h, \V, \v */ |
| 181 |
|
1, /* \X */ |
| 182 |
0, 0, 0, 0, 0, /* \Z, \z, Opt, ^, $ */ |
0, 0, 0, 0, 0, /* \Z, \z, Opt, ^, $ */ |
| 183 |
1, /* Char */ |
1, /* Char */ |
| 184 |
1, /* Charnc */ |
1, /* Charnc */ |
| 215 |
0, /* Reverse */ |
0, /* Reverse */ |
| 216 |
0, 0, 0, 0, /* ONCE, BRA, CBRA, COND */ |
0, 0, 0, 0, /* ONCE, BRA, CBRA, COND */ |
| 217 |
0, 0, 0, /* SBRA, SCBRA, SCOND */ |
0, 0, 0, /* SBRA, SCBRA, SCOND */ |
| 218 |
0, /* CREF */ |
0, 0, /* CREF, NCREF */ |
| 219 |
0, /* RREF */ |
0, 0, /* RREF, NRREF */ |
| 220 |
0, /* DEF */ |
0, /* DEF */ |
| 221 |
0, 0, /* BRAZERO, BRAMINZERO */ |
0, 0, /* BRAZERO, BRAMINZERO */ |
| 222 |
0, 0, 0, 0, /* PRUNE, SKIP, THEN, COMMIT */ |
0, 0, 0, /* MARK, PRUNE, PRUNE_ARG, */ |
| 223 |
0, 0, 0, 0 /* FAIL, ACCEPT, CLOSE, SKIPZERO */ |
0, 0, 0, 0, /* SKIP, SKIP_ARG, THEN, THEN_ARG, */ |
| 224 |
|
0, 0, 0, 0, 0 /* COMMIT, FAIL, ACCEPT, CLOSE, SKIPZERO */ |
| 225 |
}; |
}; |
| 226 |
|
|
| 227 |
/* These 2 tables allow for compact code for testing for \D, \d, \S, \s, \W, |
/* These 2 tables allow for compact code for testing for \D, \d, \S, \s, \W, |
| 259 |
#define INTS_PER_STATEBLOCK (sizeof(stateblock)/sizeof(int)) |
#define INTS_PER_STATEBLOCK (sizeof(stateblock)/sizeof(int)) |
| 260 |
|
|
| 261 |
|
|
| 262 |
#ifdef DEBUG |
#ifdef PCRE_DEBUG |
| 263 |
/************************************************* |
/************************************************* |
| 264 |
* Print character string * |
* Print character string * |
| 265 |
*************************************************/ |
*************************************************/ |
| 475 |
|
|
| 476 |
{ |
{ |
| 477 |
gone_back = (current_subject - max_back < start_subject)? |
gone_back = (current_subject - max_back < start_subject)? |
| 478 |
current_subject - start_subject : max_back; |
(int)(current_subject - start_subject) : max_back; |
| 479 |
current_subject -= gone_back; |
current_subject -= gone_back; |
| 480 |
} |
} |
| 481 |
|
|
| 492 |
int back = GET(end_code, 2+LINK_SIZE); |
int back = GET(end_code, 2+LINK_SIZE); |
| 493 |
if (back <= gone_back) |
if (back <= gone_back) |
| 494 |
{ |
{ |
| 495 |
int bstate = end_code - start_code + 2 + 2*LINK_SIZE; |
int bstate = (int)(end_code - start_code + 2 + 2*LINK_SIZE); |
| 496 |
ADD_NEW_DATA(-bstate, 0, gone_back - back); |
ADD_NEW_DATA(-bstate, 0, gone_back - back); |
| 497 |
} |
} |
| 498 |
end_code += GET(end_code, 1); |
end_code += GET(end_code, 1); |
| 528 |
((*this_start_code == OP_CBRA || *this_start_code == OP_SCBRA)? 2:0); |
((*this_start_code == OP_CBRA || *this_start_code == OP_SCBRA)? 2:0); |
| 529 |
do |
do |
| 530 |
{ |
{ |
| 531 |
ADD_NEW(end_code - start_code + length, 0); |
ADD_NEW((int)(end_code - start_code + length), 0); |
| 532 |
end_code += GET(end_code, 1); |
end_code += GET(end_code, 1); |
| 533 |
length = 1 + LINK_SIZE; |
length = 1 + LINK_SIZE; |
| 534 |
} |
} |
| 563 |
workspace[0] ^= 1; /* Remember for the restarting feature */ |
workspace[0] ^= 1; /* Remember for the restarting feature */ |
| 564 |
workspace[1] = active_count; |
workspace[1] = active_count; |
| 565 |
|
|
| 566 |
#ifdef DEBUG |
#ifdef PCRE_DEBUG |
| 567 |
printf("%.*sNext character: rest of subject = \"", rlevel*2-2, SP); |
printf("%.*sNext character: rest of subject = \"", rlevel*2-2, SP); |
| 568 |
pchars((uschar *)ptr, strlen((char *)ptr), stdout); |
pchars((uschar *)ptr, strlen((char *)ptr), stdout); |
| 569 |
printf("\"\n"); |
printf("\"\n"); |
| 609 |
int state_offset = current_state->offset; |
int state_offset = current_state->offset; |
| 610 |
int count, codevalue, rrc; |
int count, codevalue, rrc; |
| 611 |
|
|
| 612 |
#ifdef DEBUG |
#ifdef PCRE_DEBUG |
| 613 |
