| 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) 1997-2007 University of Cambridge |
Copyright (c) 1997-2008 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 |
| 44 |
applications. */ |
applications. */ |
| 45 |
|
|
| 46 |
|
|
| 47 |
|
#ifdef HAVE_CONFIG_H |
| 48 |
|
#include "config.h" |
| 49 |
|
#endif |
| 50 |
|
|
| 51 |
#define NLBLOCK md /* Block containing newline information */ |
#define NLBLOCK md /* Block containing newline information */ |
| 52 |
#define PSSTART start_subject /* Field containing processed string start */ |
#define PSSTART start_subject /* Field containing processed string start */ |
| 53 |
#define PSEND end_subject /* Field containing processed string end */ |
#define PSEND end_subject /* Field containing processed string end */ |
| 84 |
small value. ***NOTE*** If the start of this table is modified, the two tables |
small value. ***NOTE*** If the start of this table is modified, the two tables |
| 85 |
that follow must also be modified. */ |
that follow must also be modified. */ |
| 86 |
|
|
| 87 |
static uschar coptable[] = { |
static const uschar coptable[] = { |
| 88 |
0, /* End */ |
0, /* End */ |
| 89 |
0, 0, 0, 0, 0, /* \A, \G, \K, \B, \b */ |
0, 0, 0, 0, 0, /* \A, \G, \K, \B, \b */ |
| 90 |
0, 0, 0, 0, 0, 0, /* \D, \d, \S, \s, \W, \w */ |
0, 0, 0, 0, 0, 0, /* \D, \d, \S, \s, \W, \w */ |
| 91 |
0, 0, /* Any, Anybyte */ |
0, 0, 0, /* Any, AllAny, Anybyte */ |
| 92 |
0, 0, 0, /* NOTPROP, PROP, EXTUNI */ |
0, 0, 0, /* NOTPROP, PROP, EXTUNI */ |
| 93 |
0, 0, 0, 0, 0, /* \R, \H, \h, \V, \v */ |
0, 0, 0, 0, 0, /* \R, \H, \h, \V, \v */ |
| 94 |
0, 0, 0, 0, 0, /* \Z, \z, Opt, ^, $ */ |
0, 0, 0, 0, 0, /* \Z, \z, Opt, ^, $ */ |
| 130 |
0, /* CREF */ |
0, /* CREF */ |
| 131 |
0, /* RREF */ |
0, /* RREF */ |
| 132 |
0, /* DEF */ |
0, /* DEF */ |
| 133 |
0, 0 /* BRAZERO, BRAMINZERO */ |
0, 0, /* BRAZERO, BRAMINZERO */ |
| 134 |
|
0, 0, 0, 0, /* PRUNE, SKIP, THEN, COMMIT */ |
| 135 |
|
0, 0, 0 /* FAIL, ACCEPT, SKIPZERO */ |
| 136 |
}; |
}; |
| 137 |
|
|
| 138 |
/* 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, |
| 139 |
and \w */ |
and \w */ |
| 140 |
|
|
| 141 |
static uschar toptable1[] = { |
static const uschar toptable1[] = { |
| 142 |
0, 0, 0, 0, 0, 0, |
0, 0, 0, 0, 0, 0, |
| 143 |
ctype_digit, ctype_digit, |
ctype_digit, ctype_digit, |
| 144 |
ctype_space, ctype_space, |
ctype_space, ctype_space, |
| 145 |
ctype_word, ctype_word, |
ctype_word, ctype_word, |
| 146 |
0 /* OP_ANY */ |
0, 0 /* OP_ANY, OP_ALLANY */ |
| 147 |
}; |
}; |
| 148 |
|
|
| 149 |
static uschar toptable2[] = { |
static const uschar toptable2[] = { |
| 150 |
0, 0, 0, 0, 0, 0, |
0, 0, 0, 0, 0, 0, |
| 151 |
ctype_digit, 0, |
ctype_digit, 0, |
| 152 |
ctype_space, 0, |
ctype_space, 0, |
| 153 |
ctype_word, 0, |
ctype_word, 0, |
| 154 |
1 /* OP_ANY */ |
1, 1 /* OP_ANY, OP_ALLANY */ |
| 155 |
}; |
}; |
| 156 |
|
|
| 157 |
|
|
| 223 |
rlevel function call recursion level |
rlevel function call recursion level |
| 224 |
recursing regex recursive call level |
recursing regex recursive call level |
| 225 |
|
|
| 226 |
Returns: > 0 => |
Returns: > 0 => number of match offset pairs placed in offsets |
| 227 |
= 0 => |
= 0 => offsets overflowed; longest matches are present |
| 228 |
-1 => failed to match |
-1 => failed to match |
| 229 |
< -1 => some kind of unexpected problem |
< -1 => some kind of unexpected problem |
| 230 |
|
|
| 512 |
const uschar *code; |
const uschar *code; |
| 513 |
