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 */ |
67 |
|
|
68 |
/* These are offsets that are used to turn the OP_TYPESTAR and friends opcodes |
/* These are offsets that are used to turn the OP_TYPESTAR and friends opcodes |
69 |
into others, under special conditions. A gap of 20 between the blocks should be |
into others, under special conditions. A gap of 20 between the blocks should be |
70 |
enough. */ |
enough. The resulting opcodes don't have to be less than 256 because they are |
71 |
|
never stored, so we push them well clear of the normal opcodes. */ |
72 |
|
|
73 |
#define OP_PROP_EXTRA 100 |
#define OP_PROP_EXTRA 300 |
74 |
#define OP_EXTUNI_EXTRA 120 |
#define OP_EXTUNI_EXTRA 320 |
75 |
#define OP_ANYNL_EXTRA 140 |
#define OP_ANYNL_EXTRA 340 |
76 |
|
#define OP_HSPACE_EXTRA 360 |
77 |
|
#define OP_VSPACE_EXTRA 380 |
78 |
|
|
79 |
|
|
80 |
/* This table identifies those opcodes that are followed immediately by a |
/* This table identifies those opcodes that are followed immediately by a |
81 |
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 is possible to |
82 |
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 |
83 |
"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 |
84 |
small value. */ |
small value. ***NOTE*** If the start of this table is modified, the two tables |
85 |
|
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, 0, 0, 0, 0, 0, /* \A, \G, \B, \b, \D, \d, \S, \s, \W, \w */ |
0, 0, 0, 0, 0, /* \A, \G, \K, \B, \b */ |
90 |
0, 0, /* Any, Anybyte */ |
0, 0, 0, 0, 0, 0, /* \D, \d, \S, \s, \W, \w */ |
91 |
0, 0, 0, 0, /* NOTPROP, PROP, EXTUNI, ANYNL */ |
0, 0, 0, /* Any, AllAny, Anybyte */ |
92 |
|
0, 0, 0, /* NOTPROP, PROP, EXTUNI */ |
93 |
|
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, ^, $ */ |
95 |
1, /* Char */ |
1, /* Char */ |
96 |
1, /* Charnc */ |
1, /* Charnc */ |
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, |
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, |
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); |
566 |
permitted. |
permitted. |
567 |
|
|
568 |
We also use this mechanism for opcodes such as OP_TYPEPLUS that take an |
We also use this mechanism for opcodes such as OP_TYPEPLUS that take an |
569 |
argument that is not a data character - but is always one byte long. |
argument that is not a data character - but is always one byte long. We |
570 |
Unfortunately, we have to take special action to deal with \P, \p, and |
have to take special action to deal with \P, \p, \H, \h, \V, \v and \X in |
571 |
\X in this case. To keep the other cases fast, convert these ones to new |
this case. To keep the other cases fast, convert these ones to new opcodes. |
572 |
opcodes. */ |
*/ |
573 |
|
|
574 |
if (coptable[codevalue] > 0) |
if (coptable[codevalue] > 0) |
575 |
{ |
{ |
587 |
case OP_PROP: codevalue += OP_PROP_EXTRA; break; |
case OP_PROP: codevalue += OP_PROP_EXTRA; break; |
588 |
case OP_ANYNL: codevalue += OP_ANYNL_EXTRA; break; |
case OP_ANYNL: codevalue += OP_ANYNL_EXTRA; break; |
589 |
case OP_EXTUNI: codevalue += OP_EXTUNI_EXTRA; break; |
case OP_EXTUNI: codevalue += OP_EXTUNI_EXTRA; break; |
590 |
|
case OP_NOT_HSPACE: |
591 |
|
case OP_HSPACE: codevalue += OP_HSPACE_EXTRA; break; |
592 |
|
case OP_NOT_VSPACE: |
593 |
|
case OP_VSPACE: codevalue += OP_VSPACE_EXTRA; break; |
594 |
default: break; |
default: break; |
595 |
} |
} |
596 |
} |
} |
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 |
{ |
{ |
1078 |
|
case 0x000b: |
1079 |
|
case 0x000c: |
1080 |
|
case 0x0085: |
1081 |
|
case 0x2028: |
1082 |
|
case 0x2029: |
1083 |
|
if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; |