printf ("%.*sProcessing state %d c=", rlevel*2-2, SP, state_offset); |
printf ("%.*sProcessing state %d c=", rlevel*2-2, SP, state_offset); |
| 614 |
if (clen == 0) printf("EOL\n"); |
if (clen == 0) printf("EOL\n"); |
| 615 |
else if (c > 32 && c < 127) printf("'%c'\n", c); |
else if (c > 32 && c < 127) printf("'%c'\n", c); |
| 710 |
|
|
| 711 |
switch (codevalue) |
switch (codevalue) |
| 712 |
{ |
{ |
| 713 |
|
/* ========================================================================== */ |
| 714 |
|
/* These cases are never obeyed. This is a fudge that causes a compile- |
| 715 |
|
time error if the vectors coptable or poptable, which are indexed by |
| 716 |
|
opcode, are not the correct length. It seems to be the only way to do |
| 717 |
|
such a check at compile time, as the sizeof() operator does not work |
| 718 |
|
in the C preprocessor. */ |
| 719 |
|
|
| 720 |
|
case OP_TABLE_LENGTH: |
| 721 |
|
case OP_TABLE_LENGTH + |
| 722 |
|
((sizeof(coptable) == OP_TABLE_LENGTH) && |
| 723 |
|
(sizeof(poptable) == OP_TABLE_LENGTH)): |
| 724 |
|
break; |
| 725 |
|
|
| 726 |
/* ========================================================================== */ |
/* ========================================================================== */ |
| 727 |
/* Reached a closing bracket. If not at the end of the pattern, carry |
/* Reached a closing bracket. If not at the end of the pattern, carry |
| 755 |
if (count > 0) memmove(offsets + 2, offsets, count * sizeof(int)); |
if (count > 0) memmove(offsets + 2, offsets, count * sizeof(int)); |
| 756 |
if (offsetcount >= 2) |
if (offsetcount >= 2) |
| 757 |
{ |
{ |
| 758 |
offsets[0] = current_subject - start_subject; |
offsets[0] = (int)(current_subject - start_subject); |
| 759 |
offsets[1] = ptr - start_subject; |
offsets[1] = (int)(ptr - start_subject); |
| 760 |
DPRINTF(("%.*sSet matched string = \"%.*s\"\n", rlevel*2-2, SP, |
DPRINTF(("%.*sSet matched string = \"%.*s\"\n", rlevel*2-2, SP, |
| 761 |
offsets[1] - offsets[0], current_subject)); |
offsets[1] - offsets[0], current_subject)); |
| 762 |
} |
} |
| 778 |
/*-----------------------------------------------------------------*/ |
/*-----------------------------------------------------------------*/ |
| 779 |
case OP_ALT: |
case OP_ALT: |
| 780 |
do { code += GET(code, 1); } while (*code == OP_ALT); |
do { code += GET(code, 1); } while (*code == OP_ALT); |
| 781 |
ADD_ACTIVE(code - start_code, 0); |
ADD_ACTIVE((int)(code - start_code), 0); |
| 782 |
break; |
break; |
| 783 |
|
|
| 784 |
/*-----------------------------------------------------------------*/ |
/*-----------------------------------------------------------------*/ |
| 786 |
case OP_SBRA: |
case OP_SBRA: |
| 787 |
do |
do |
| 788 |
{ |
{ |
| 789 |
ADD_ACTIVE(code - start_code + 1 + LINK_SIZE, 0); |
ADD_ACTIVE((int)(code - start_code + 1 + LINK_SIZE), 0); |
| 790 |
code += GET(code, 1); |
code += GET(code, 1); |
| 791 |
} |
} |
| 792 |
while (*code == OP_ALT); |
while (*code == OP_ALT); |
| 795 |
/*-----------------------------------------------------------------*/ |
/*-----------------------------------------------------------------*/ |
| 796 |
case OP_CBRA: |
case OP_CBRA: |
| 797 |
case OP_SCBRA: |
case OP_SCBRA: |
| 798 |
ADD_ACTIVE(code - start_code + 3 + LINK_SIZE, 0); |
ADD_ACTIVE((int)(code - start_code + 3 + LINK_SIZE), 0); |
| 799 |
code += GET(code, 1); |
code += GET(code, 1); |
| 800 |
while (*code == OP_ALT) |
while (*code == OP_ALT) |
| 801 |
{ |
{ |
| 802 |
ADD_ACTIVE(code - start_code + 1 + LINK_SIZE, 0); |
ADD_ACTIVE((int)(code - start_code + 1 + LINK_SIZE), 0); |
| 803 |
code += GET(code, 1); |
code += GET(code, 1); |
| 804 |
} |
} |
| 805 |
break; |
break; |