int state_offset = current_state->offset; |
int state_offset = current_state->offset; |
| 514 |
int count, codevalue; |
int count, codevalue; |
|
#ifdef SUPPORT_UCP |
|
|
int chartype, script; |
|
|
#endif |
|
| 515 |
|
|
| 516 |
#ifdef DEBUG |
#ifdef DEBUG |
| 517 |
printf ("%.*sProcessing state %d c=", rlevel*2-2, SP, state_offset); |
printf ("%.*sProcessing state %d c=", rlevel*2-2, SP, state_offset); |
| 691 |
break; |
break; |
| 692 |
|
|
| 693 |
/*-----------------------------------------------------------------*/ |
/*-----------------------------------------------------------------*/ |
| 694 |
|
case OP_SKIPZERO: |
| 695 |
|
code += 1 + GET(code, 2); |
| 696 |
|
while (*code == OP_ALT) code += GET(code, 1); |
| 697 |
|
ADD_ACTIVE(code - start_code + 1 + LINK_SIZE, 0); |
| 698 |
|
break; |
| 699 |
|
|
| 700 |
|
/*-----------------------------------------------------------------*/ |
| 701 |
case OP_CIRC: |
case OP_CIRC: |
| 702 |
if ((ptr == start_subject && (md->moptions & PCRE_NOTBOL) == 0) || |
if ((ptr == start_subject && (md->moptions & PCRE_NOTBOL) == 0) || |
| 703 |
((ims & PCRE_MULTILINE) != 0 && |
((ims & PCRE_MULTILINE) != 0 && |
| 736 |
|
|
| 737 |
/*-----------------------------------------------------------------*/ |
/*-----------------------------------------------------------------*/ |
| 738 |
case OP_ANY: |
case OP_ANY: |
| 739 |
if (clen > 0 && ((ims & PCRE_DOTALL) != 0 || !IS_NEWLINE(ptr))) |
if (clen > 0 && !IS_NEWLINE(ptr)) |
| 740 |
|
{ ADD_NEW(state_offset + 1, 0); } |
| 741 |
|
break; |
| 742 |
|
|
| 743 |
|
/*-----------------------------------------------------------------*/ |
| 744 |
|
case OP_ALLANY: |
| 745 |
|
if (clen > 0) |
| 746 |
{ ADD_NEW(state_offset + 1, 0); } |
{ ADD_NEW(state_offset + 1, 0); } |
| 747 |
break; |
break; |
| 748 |
|
|
| 822 |
if (clen > 0) |
if (clen > 0) |
| 823 |
{ |
{ |
| 824 |
BOOL OK; |
BOOL OK; |
| 825 |
int category = _pcre_ucp_findprop(c, &chartype, &script); |
const ucd_record * prop = GET_UCD(c); |
| 826 |
switch(code[1]) |
switch(code[1]) |
| 827 |
{ |
{ |
| 828 |
case PT_ANY: |
case PT_ANY: |
| 830 |
break; |
break; |
| 831 |
|
|
| 832 |
case PT_LAMP: |
case PT_LAMP: |
| 833 |
OK = chartype == ucp_Lu || chartype == ucp_Ll || chartype == ucp_Lt; |
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || prop->chartype == ucp_Lt; |
| 834 |
break; |
break; |
| 835 |
|
|
| 836 |
case PT_GC: |
case PT_GC: |
| 837 |
OK = category == code[2]; |
OK = ucp_gentype[prop->chartype] == code[2]; |
| 838 |
break; |
break; |
| 839 |
|
|
| 840 |
case PT_PC: |
case PT_PC: |
| 841 |
OK = chartype == code[2]; |
OK = prop->chartype == code[2]; |
| 842 |
break; |
break; |
| 843 |
|
|
| 844 |
case PT_SC: |
case PT_SC: |
| 845 |
OK = script == code[2]; |
OK = prop->script == code[2]; |
| 846 |
break; |
break; |
| 847 |
|
|
| 848 |
/* Should never occur, but keep compilers from grumbling. */ |
/* Should never occur, but keep compilers from grumbling. */ |
| 862 |
/* ========================================================================== */ |
/* ========================================================================== */ |
| 863 |
/* These opcodes likewise inspect the subject character, but have an |
/* These opcodes likewise inspect the subject character, but have an |
| 864 |
argument that is not a data character. It is one of these opcodes: |
argument that is not a data character. It is one of these opcodes: |
| 865 |
OP_ANY, OP_DIGIT, OP_NOT_DIGIT, OP_WHITESPACE, OP_NOT_SPACE, OP_WORDCHAR, |