1084 |
|
goto ANYNL01; |
1085 |
|
|
1086 |
case 0x000d: |
case 0x000d: |
1087 |
if (ptr + 1 < end_subject && ptr[1] == 0x0a) ncount = 1; |
if (ptr + 1 < end_subject && ptr[1] == 0x0a) ncount = 1; |
1088 |
/* Fall through */ |
/* Fall through */ |
1089 |
|
|
1090 |
|
ANYNL01: |
1091 |
|
case 0x000a: |
1092 |
|
if (count > 0 && codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSPLUS) |
1093 |
|
{ |
1094 |
|
active_count--; /* Remove non-match possibility */ |
1095 |
|
next_active_state--; |
1096 |
|
} |
1097 |
|
count++; |
1098 |
|
ADD_NEW_DATA(-state_offset, count, ncount); |
1099 |
|
break; |
1100 |
|
|
1101 |
|
default: |
1102 |
|
break; |
1103 |
|
} |
1104 |
|
} |
1105 |
|
break; |
1106 |
|
|
1107 |
|
/*-----------------------------------------------------------------*/ |
1108 |
|
case OP_VSPACE_EXTRA + OP_TYPEPLUS: |
1109 |
|
case OP_VSPACE_EXTRA + OP_TYPEMINPLUS: |
1110 |
|
case OP_VSPACE_EXTRA + OP_TYPEPOSPLUS: |
1111 |
|
count = current_state->count; /* Already matched */ |
1112 |
|
if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); } |
1113 |
|
if (clen > 0) |
1114 |
|
{ |
1115 |
|
BOOL OK; |
1116 |
|
switch (c) |
1117 |
|
{ |
1118 |
case 0x000a: |
case 0x000a: |
1119 |
case 0x000b: |
case 0x000b: |
1120 |
case 0x000c: |
case 0x000c: |
1121 |
|
case 0x000d: |
1122 |
case 0x0085: |
case 0x0085: |
1123 |
case 0x2028: |
case 0x2028: |
1124 |
case 0x2029: |
case 0x2029: |
1125 |
if (count > 0 && codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSPLUS) |
OK = TRUE; |
1126 |
|
break; |
1127 |
|
|
1128 |
|
default: |
1129 |
|
OK = FALSE; |
1130 |
|
break; |
1131 |
|
} |
1132 |
|
|
1133 |
|
if (OK == (d == OP_VSPACE)) |
1134 |
|
{ |
1135 |
|
if (count > 0 && codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSPLUS) |
1136 |
{ |
{ |
1137 |
active_count--; /* Remove non-match possibility */ |
active_count--; /* Remove non-match possibility */ |
1138 |
next_active_state--; |
next_active_state--; |
1139 |
} |
} |
1140 |
count++; |
count++; |
1141 |
ADD_NEW_DATA(-state_offset, count, ncount); |
ADD_NEW_DATA(-state_offset, count, 0); |
1142 |
|
} |
1143 |
|
} |
1144 |
|
break; |
1145 |
|
|
1146 |
|
/*-----------------------------------------------------------------*/ |
1147 |
|
case OP_HSPACE_EXTRA + OP_TYPEPLUS: |
1148 |
|
case OP_HSPACE_EXTRA + OP_TYPEMINPLUS: |
1149 |
|
case OP_HSPACE_EXTRA + OP_TYPEPOSPLUS: |
1150 |
|
count = current_state->count; /* Already matched */ |
1151 |
|
if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); } |
1152 |
|
if (clen > 0) |
1153 |
|
{ |
1154 |
|
BOOL OK; |
1155 |
|
switch (c) |
1156 |
|
{ |
1157 |
|
case 0x09: /* HT */ |
1158 |
|
case 0x20: /* SPACE */ |
1159 |
|
case 0xa0: /* NBSP */ |
1160 |
|
case 0x1680: /* OGHAM SPACE MARK */ |
1161 |
|
case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */ |
1162 |
|
case 0x2000: /* EN QUAD */ |
1163 |
|
case 0x2001: /* EM QUAD */ |
1164 |
|
case 0x2002: /* EN SPACE */ |
1165 |
|
case 0x2003: /* EM SPACE */ |
1166 |
|
case 0x2004: /* THREE-PER-EM SPACE */ |
1167 |
|
case 0x2005: /* FOUR-PER-EM SPACE */ |
1168 |
|
case 0x2006: /* SIX-PER-EM SPACE */ |
1169 |
|
case 0x2007: /* FIGURE SPACE */ |
1170 |
|
case 0x2008: /* PUNCTUATION SPACE */ |
1171 |
|
case 0x2009: /* THIN SPACE */ |
1172 |
|
case 0x200A: /* HAIR SPACE */ |
1173 |
|
case 0x202f: /* NARROW NO-BREAK SPACE */ |
1174 |
|
case 0x205f: /* MEDIUM MATHEMATICAL SPACE */ |
1175 |
|
case 0x3000: /* IDEOGRAPHIC SPACE */ |
1176 |
|
OK = TRUE; |
1177 |
break; |
break; |
1178 |
|
|
1179 |
default: |
default: |
1180 |
|
OK = FALSE; |
1181 |
break; |
break; |
1182 |
} |
} |
1183 |
|
|
1184 |
|