| 810 |
ADD_ACTIVE(state_offset + 1, 0); |
ADD_ACTIVE(state_offset + 1, 0); |
| 811 |
code += 1 + GET(code, 2); |
code += 1 + GET(code, 2); |
| 812 |
while (*code == OP_ALT) code += GET(code, 1); |
while (*code == OP_ALT) code += GET(code, 1); |
| 813 |
ADD_ACTIVE(code - start_code + 1 + LINK_SIZE, 0); |
ADD_ACTIVE((int)(code - start_code + 1 + LINK_SIZE), 0); |
| 814 |
break; |
break; |
| 815 |
|
|
| 816 |
/*-----------------------------------------------------------------*/ |
/*-----------------------------------------------------------------*/ |
| 817 |
case OP_SKIPZERO: |
case OP_SKIPZERO: |
| 818 |
code += 1 + GET(code, 2); |
code += 1 + GET(code, 2); |
| 819 |
while (*code == OP_ALT) code += GET(code, 1); |
while (*code == OP_ALT) code += GET(code, 1); |
| 820 |
ADD_ACTIVE(code - start_code + 1 + LINK_SIZE, 0); |
ADD_ACTIVE((int)(code - start_code + 1 + LINK_SIZE), 0); |
| 821 |
break; |
break; |
| 822 |
|
|
| 823 |
/*-----------------------------------------------------------------*/ |
/*-----------------------------------------------------------------*/ |
| 831 |
|
|
| 832 |
/*-----------------------------------------------------------------*/ |
/*-----------------------------------------------------------------*/ |
| 833 |
case OP_EOD: |
case OP_EOD: |
| 834 |
if (ptr >= end_subject) { ADD_ACTIVE(state_offset + 1, 0); } |
if (ptr >= end_subject) |
| 835 |
|
{ |
| 836 |
|
if ((md->moptions & PCRE_PARTIAL_HARD) != 0) |
| 837 |
|
could_continue = TRUE; |
| 838 |
|
else { ADD_ACTIVE(state_offset + 1, 0); } |
| 839 |
|
} |
| 840 |
break; |
break; |
| 841 |
|
|
| 842 |
/*-----------------------------------------------------------------*/ |
/*-----------------------------------------------------------------*/ |
| 876 |
|
|
| 877 |
/*-----------------------------------------------------------------*/ |
/*-----------------------------------------------------------------*/ |
| 878 |
case OP_EODN: |
case OP_EODN: |
| 879 |
if (clen == 0 || (IS_NEWLINE(ptr) && ptr == end_subject - md->nllen)) |
if (clen == 0 && (md->moptions & PCRE_PARTIAL_HARD) != 0) |
| 880 |
|
could_continue = TRUE; |
| 881 |
|
else if (clen == 0 || (IS_NEWLINE(ptr) && ptr == end_subject - md->nllen)) |
| 882 |
{ ADD_ACTIVE(state_offset + 1, 0); } |
{ ADD_ACTIVE(state_offset + 1, 0); } |
| 883 |
break; |
break; |
| 884 |
|
|
| 886 |
case OP_DOLL: |
case OP_DOLL: |
| 887 |
if ((md->moptions & PCRE_NOTEOL) == 0) |
if ((md->moptions & PCRE_NOTEOL) == 0) |
| 888 |
{ |
{ |
| 889 |
if (clen == 0 || |
if (clen == 0 && (md->moptions & PCRE_PARTIAL_HARD) != 0) |
| 890 |
|
could_continue = TRUE; |
| 891 |
|
else if (clen == 0 || |
| 892 |
((md->poptions & PCRE_DOLLAR_ENDONLY) == 0 && IS_NEWLINE(ptr) && |
((md->poptions & PCRE_DOLLAR_ENDONLY) == 0 && IS_NEWLINE(ptr) && |
| 893 |
((ims & PCRE_MULTILINE) != 0 || ptr == end_subject - md->nllen) |
((ims & PCRE_MULTILINE) != 0 || ptr == end_subject - md->nllen) |
| 894 |
)) |
)) |
| 931 |
if (utf8) BACKCHAR(temp); |
if (utf8) BACKCHAR(temp); |
| 932 |
#endif |
#endif |
| 933 |
GETCHARTEST(d, temp); |
GETCHARTEST(d, temp); |
| 934 |
|
#ifdef SUPPORT_UCP |
| 935 |
|
if ((md->poptions & PCRE_UCP) != 0) |
| 936 |
|
{ |
| 937 |
|
if (d == '_') left_word = TRUE; else |
| 938 |
|
{ |
| 939 |
|
int cat = UCD_CATEGORY(d); |
| 940 |
|
left_word = (cat == ucp_L || cat == ucp_N); |
| 941 |
|
} |
| 942 |
|
} |
| 943 |
|
else |
| 944 |
|
#endif |
| 945 |
left_word = d < 256 && (ctypes[d] & ctype_word) != 0; |
left_word = d < 256 && (ctypes[d] & ctype_word) != 0; |
| 946 |
} |
} |
| 947 |
else left_word = 0; |
else left_word = FALSE; |
| 948 |
|
|
| 949 |
if (clen > 0) |
if (clen > 0) |
| 950 |
|
{ |
| 951 |
|
#ifdef SUPPORT_UCP |
| 952 |
|