OP_ANY, OP_ALLANY, OP_DIGIT, OP_NOT_DIGIT, OP_WHITESPACE, OP_NOT_SPACE, |
| 866 |
OP_NOT_WORDCHAR. The value is loaded into d. */ |
OP_WORDCHAR, OP_NOT_WORDCHAR. The value is loaded into d. */ |
| 867 |
|
|
| 868 |
case OP_TYPEPLUS: |
case OP_TYPEPLUS: |
| 869 |
case OP_TYPEMINPLUS: |
case OP_TYPEMINPLUS: |
| 874 |
{ |
{ |
| 875 |
if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || |
if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || |
| 876 |
(c < 256 && |
(c < 256 && |
| 877 |
(d != OP_ANY || |
(d != OP_ANY || !IS_NEWLINE(ptr)) && |
|
(ims & PCRE_DOTALL) != 0 || |
|
|
!IS_NEWLINE(ptr) |
|
|
) && |
|
| 878 |
((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) |
((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) |
| 879 |
{ |
{ |
| 880 |
if (count > 0 && codevalue == OP_TYPEPOSPLUS) |
if (count > 0 && codevalue == OP_TYPEPOSPLUS) |
| 897 |
{ |
{ |
| 898 |
if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || |
if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || |
| 899 |
(c < 256 && |
(c < 256 && |
| 900 |
(d != OP_ANY || |
(d != OP_ANY || !IS_NEWLINE(ptr)) && |
|
(ims & PCRE_DOTALL) != 0 || |
|
|
!IS_NEWLINE(ptr) |
|
|
) && |
|
| 901 |
((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) |
((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) |
| 902 |
{ |
{ |
| 903 |
if (codevalue == OP_TYPEPOSQUERY) |
if (codevalue == OP_TYPEPOSQUERY) |
| 919 |
{ |
{ |
| 920 |
if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || |
if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || |
| 921 |
(c < 256 && |
(c < 256 && |
| 922 |
(d != OP_ANY || |
(d != OP_ANY || !IS_NEWLINE(ptr)) && |
|
(ims & PCRE_DOTALL) != 0 || |
|
|
!IS_NEWLINE(ptr) |
|
|
) && |
|
| 923 |
((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) |
((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) |
| 924 |
{ |
{ |
| 925 |
if (codevalue == OP_TYPEPOSSTAR) |
if (codevalue == OP_TYPEPOSSTAR) |
| 939 |
{ |
{ |
| 940 |
if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || |
if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || |
| 941 |
(c < 256 && |
(c < 256 && |
| 942 |
(d != OP_ANY || |
(d != OP_ANY || !IS_NEWLINE(ptr)) && |
|
(ims & PCRE_DOTALL) != 0 || |
|
|
!IS_NEWLINE(ptr) |
|
|
) && |
|
| 943 |
((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) |
((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) |
| 944 |
{ |
{ |
| 945 |
if (++count >= GET2(code, 1)) |
if (++count >= GET2(code, 1)) |
| 960 |
{ |
{ |
| 961 |
if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || |
if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || |
| 962 |
(c < 256 && |
(c < 256 && |
| 963 |
(d != OP_ANY || |
(d != OP_ANY || !IS_NEWLINE(ptr)) && |
|
(ims & PCRE_DOTALL) != 0 || |
|
|
!IS_NEWLINE(ptr) |
|
|
) && |
|
| 964 |
((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) |
((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) |
| 965 |
{ |
{ |
| 966 |
if (codevalue == OP_TYPEPOSUPTO) |
if (codevalue == OP_TYPEPOSUPTO) |
| 991 |
if (clen > 0) |
if (clen > 0) |
| 992 |
{ |
{ |
| 993 |
BOOL OK; |
BOOL OK; |
| 994 |
int category = _pcre_ucp_findprop(c, &chartype, &script); |
const ucd_record * prop = GET_UCD(c); |
| 995 |
switch(code[2]) |
switch(code[2]) |
| 996 |
{ |
{ |
| 997 |
case PT_ANY: |
case PT_ANY: |
| 999 |
break; |
break; |
| 1000 |
|
|
| 1001 |
case PT_LAMP: |
case PT_LAMP: |
| 1002 |
OK = chartype == ucp_Lu || chartype == ucp_Ll || chartype == ucp_Lt; |
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || prop->chartype == ucp_Lt; |
| 1003 |
break; |
break; |
| 1004 |
|
|
| 1005 |
case PT_GC: |
case PT_GC: |
| 1006 |
OK = category == code[3]; |
OK = ucp_gentype[prop->chartype] == code[3]; |
| 1007 |
break; |
break; |
| 1008 |
|
|
| 1009 |
case PT_PC: |
case PT_PC: |
| 1010 |
OK = chartype == code[3]; |
OK = prop->chartype == code[3]; |
| 1011 |
break; |
break; |
| 1012 |
|
|
| 1013 |
case PT_SC: |
case PT_SC: |
| 1014 |
OK = script == code[3]; |
OK = prop->script == code[3]; |
| 1015 |
break; |
break; |
| 1016 |
|
|
| 1017 |
/* Should never occur, but keep compilers from grumbling. */ |
/* Should never occur, but keep compilers from grumbling. */ |
| 1040 |
case OP_EXTUNI_EXTRA + OP_TYPEPOSPLUS: |
case OP_EXTUNI_EXTRA + OP_TYPEPOSPLUS: |
| 1041 |
count = current_state->count; /* Already matched */ |
count = current_state->count; /* Already matched */ |
| 1042 |
if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); } |
if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); } |
| 1043 |
if (clen > 0 && _pcre_ucp_findprop(c, &chartype, &script) != ucp_M) |
if (clen > 0 && UCD_CATEGORY(c) != ucp_M) |
| 1044 |
{ |
{ |
| 1045 |
const uschar *nptr = ptr + clen; |
const uschar *nptr = ptr + clen; |
| 1046 |
int ncount = 0; |
int ncount = 0; |
| 1054 |
int nd; |
int nd; |
| 1055 |
int ndlen = 1; |
int ndlen = 1; |
| 1056 |
GETCHARLEN(nd, nptr, ndlen); |
GETCHARLEN(nd, nptr, ndlen); |
| 1057 |
if (_pcre_ucp_findprop(nd, &chartype, &script) != ucp_M) break; |
if (UCD_CATEGORY(nd) != ucp_M) break; |
| 1058 |
ncount++; |
ncount++; |
| 1059 |
nptr += ndlen; |
nptr += ndlen; |
| 1060 |
} |
} |
| 1075 |
int ncount = 0; |
int ncount = 0; |
| 1076 |
switch (c) |
switch (c) |
| 1077 |
{ |
{ |
|
case 0x000d: |
|
|
if (ptr + 1 < end_subject && ptr[1] == 0x0a) ncount = 1; |
|
|
/* Fall through */ |
|
|
case 0x000a: |
|
| 1078 |
case 0x000b: |
case 0x000b: |
| 1079 |
case 0x000c: |
case 0x000c: |
| 1080 |
case 0x0085: |
case 0x0085: |
| 1081 |
case 0x2028: |
case 0x2028: |
| 1082 |
case 0x2029: |
case 0x2029: |
| 1083 |
|
if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; |
| 1084 |
|
goto ANYNL01; |
| 1085 |
|
|
| 1086 |
|
case 0x000d: |
| 1087 |
|
if (ptr + 1 < end_subject && ptr[1] == 0x0a) ncount = 1; |
| 1088 |
|
/* Fall through */ |
| 1089 |
|
|
| 1090 |
|
ANYNL01: |
| 1091 |
|
case 0x000a: |
| 1092 |
if (count > 0 && codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSPLUS) |
if (count > 0 && codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSPLUS) |
| 1093 |
{ |
{ |
| 1094 |
active_count--; /* Remove non-match possibility */ |
active_count--; /* Remove non-match possibility */ |
| 1097 |
count++; |
count++; |
| 1098 |
ADD_NEW_DATA(-state_offset, count, ncount); |
ADD_NEW_DATA(-state_offset, count, ncount); |
| 1099 |
break; |
break; |
| 1100 |
|
|
| 1101 |
default: |
default: |
| 1102 |
break; |
break; |
| 1103 |
} |
} |
| 1213 |
if (clen > 0) |
if (clen > 0) |
| 1214 |
{ |
{ |
| 1215 |
BOOL OK; |
BOOL OK; |
| 1216 |
int category = _pcre_ucp_findprop(c, &chartype, &script); |
const ucd_record * prop = GET_UCD(c); |
| 1217 |
switch(code[2]) |
switch(code[2]) |
| 1218 |
{ |
{ |
| 1219 |
case PT_ANY: |
case PT_ANY: |
| 1221 |
break; |
break; |
| 1222 |
|
|
| 1223 |
case PT_LAMP: |
case PT_LAMP: |
| 1224 |
OK = chartype == ucp_Lu || chartype == ucp_Ll || chartype == ucp_Lt; |
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || prop->chartype == ucp_Lt; |