if ((md->poptions & PCRE_UCP) != 0) |
| 953 |
|
{ |
| 954 |
|
if (c == '_') right_word = TRUE; else |
| 955 |
|
{ |
| 956 |
|
int cat = UCD_CATEGORY(c); |
| 957 |
|
right_word = (cat == ucp_L || cat == ucp_N); |
| 958 |
|
} |
| 959 |
|
} |
| 960 |
|
else |
| 961 |
|
#endif |
| 962 |
right_word = c < 256 && (ctypes[c] & ctype_word) != 0; |
right_word = c < 256 && (ctypes[c] & ctype_word) != 0; |
| 963 |
else right_word = 0; |
} |
| 964 |
|
else right_word = FALSE; |
| 965 |
|
|
| 966 |
if ((left_word == right_word) == (codevalue == OP_NOT_WORD_BOUNDARY)) |
if ((left_word == right_word) == (codevalue == OP_NOT_WORD_BOUNDARY)) |
| 967 |
{ ADD_ACTIVE(state_offset + 1, 0); } |
{ ADD_ACTIVE(state_offset + 1, 0); } |
| 988 |
break; |
break; |
| 989 |
|
|
| 990 |
case PT_LAMP: |
case PT_LAMP: |
| 991 |
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || prop->chartype == ucp_Lt; |
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || |
| 992 |
|
prop->chartype == ucp_Lt; |
| 993 |
break; |
break; |
| 994 |
|
|
| 995 |
case PT_GC: |
case PT_GC: |
| 1004 |
OK = prop->script == code[2]; |
OK = prop->script == code[2]; |
| 1005 |
break; |
break; |
| 1006 |
|
|
| 1007 |
|
/* These are specials for combination cases. */ |
| 1008 |
|
|
| 1009 |
|
case PT_ALNUM: |
| 1010 |
|
OK = _pcre_ucp_gentype[prop->chartype] == ucp_L || |
| 1011 |
|
_pcre_ucp_gentype[prop->chartype] == ucp_N; |
| 1012 |
|
break; |
| 1013 |
|
|
| 1014 |
|
case PT_SPACE: /* Perl space */ |
| 1015 |
|
OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z || |
| 1016 |
|
c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR; |
| 1017 |
|
break; |
| 1018 |
|
|
| 1019 |
|
case PT_PXSPACE: /* POSIX space */ |
| 1020 |
|
OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z || |
| 1021 |
|
c == CHAR_HT || c == CHAR_NL || c == CHAR_VT || |
| 1022 |
|
c == CHAR_FF || c == CHAR_CR; |
| 1023 |
|
break; |
| 1024 |
|
|
| 1025 |
|
case PT_WORD: |
| 1026 |
|
OK = _pcre_ucp_gentype[prop->chartype] == ucp_L || |
| 1027 |
|
_pcre_ucp_gentype[prop->chartype] == ucp_N || |
| 1028 |
|
c == CHAR_UNDERSCORE; |
| 1029 |
|
break; |
| 1030 |
|
|
| 1031 |
/* Should never occur, but keep compilers from grumbling. */ |
/* Should never occur, but keep compilers from grumbling. */ |
| 1032 |
|
|
| 1033 |
default: |
default: |
| 1182 |
break; |
break; |
| 1183 |
|
|
| 1184 |
case PT_LAMP: |
case PT_LAMP: |
| 1185 |
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || prop->chartype == ucp_Lt; |
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || |
| 1186 |
|
prop->chartype == ucp_Lt; |
| 1187 |
break; |
break; |
| 1188 |
|
|
| 1189 |
case PT_GC: |
case PT_GC: |
| 1198 |
OK = prop->script == code[3]; |
OK = prop->script == code[3]; |
| 1199 |
break; |
break; |
| 1200 |
|
|
| 1201 |
|
/* These are specials for combination cases. */ |
| 1202 |
|
|
| 1203 |
|
case PT_ALNUM: |
| 1204 |
|
OK = _pcre_ucp_gentype[prop->chartype] == ucp_L || |
| 1205 |
|
_pcre_ucp_gentype[prop->chartype] == ucp_N; |
| 1206 |
|
break; |
| 1207 |
|
|
| 1208 |
|
case PT_SPACE: /* Perl space */ |
| 1209 |
|
OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z || |
| 1210 |
|
c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR; |
| 1211 |
|
break; |
| 1212 |
|
|
| 1213 |
|
case PT_PXSPACE: /* POSIX space */ |
| 1214 |
|
OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z || |
| 1215 |
|
c == CHAR_HT || c == CHAR_NL || c == CHAR_VT || |
| 1216 |
|
c == CHAR_FF || c == CHAR_CR; |
| 1217 |
|
break; |
| 1218 |
|
|
| 1219 |
|
case PT_WORD: |
| 1220 |
|
OK = _pcre_ucp_gentype[prop->chartype] == ucp_L || |
| 1221 |
|
_pcre_ucp_gentype[prop->chartype] == ucp_N || |
| 1222 |
|
c == CHAR_UNDERSCORE; |
| 1223 |
|