| 1225 |
break; |
break; |
| 1226 |
|
|
| 1227 |
case PT_GC: |
case PT_GC: |
| 1228 |
OK = category == code[3]; |
OK = ucp_gentype[prop->chartype] == code[3]; |
| 1229 |
break; |
break; |
| 1230 |
|
|
| 1231 |
case PT_PC: |
case PT_PC: |
| 1232 |
OK = chartype == code[3]; |
OK = prop->chartype == code[3]; |
| 1233 |
break; |
break; |
| 1234 |
|
|
| 1235 |
case PT_SC: |
case PT_SC: |
| 1236 |
OK = script == code[3]; |
OK = prop->script == code[3]; |
| 1237 |
break; |
break; |
| 1238 |
|
|
| 1239 |
/* Should never occur, but keep compilers from grumbling. */ |
/* Should never occur, but keep compilers from grumbling. */ |
| 1271 |
QS2: |
QS2: |
| 1272 |
|
|
| 1273 |
ADD_ACTIVE(state_offset + 2, 0); |
ADD_ACTIVE(state_offset + 2, 0); |
| 1274 |
if (clen > 0 && _pcre_ucp_findprop(c, &chartype, &script) != ucp_M) |
if (clen > 0 && UCD_CATEGORY(c) != ucp_M) |
| 1275 |
{ |
{ |
| 1276 |
const uschar *nptr = ptr + clen; |
const uschar *nptr = ptr + clen; |
| 1277 |
int ncount = 0; |
int ncount = 0; |
| 1286 |
int nd; |
int nd; |
| 1287 |
int ndlen = 1; |
int ndlen = 1; |
| 1288 |
GETCHARLEN(nd, nptr, ndlen); |
GETCHARLEN(nd, nptr, ndlen); |
| 1289 |
if (_pcre_ucp_findprop(nd, &chartype, &script) != ucp_M) break; |
if (UCD_CATEGORY(nd) != ucp_M) break; |
| 1290 |
ncount++; |
ncount++; |
| 1291 |
nptr += ndlen; |
nptr += ndlen; |
| 1292 |
} |
} |
| 1314 |
int ncount = 0; |
int ncount = 0; |
| 1315 |
switch (c) |
switch (c) |
| 1316 |
{ |
{ |
|
case 0x000d: |
|
|
if (ptr + 1 < end_subject && ptr[1] == 0x0a) ncount = 1; |
|
|
/* Fall through */ |
|
|
case 0x000a: |
|
| 1317 |
case 0x000b: |
case 0x000b: |
| 1318 |
case 0x000c: |
case 0x000c: |
| 1319 |
case 0x0085: |
case 0x0085: |
| 1320 |
case 0x2028: |
case 0x2028: |
| 1321 |
case 0x2029: |
case 0x2029: |
| 1322 |
|
if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; |
| 1323 |
|
goto ANYNL02; |
| 1324 |
|
|
| 1325 |
|
case 0x000d: |
| 1326 |
|
if (ptr + 1 < end_subject && ptr[1] == 0x0a) ncount = 1; |
| 1327 |
|
/* Fall through */ |
| 1328 |
|
|
| 1329 |
|
ANYNL02: |
| 1330 |
|
case 0x000a: |
| 1331 |
if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSSTAR || |
if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSSTAR || |
| 1332 |
codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSQUERY) |
codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSQUERY) |
| 1333 |
{ |
{ |
| 1336 |
} |
} |
| 1337 |
ADD_NEW_DATA(-(state_offset + count), 0, ncount); |
ADD_NEW_DATA(-(state_offset + count), 0, ncount); |
| 1338 |
break; |
break; |
| 1339 |
|
|
| 1340 |
default: |
default: |
| 1341 |
break; |
break; |
| 1342 |
} |
} |
| 1460 |
if (clen > 0) |
if (clen > 0) |
| 1461 |
{ |
{ |
| 1462 |
BOOL OK; |
BOOL OK; |
| 1463 |
int category = _pcre_ucp_findprop(c, &chartype, &script); |
const ucd_record * prop = GET_UCD(c); |
| 1464 |
switch(code[4]) |
switch(code[4]) |
| 1465 |
{ |
{ |
| 1466 |
case PT_ANY: |
case PT_ANY: |
| 1468 |
break; |
break; |
| 1469 |
|
|
| 1470 |
case PT_LAMP: |
case PT_LAMP: |
| 1471 |
OK = chartype == ucp_Lu || chartype == ucp_Ll || chartype == ucp_Lt; |
OK = prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || prop->chartype == ucp_Lt; |
| 1472 |
break; |
break; |
| 1473 |
|
|
| 1474 |
case PT_GC: |
case PT_GC: |
| 1475 |
OK = category == code[5]; |
OK = ucp_gentype[prop->chartype] == code[5]; |
| 1476 |
break; |
break; |
| 1477 |
|
|
| 1478 |
case PT_PC: |
case PT_PC: |
| 1479 |
OK = chartype == code[5]; |