break; |
| 1224 |
|
|
| 1225 |
/* Should never occur, but keep compilers from grumbling. */ |
/* Should never occur, but keep compilers from grumbling. */ |
| 1226 |
|
|
| 1227 |
default: |
default: |
| 1429 |
break; |
break; |
| 1430 |
|
|
| 1431 |
case PT_LAMP: |
case PT_LAMP: |
| 1432 |
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || prop->chartype == ucp_Lt; |
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || |
| 1433 |
|
prop->chartype == ucp_Lt; |
| 1434 |
break; |
break; |
| 1435 |
|
|
| 1436 |
case PT_GC: |
case PT_GC: |
| 1445 |
OK = prop->script == code[3]; |
OK = prop->script == code[3]; |
| 1446 |
break; |
break; |
| 1447 |
|
|
| 1448 |
|
/* These are specials for combination cases. */ |
| 1449 |
|
|
| 1450 |
|
case PT_ALNUM: |
| 1451 |
|
OK = _pcre_ucp_gentype[prop->chartype] == ucp_L || |
| 1452 |
|
_pcre_ucp_gentype[prop->chartype] == ucp_N; |
| 1453 |
|
break; |
| 1454 |
|
|
| 1455 |
|
case PT_SPACE: /* Perl space */ |
| 1456 |
|
OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z || |
| 1457 |
|
c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR; |
| 1458 |
|
break; |
| 1459 |
|
|
| 1460 |
|
case PT_PXSPACE: /* POSIX space */ |
| 1461 |
|
OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z || |
| 1462 |
|
c == CHAR_HT || c == CHAR_NL || c == CHAR_VT || |
| 1463 |
|
c == CHAR_FF || c == CHAR_CR; |
| 1464 |
|
break; |
| 1465 |
|
|
| 1466 |
|
case PT_WORD: |
| 1467 |
|
OK = _pcre_ucp_gentype[prop->chartype] == ucp_L || |
| 1468 |
|
_pcre_ucp_gentype[prop->chartype] == ucp_N || |
| 1469 |
|
c == CHAR_UNDERSCORE; |
| 1470 |
|
break; |
| 1471 |
|
|
| 1472 |
/* Should never occur, but keep compilers from grumbling. */ |
/* Should never occur, but keep compilers from grumbling. */ |
| 1473 |
|
|
| 1474 |
default: |
default: |
| 1701 |
break; |
break; |
| 1702 |
|
|
| 1703 |
case PT_LAMP: |
case PT_LAMP: |
| 1704 |
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || prop->chartype == ucp_Lt; |
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || |
| 1705 |
|
prop->chartype == ucp_Lt; |
| 1706 |
break; |
break; |
| 1707 |
|
|
| 1708 |
case PT_GC: |
case PT_GC: |
| 1717 |
OK = prop->script == code[5]; |
OK = prop->script == code[5]; |
| 1718 |
break; |
break; |
| 1719 |
|
|
| 1720 |
|
/* These are specials for combination cases. */ |
| 1721 |
|
|
| 1722 |
|
case PT_ALNUM: |
| 1723 |
|
OK = _pcre_ucp_gentype[prop->chartype] == ucp_L || |
| 1724 |
|
_pcre_ucp_gentype[prop->chartype] == ucp_N; |
| 1725 |
|
break; |
| 1726 |
|
|
| 1727 |
|
case PT_SPACE: /* Perl space */ |
| 1728 |
|
OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z || |
| 1729 |
|
c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR; |
| 1730 |
|
break; |
| 1731 |
|
|
| 1732 |
|
case PT_PXSPACE: /* POSIX space */ |
| 1733 |
|
OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z || |
| 1734 |
|
c == CHAR_HT || c == CHAR_NL || c == CHAR_VT || |
| 1735 |
|
c == CHAR_FF || c == CHAR_CR; |
| 1736 |
|
break; |
| 1737 |
|
|
| 1738 |
|
case PT_WORD: |
| 1739 |
|
OK = _pcre_ucp_gentype[prop->chartype] == ucp_L || |
| 1740 |
|
_pcre_ucp_gentype[prop->chartype] == ucp_N || |
| 1741 |
|
c == CHAR_UNDERSCORE; |
| 1742 |
|
break; |
| 1743 |
|
|
| 1744 |
/* Should never occur, but keep compilers from grumbling. */ |
/* Should never occur, but keep compilers from grumbling. */ |
| 1745 |
|
|
| 1746 |
default: |
default: |
| 2368 |
points to the byte after the end of the class. If there is a |
points to the byte after the end of the class. If there is a |
| 2369 |
quantifier, this is where it will be. */ |
quantifier, this is where it will be. */ |
| 2370 |
|
|
| 2371 |
next_state_offset = ecode - start_code; |