OK = prop->chartype == code[5]; |
| 1480 |
break; |
break; |
| 1481 |
|
|
| 1482 |
case PT_SC: |
case PT_SC: |
| 1483 |
OK = script == code[5]; |
OK = prop->script == code[5]; |
| 1484 |
break; |
break; |
| 1485 |
|
|
| 1486 |
/* Should never occur, but keep compilers from grumbling. */ |
/* Should never occur, but keep compilers from grumbling. */ |
| 1513 |
if (codevalue != OP_EXTUNI_EXTRA + OP_TYPEEXACT) |
if (codevalue != OP_EXTUNI_EXTRA + OP_TYPEEXACT) |
| 1514 |
{ ADD_ACTIVE(state_offset + 4, 0); } |
{ ADD_ACTIVE(state_offset + 4, 0); } |
| 1515 |
count = current_state->count; /* Number already matched */ |
count = current_state->count; /* Number already matched */ |
| 1516 |
if (clen > 0 && _pcre_ucp_findprop(c, &chartype, &script) != ucp_M) |
if (clen > 0 && UCD_CATEGORY(c) != ucp_M) |
| 1517 |
{ |
{ |
| 1518 |
const uschar *nptr = ptr + clen; |
const uschar *nptr = ptr + clen; |
| 1519 |
int ncount = 0; |
int ncount = 0; |
| 1527 |
int nd; |
int nd; |
| 1528 |
int ndlen = 1; |
int ndlen = 1; |
| 1529 |
GETCHARLEN(nd, nptr, ndlen); |
GETCHARLEN(nd, nptr, ndlen); |
| 1530 |
if (_pcre_ucp_findprop(nd, &chartype, &script) != ucp_M) break; |
if (UCD_CATEGORY(nd) != ucp_M) break; |
| 1531 |
ncount++; |
ncount++; |
| 1532 |
nptr += ndlen; |
nptr += ndlen; |
| 1533 |
} |
} |
| 1552 |
int ncount = 0; |
int ncount = 0; |
| 1553 |
switch (c) |
switch (c) |
| 1554 |
{ |
{ |
|
case 0x000d: |
|
|
if (ptr + 1 < end_subject && ptr[1] == 0x0a) ncount = 1; |
|
|
/* Fall through */ |
|
|
case 0x000a: |
|
| 1555 |
case 0x000b: |
case 0x000b: |
| 1556 |
case 0x000c: |
case 0x000c: |
| 1557 |
case 0x0085: |
case 0x0085: |
| 1558 |
case 0x2028: |
case 0x2028: |
| 1559 |
case 0x2029: |
case 0x2029: |
| 1560 |
|
if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; |
| 1561 |
|
goto ANYNL03; |
| 1562 |
|
|
| 1563 |
|
case 0x000d: |
| 1564 |
|
if (ptr + 1 < end_subject && ptr[1] == 0x0a) ncount = 1; |
| 1565 |
|
/* Fall through */ |
| 1566 |
|
|
| 1567 |
|
ANYNL03: |
| 1568 |
|
case 0x000a: |
| 1569 |
if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSUPTO) |
if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSUPTO) |
| 1570 |
{ |
{ |
| 1571 |
active_count--; /* Remove non-match possibility */ |
active_count--; /* Remove non-match possibility */ |
| 1576 |
else |
else |
| 1577 |
{ ADD_NEW_DATA(-state_offset, count, ncount); } |
{ ADD_NEW_DATA(-state_offset, count, ncount); } |
| 1578 |
break; |
break; |
| 1579 |
|
|
| 1580 |
default: |
default: |
| 1581 |
break; |
break; |
| 1582 |
} |
} |
| 1707 |
other case of the character. */ |
other case of the character. */ |
| 1708 |
|
|
| 1709 |
#ifdef SUPPORT_UCP |
#ifdef SUPPORT_UCP |
| 1710 |
othercase = _pcre_ucp_othercase(c); |
othercase = UCD_OTHERCASE(c); |
| 1711 |
#else |
#else |
| 1712 |
othercase = NOTACHAR; |
othercase = NOTACHAR; |
| 1713 |
#endif |
#endif |
| 1732 |
to wait for them to pass before continuing. */ |
to wait for them to pass before continuing. */ |
| 1733 |
|
|
| 1734 |
case OP_EXTUNI: |
case OP_EXTUNI: |
| 1735 |
if (clen > 0 && _pcre_ucp_findprop(c, &chartype, &script) != ucp_M) |
if (clen > 0 && UCD_CATEGORY(c) != ucp_M) |
| 1736 |
{ |
{ |
| 1737 |
const uschar *nptr = ptr + clen; |
const uschar *nptr = ptr + clen; |
| 1738 |
int ncount = 0; |
int ncount = 0; |
| 1740 |
{ |
{ |
| 1741 |
int nclen = 1; |
int nclen = 1; |
| 1742 |
GETCHARLEN(c, nptr, nclen); |
GETCHARLEN(c, nptr, nclen); |
| 1743 |
if (_pcre_ucp_findprop(c, &chartype, &script) != ucp_M) break; |