next_state_offset = (int)(ecode - start_code); |
| 2372 |
|
|
| 2373 |
switch (*ecode) |
switch (*ecode) |
| 2374 |
{ |
{ |
| 2439 |
md, /* static match data */ |
md, /* static match data */ |
| 2440 |
code, /* this subexpression's code */ |
code, /* this subexpression's code */ |
| 2441 |
ptr, /* where we currently are */ |
ptr, /* where we currently are */ |
| 2442 |
ptr - start_subject, /* start offset */ |
(int)(ptr - start_subject), /* start offset */ |
| 2443 |
local_offsets, /* offset vector */ |
local_offsets, /* offset vector */ |
| 2444 |
sizeof(local_offsets)/sizeof(int), /* size of same */ |
sizeof(local_offsets)/sizeof(int), /* size of same */ |
| 2445 |
local_workspace, /* workspace vector */ |
local_workspace, /* workspace vector */ |
| 2448 |
rlevel, /* function recursion level */ |
rlevel, /* function recursion level */ |
| 2449 |
recursing); /* pass on regex recursion */ |
recursing); /* pass on regex recursion */ |
| 2450 |
|
|
| 2451 |
|
if (rc == PCRE_ERROR_DFA_UITEM) return rc; |
| 2452 |
if ((rc >= 0) == (codevalue == OP_ASSERT || codevalue == OP_ASSERTBACK)) |
if ((rc >= 0) == (codevalue == OP_ASSERT || codevalue == OP_ASSERTBACK)) |
| 2453 |
{ ADD_ACTIVE(endasscode + LINK_SIZE + 1 - start_code, 0); } |
{ ADD_ACTIVE((int)(endasscode + LINK_SIZE + 1 - start_code), 0); } |
| 2454 |
} |
} |
| 2455 |
break; |
break; |
| 2456 |
|
|
| 2477 |
cb.callout_number = code[LINK_SIZE+2]; |
cb.callout_number = code[LINK_SIZE+2]; |
| 2478 |
cb.offset_vector = offsets; |
cb.offset_vector = offsets; |
| 2479 |
cb.subject = (PCRE_SPTR)start_subject; |
cb.subject = (PCRE_SPTR)start_subject; |
| 2480 |
cb.subject_length = end_subject - start_subject; |
cb.subject_length = (int)(end_subject - start_subject); |
| 2481 |
cb.start_match = current_subject - start_subject; |
cb.start_match = (int)(current_subject - start_subject); |
| 2482 |
cb.current_position = ptr - start_subject; |
cb.current_position = (int)(ptr - start_subject); |
| 2483 |
cb.pattern_position = GET(code, LINK_SIZE + 3); |
cb.pattern_position = GET(code, LINK_SIZE + 3); |
| 2484 |
cb.next_item_length = GET(code, 3 + 2*LINK_SIZE); |
cb.next_item_length = GET(code, 3 + 2*LINK_SIZE); |
| 2485 |
cb.capture_top = 1; |
cb.capture_top = 1; |
| 2530 |
md, /* fixed match data */ |
md, /* fixed match data */ |
| 2531 |
asscode, /* this subexpression's code */ |
asscode, /* this subexpression's code */ |
| 2532 |
ptr, /* where we currently are */ |
ptr, /* where we currently are */ |
| 2533 |
ptr - start_subject, /* start offset */ |
(int)(ptr - start_subject), /* start offset */ |
| 2534 |
local_offsets, /* offset vector */ |
local_offsets, /* offset vector */ |
| 2535 |
sizeof(local_offsets)/sizeof(int), /* size of same */ |
sizeof(local_offsets)/sizeof(int), /* size of same */ |
| 2536 |
local_workspace, /* workspace vector */ |
local_workspace, /* workspace vector */ |
| 2539 |
rlevel, /* function recursion level */ |
rlevel, /* function recursion level */ |
| 2540 |
recursing); /* pass on regex recursion */ |
recursing); /* pass on regex recursion */ |
| 2541 |
|
|
| 2542 |
|
if (rc == PCRE_ERROR_DFA_UITEM) return rc; |
| 2543 |
if ((rc >= 0) == |
if ((rc >= 0) == |
| 2544 |
(condcode == OP_ASSERT || condcode == OP_ASSERTBACK)) |
(condcode == OP_ASSERT || condcode == OP_ASSERTBACK)) |
| 2545 |
{ ADD_ACTIVE(endasscode + LINK_SIZE + 1 - start_code, 0); } |
{ ADD_ACTIVE((int)(endasscode + LINK_SIZE + 1 - start_code), 0); } |
| 2546 |
else |
else |
| 2547 |
{ ADD_ACTIVE(state_offset + codelink + LINK_SIZE + 1, 0); } |
{ ADD_ACTIVE(state_offset + codelink + LINK_SIZE + 1, 0); } |
| 2548 |
} |
} |
| 2563 |
md, /* fixed match data */ |
md, /* fixed match data */ |
| 2564 |