if (UCD_CATEGORY(c) != ucp_M) break; |
| 1744 |
ncount++; |
ncount++; |
| 1745 |
nptr += nclen; |
nptr += nclen; |
| 1746 |
} |
} |
| 1757 |
case OP_ANYNL: |
case OP_ANYNL: |
| 1758 |
if (clen > 0) switch(c) |
if (clen > 0) switch(c) |
| 1759 |
{ |
{ |
|
case 0x000a: |
|
| 1760 |
case 0x000b: |
case 0x000b: |
| 1761 |
case 0x000c: |
case 0x000c: |
| 1762 |
case 0x0085: |
case 0x0085: |
| 1763 |
case 0x2028: |
case 0x2028: |
| 1764 |
case 0x2029: |
case 0x2029: |
| 1765 |
|
if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; |
| 1766 |
|
|
| 1767 |
|
case 0x000a: |
| 1768 |
ADD_NEW(state_offset + 1, 0); |
ADD_NEW(state_offset + 1, 0); |
| 1769 |
break; |
break; |
| 1770 |
|
|
| 1771 |
case 0x000d: |
case 0x000d: |
| 1772 |
if (ptr + 1 < end_subject && ptr[1] == 0x0a) |
if (ptr + 1 < end_subject && ptr[1] == 0x0a) |
| 1773 |
{ |
{ |
| 1908 |
if (utf8 && d >= 128) |
if (utf8 && d >= 128) |
| 1909 |
{ |
{ |
| 1910 |
#ifdef SUPPORT_UCP |
#ifdef SUPPORT_UCP |
| 1911 |
otherd = _pcre_ucp_othercase(d); |
otherd = UCD_OTHERCASE(d); |
| 1912 |
#endif /* SUPPORT_UCP */ |
#endif /* SUPPORT_UCP */ |
| 1913 |
} |
} |
| 1914 |
else |
else |
| 1946 |
if (utf8 && d >= 128) |
if (utf8 && d >= 128) |
| 1947 |
{ |
{ |
| 1948 |
#ifdef SUPPORT_UCP |
#ifdef SUPPORT_UCP |
| 1949 |
otherd = _pcre_ucp_othercase(d); |
otherd = UCD_OTHERCASE(d); |
| 1950 |
#endif /* SUPPORT_UCP */ |
#endif /* SUPPORT_UCP */ |
| 1951 |
} |
} |
| 1952 |
else |
else |
| 1982 |
if (utf8 && d >= 128) |
if (utf8 && d >= 128) |
| 1983 |
{ |
{ |
| 1984 |
#ifdef SUPPORT_UCP |
#ifdef SUPPORT_UCP |
| 1985 |
otherd = _pcre_ucp_othercase(d); |
otherd = UCD_OTHERCASE(d); |
| 1986 |
#endif /* SUPPORT_UCP */ |
#endif /* SUPPORT_UCP */ |
| 1987 |
} |
} |
| 1988 |
else |
else |
| 2014 |
if (utf8 && d >= 128) |
if (utf8 && d >= 128) |
| 2015 |
{ |
{ |
| 2016 |
#ifdef SUPPORT_UCP |
#ifdef SUPPORT_UCP |
| 2017 |
otherd = _pcre_ucp_othercase(d); |
otherd = UCD_OTHERCASE(d); |
| 2018 |
#endif /* SUPPORT_UCP */ |
#endif /* SUPPORT_UCP */ |
| 2019 |
} |
} |
| 2020 |
else |
else |
| 2049 |
if (utf8 && d >= 128) |
if (utf8 && d >= 128) |
| 2050 |
{ |
{ |
| 2051 |
#ifdef SUPPORT_UCP |
#ifdef SUPPORT_UCP |
| 2052 |
otherd = _pcre_ucp_othercase(d); |
otherd = UCD_OTHERCASE(d); |
| 2053 |
#endif /* SUPPORT_UCP */ |
#endif /* SUPPORT_UCP */ |
| 2054 |
} |
} |
| 2055 |
else |
else |
| 2157 |
|
|
| 2158 |
/* ========================================================================== */ |
/* ========================================================================== */ |
| 2159 |
/* These are the opcodes for fancy brackets of various kinds. We have |
/* These are the opcodes for fancy brackets of various kinds. We have |
| 2160 |
to use recursion in order to handle them. */ |
to use recursion in order to handle them. The "always failing" assersion |
| 2161 |
|
(?!) is optimised when compiling to OP_FAIL, so we have to support that, |
| 2162 |
|
though the other "backtracking verbs" are not supported. */ |
| 2163 |
|
|
| 2164 |
|
case OP_FAIL: |
| 2165 |
|
break; |
| 2166 |
|
|
| 2167 |
case OP_ASSERT: |
case OP_ASSERT: |
| 2168 |
case OP_ASSERT_NOT: |
case OP_ASSERT_NOT: |
| 2595 |
md->moptions = options; |
md->moptions = options; |
| 2596 |
md->poptions = re->options; |
md->poptions = re->options; |
| 2597 |
|
|
| 2598 |
|
/* If the BSR option is not set at match time, copy what was set |
| 2599 |
|
at compile time. */ |
| 2600 |
|
|
| 2601 |
|