start_code + GET(code, 1), /* this subexpression's code */ |
start_code + GET(code, 1), /* this subexpression's code */ |
| 2565 |
ptr, /* where we currently are */ |
ptr, /* where we currently are */ |
| 2566 |
ptr - start_subject, /* start offset */ |
(int)(ptr - start_subject), /* start offset */ |
| 2567 |
local_offsets, /* offset vector */ |
local_offsets, /* offset vector */ |
| 2568 |
sizeof(local_offsets)/sizeof(int), /* size of same */ |
sizeof(local_offsets)/sizeof(int), /* size of same */ |
| 2569 |
local_workspace, /* workspace vector */ |
local_workspace, /* workspace vector */ |
| 2615 |
md, /* fixed match data */ |
md, /* fixed match data */ |
| 2616 |
code, /* this subexpression's code */ |
code, /* this subexpression's code */ |
| 2617 |
ptr, /* where we currently are */ |
ptr, /* where we currently are */ |
| 2618 |
ptr - start_subject, /* start offset */ |
(int)(ptr - start_subject), /* start offset */ |
| 2619 |
local_offsets, /* offset vector */ |
local_offsets, /* offset vector */ |
| 2620 |
sizeof(local_offsets)/sizeof(int), /* size of same */ |
sizeof(local_offsets)/sizeof(int), /* size of same */ |
| 2621 |
local_workspace, /* workspace vector */ |
local_workspace, /* workspace vector */ |
| 2632 |
|
|
| 2633 |
do { end_subpattern += GET(end_subpattern, 1); } |
do { end_subpattern += GET(end_subpattern, 1); } |
| 2634 |
while (*end_subpattern == OP_ALT); |
while (*end_subpattern == OP_ALT); |
| 2635 |
next_state_offset = end_subpattern - start_code + LINK_SIZE + 1; |
next_state_offset = |
| 2636 |
|
(int)(end_subpattern - start_code + LINK_SIZE + 1); |
| 2637 |
|
|
| 2638 |
/* If the end of this subpattern is KETRMAX or KETRMIN, we must |
/* If the end of this subpattern is KETRMAX or KETRMIN, we must |
| 2639 |
arrange for the repeat state also to be added to the relevant list. |
arrange for the repeat state also to be added to the relevant list. |
| 2641 |
|
|
| 2642 |
repeat_state_offset = (*end_subpattern == OP_KETRMAX || |
repeat_state_offset = (*end_subpattern == OP_KETRMAX || |
| 2643 |
*end_subpattern == OP_KETRMIN)? |
*end_subpattern == OP_KETRMIN)? |
| 2644 |
end_subpattern - start_code - GET(end_subpattern, 1) : -1; |
(int)(end_subpattern - start_code - GET(end_subpattern, 1)) : -1; |
| 2645 |
|
|
| 2646 |
/* If we have matched an empty string, add the next state at the |
/* If we have matched an empty string, add the next state at the |
| 2647 |
current character pointer. This is important so that the duplicate |
current character pointer. This is important so that the duplicate |
| 2705 |
cb.callout_number = code[1]; |
cb.callout_number = code[1]; |
| 2706 |
cb.offset_vector = offsets; |
cb.offset_vector = offsets; |
| 2707 |
cb.subject = (PCRE_SPTR)start_subject; |
cb.subject = (PCRE_SPTR)start_subject; |
| 2708 |
cb.subject_length = end_subject - start_subject; |
cb.subject_length = (int)(end_subject - start_subject); |
| 2709 |
cb.start_match = current_subject - start_subject; |
cb.start_match = (int)(current_subject - start_subject); |
| 2710 |
cb.current_position = ptr - start_subject; |
cb.current_position = (int)(ptr - start_subject); |
| 2711 |
cb.pattern_position = GET(code, 2); |
cb.pattern_position = GET(code, 2); |
| 2712 |
cb.next_item_length = GET(code, 2 + LINK_SIZE); |
cb.next_item_length = GET(code, 2 + LINK_SIZE); |
| 2713 |
cb.capture_top = 1; |
cb.capture_top = 1; |
| 2753 |
((md->moptions & PCRE_PARTIAL_SOFT) != 0 && /* Soft partial and */ |
((md->moptions & PCRE_PARTIAL_SOFT) != 0 && /* Soft partial and */ |
| 2754 |
match_count < 0) /* no matches */ |
match_count < 0) /* no matches */ |
| 2755 |
) && /* And... */ |
) && /* And... */ |
| 2756 |
ptr >= end_subject && /* Reached end of subject */ |