if ((md->moptions & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) == 0) |
| 2602 |
|
{ |
| 2603 |
|
if ((re->options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) != 0) |
| 2604 |
|
md->moptions |= re->options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE); |
| 2605 |
|
#ifdef BSR_ANYCRLF |
| 2606 |
|
else md->moptions |= PCRE_BSR_ANYCRLF; |
| 2607 |
|
#endif |
| 2608 |
|
} |
| 2609 |
|
|
| 2610 |
/* Handle different types of newline. The three bits give eight cases. If |
/* Handle different types of newline. The three bits give eight cases. If |
| 2611 |
nothing is set at run time, whatever was used at compile time applies. */ |
nothing is set at run time, whatever was used at compile time applies. */ |
| 2612 |
|
|
| 2677 |
used in a loop when finding where to start. */ |
used in a loop when finding where to start. */ |
| 2678 |
|
|
| 2679 |
lcc = md->tables + lcc_offset; |
lcc = md->tables + lcc_offset; |
| 2680 |
startline = (re->options & PCRE_STARTLINE) != 0; |
startline = (re->flags & PCRE_STARTLINE) != 0; |
| 2681 |
firstline = (re->options & PCRE_FIRSTLINE) != 0; |
firstline = (re->options & PCRE_FIRSTLINE) != 0; |
| 2682 |
|
|
| 2683 |
/* Set up the first character to match, if available. The first_byte value is |
/* Set up the first character to match, if available. The first_byte value is |
| 2688 |
|
|
| 2689 |
if (!anchored) |
if (!anchored) |
| 2690 |
{ |
{ |
| 2691 |
if ((re->options & PCRE_FIRSTSET) != 0) |
if ((re->flags & PCRE_FIRSTSET) != 0) |
| 2692 |
{ |
{ |
| 2693 |
first_byte = re->first_byte & 255; |
first_byte = re->first_byte & 255; |
| 2694 |
if ((first_byte_caseless = ((re->first_byte & REQ_CASELESS) != 0)) == TRUE) |
if ((first_byte_caseless = ((re->first_byte & REQ_CASELESS) != 0)) == TRUE) |
| 2705 |
/* For anchored or unanchored matches, there may be a "last known required |
/* For anchored or unanchored matches, there may be a "last known required |
| 2706 |
character" set. */ |
character" set. */ |
| 2707 |
|
|
| 2708 |
if ((re->options & PCRE_REQCHSET) != 0) |
if ((re->flags & PCRE_REQCHSET) != 0) |
| 2709 |
{ |
{ |
| 2710 |
req_byte = re->req_byte & 255; |
req_byte = re->req_byte & 255; |
| 2711 |
req_byte_caseless = (re->req_byte & REQ_CASELESS) != 0; |
req_byte_caseless = (re->req_byte & REQ_CASELESS) != 0; |
| 2875 |
} |
} |
| 2876 |
if (current_subject > end_subject) break; |
if (current_subject > end_subject) break; |
| 2877 |
|
|
| 2878 |
/* If we have just passed a CR and the newline option is CRLF or ANY or |
/* If we have just passed a CR and we are now at a LF, and the pattern does |
| 2879 |
ANYCRLF, and we are now at a LF, advance the match position by one more |
not contain any explicit matches for \r or \n, and the newline option is CRLF |
| 2880 |
character. */ |
or ANY or ANYCRLF, advance the match position by one more character. */ |
| 2881 |
|
|
| 2882 |
if (current_subject[-1] == '\r' && |
if (current_subject[-1] == '\r' && |
| 2883 |
(md->nltype == NLTYPE_ANY || |
current_subject < end_subject && |
| 2884 |
md->nltype == NLTYPE_ANYCRLF || |
*current_subject == '\n' && |
| 2885 |
md->nllen == 2) && |
(re->flags & PCRE_HASCRORLF) == 0 && |
| 2886 |
current_subject < end_subject && |
(md->nltype == NLTYPE_ANY || |
| 2887 |
*current_subject == '\n') |
md->nltype == NLTYPE_ANYCRLF || |
| 2888 |
|
md->nllen == 2)) |
| 2889 |
current_subject++; |
current_subject++; |
| 2890 |
|
|
| 2891 |
} /* "Bumpalong" loop */ |
} /* "Bumpalong" loop */ |