ptr >= end_subject && /* Reached end of subject */ |
| 2757 |
ptr > current_subject) /* Matched non-empty string */ |
ptr > md->start_used_ptr) /* Inspected non-empty string */ |
| 2758 |
{ |
{ |
| 2759 |
if (offsetcount >= 2) |
if (offsetcount >= 2) |
| 2760 |
{ |
{ |
| 2761 |
offsets[0] = md->start_used_ptr - start_subject; |
offsets[0] = (int)(md->start_used_ptr - start_subject); |
| 2762 |
offsets[1] = end_subject - start_subject; |
offsets[1] = (int)(end_subject - start_subject); |
| 2763 |
} |
} |
| 2764 |
match_count = PCRE_ERROR_PARTIAL; |
match_count = PCRE_ERROR_PARTIAL; |
| 2765 |
} |
} |
| 2844 |
(offsets == NULL && offsetcount > 0)) return PCRE_ERROR_NULL; |
(offsets == NULL && offsetcount > 0)) return PCRE_ERROR_NULL; |
| 2845 |
if (offsetcount < 0) return PCRE_ERROR_BADCOUNT; |
if (offsetcount < 0) return PCRE_ERROR_BADCOUNT; |
| 2846 |
if (wscount < 20) return PCRE_ERROR_DFA_WSSIZE; |
if (wscount < 20) return PCRE_ERROR_DFA_WSSIZE; |
| 2847 |
|
if (start_offset < 0 || start_offset > length) return PCRE_ERROR_BADOFFSET; |
| 2848 |
|
|
| 2849 |
/* We need to find the pointer to any study data before we test for byte |
/* We need to find the pointer to any study data before we test for byte |
| 2850 |
flipping, so we scan the extra_data block first. This may set two fields in the |
flipping, so we scan the extra_data block first. This may set two fields in the |
| 2963 |
#ifdef SUPPORT_UTF8 |
#ifdef SUPPORT_UTF8 |
| 2964 |
if (utf8 && (options & PCRE_NO_UTF8_CHECK) == 0) |
if (utf8 && (options & PCRE_NO_UTF8_CHECK) == 0) |
| 2965 |
{ |
{ |
| 2966 |
if (_pcre_valid_utf8((uschar *)subject, length) >= 0) |
int tb; |
| 2967 |
return PCRE_ERROR_BADUTF8; |
if ((tb = _pcre_valid_utf8((uschar *)subject, length)) >= 0) |
| 2968 |
|
return (tb == length && (options & PCRE_PARTIAL_HARD) != 0)? |
| 2969 |
|
PCRE_ERROR_SHORTUTF8 : PCRE_ERROR_BADUTF8; |
| 2970 |
if (start_offset > 0 && start_offset < length) |
if (start_offset > 0 && start_offset < length) |
| 2971 |
{ |
{ |
| 2972 |
int tb = ((uschar *)subject)[start_offset]; |
tb = ((USPTR)subject)[start_offset] & 0xc0; |
| 2973 |
if (tb > 127) |
if (tb == 0x80) return PCRE_ERROR_BADUTF8_OFFSET; |
|
{ |
|
|
tb &= 0xc0; |
|
|
if (tb != 0 && tb != 0xc0) return PCRE_ERROR_BADUTF8_OFFSET; |
|
|
} |
|
| 2974 |
} |
} |
| 2975 |
} |
} |
| 2976 |
#endif |
#endif |
| 3117 |
while (current_subject < end_subject) |
while (current_subject < end_subject) |
| 3118 |
{ |
{ |
| 3119 |
register unsigned int c = *current_subject; |
register unsigned int c = *current_subject; |
| 3120 |
if ((start_bits[c/8] & (1 << (c&7))) == 0) current_subject++; |
if ((start_bits[c/8] & (1 << (c&7))) == 0) |
| 3121 |
else break; |
{ |
| 3122 |
|
current_subject++; |
| 3123 |
|
#ifdef SUPPORT_UTF8 |
| 3124 |
|
if (utf8) |
| 3125 |
|
while(current_subject < end_subject && |
| 3126 |
|
(*current_subject & 0xc0) == 0x80) current_subject++; |
| 3127 |
|
#endif |
| 3128 |
|
} |
| 3129 |
|
else break; |
| 3130 |
} |
} |
| 3131 |
} |
} |
| 3132 |
} |
} |
| 3148 |
bytes to avoid spending too much time in this optimization. */ |
bytes to avoid spending too much time in this optimization. */ |
| 3149 |
|
|
| 3150 |
if (study != NULL && (study->flags & PCRE_STUDY_MINLEN) != 0 && |
if (study != NULL && (study->flags & PCRE_STUDY_MINLEN) != 0 && |
| 3151 |
end_subject - current_subject < study->minlength) |
(pcre_uint32)(end_subject - current_subject) < study->minlength) |
| 3152 |
return PCRE_ERROR_NOMATCH; |
return PCRE_ERROR_NOMATCH; |
| 3153 |
|
|
| 3154 |
/* If req_byte is set, we know that that character must appear in the |
/* If req_byte is set, we know that that character must appear in the |