if (OK == (d == OP_HSPACE)) |
1185 |
|
{ |
1186 |
|
if (count > 0 && codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSPLUS) |
1187 |
|
{ |
1188 |
|
active_count--; /* Remove non-match possibility */ |
1189 |
|
next_active_state--; |
1190 |
|
} |
1191 |
|
count++; |
1192 |
|
ADD_NEW_DATA(-state_offset, count, 0); |
1193 |
|
} |
1194 |
} |
} |
1195 |
break; |
break; |
1196 |
|
|
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 |
{ |
{ |
1317 |
|
case 0x000b: |
1318 |
|
case 0x000c: |
1319 |
|
case 0x0085: |
1320 |
|
case 0x2028: |
1321 |
|
case 0x2029: |
1322 |
|
if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; |
1323 |
|
goto ANYNL02; |
1324 |
|
|
1325 |
case 0x000d: |
case 0x000d: |
1326 |
if (ptr + 1 < end_subject && ptr[1] == 0x0a) ncount = 1; |
if (ptr + 1 < end_subject && ptr[1] == 0x0a) ncount = 1; |
1327 |
/* Fall through */ |
/* Fall through */ |
1328 |
|
|
1329 |
|
ANYNL02: |
1330 |
|
case 0x000a: |
1331 |
|
if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSSTAR || |
1332 |
|
codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSQUERY) |
1333 |
|
{ |
1334 |
|
active_count--; /* Remove non-match possibility */ |
1335 |
|
next_active_state--; |
1336 |
|
} |
1337 |
|
ADD_NEW_DATA(-(state_offset + count), 0, ncount); |
1338 |
|
break; |
1339 |
|
|
1340 |
|
default: |
1341 |
|
break; |
1342 |
|
} |
1343 |
|
} |
1344 |
|
break; |
1345 |
|
|
1346 |
|
/*-----------------------------------------------------------------*/ |
1347 |
|
case OP_VSPACE_EXTRA + OP_TYPEQUERY: |
1348 |
|
case OP_VSPACE_EXTRA + OP_TYPEMINQUERY: |
1349 |
|
case OP_VSPACE_EXTRA + OP_TYPEPOSQUERY: |
1350 |
|
count = 2; |
1351 |
|
goto QS4; |
1352 |
|
|
1353 |
|
case OP_VSPACE_EXTRA + OP_TYPESTAR: |
1354 |
|
case OP_VSPACE_EXTRA + OP_TYPEMINSTAR: |
1355 |
|
case OP_VSPACE_EXTRA + OP_TYPEPOSSTAR: |
1356 |
|
count = 0; |
1357 |
|
|
1358 |
|
QS4: |
1359 |
|
ADD_ACTIVE(state_offset + 2, 0); |
1360 |
|
if (clen > 0) |
1361 |
|
{ |
1362 |
|
BOOL OK; |
1363 |
|
switch (c) |
1364 |
|
{ |
1365 |
case 0x000a: |
case 0x000a: |
1366 |
case 0x000b: |
case 0x000b: |
1367 |
case 0x000c: |
case 0x000c: |
1368 |
|
case 0x000d: |
1369 |
case 0x0085: |
case 0x0085: |
1370 |
case 0x2028: |
case 0x2028: |
1371 |
case 0x2029: |
case 0x2029: |
1372 |
if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSSTAR || |
OK = TRUE; |
1373 |
codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSQUERY) |
break; |
1374 |
|
|
1375 |
|
default: |
1376 |
|
OK = FALSE; |
1377 |
|
break; |
1378 |
|
} |
1379 |
|
if (OK == (d == OP_VSPACE)) |
1380 |
|
{ |
1381 |
|
if (codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSSTAR || |
1382 |
|
codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSQUERY) |
1383 |
{ |
{ |
1384 |
active_count--; /* Remove non-match possibility */ |
active_count--; /* Remove non-match possibility */ |
1385 |
next_active_state--; |
next_active_state--; |
1386 |
} |
} |
1387 |
ADD_NEW_DATA(-(state_offset + count), 0, ncount); |
ADD_NEW_DATA(-(state_offset + count), 0, 0); |
1388 |
|
} |
1389 |
|
} |
1390 |
|
break; |
1391 |
|
|
1392 |
|
/*-----------------------------------------------------------------*/ |
1393 |
|
case OP_HSPACE_EXTRA + OP_TYPEQUERY: |
1394 |
|
case OP_HSPACE_EXTRA + OP_TYPEMINQUERY: |
1395 |
|
case OP_HSPACE_EXTRA + OP_TYPEPOSQUERY: |
1396 |
|
count = 2; |
1397 |
|
goto QS5; |
1398 |
|
|
1399 |
|
case OP_HSPACE_EXTRA + OP_TYPESTAR: |
1400 |
|
case OP_HSPACE_EXTRA + OP_TYPEMINSTAR: |
1401 |
|
case OP_HSPACE_EXTRA + OP_TYPEPOSSTAR: |
1402 |
|
count = 0; |
1403 |
|
|
1404 |
|
QS5: |
1405 |
|
ADD_ACTIVE(state_offset + 2, 0); |
1406 |
|
if (clen > 0) |
1407 |
|
{ |
1408 |
|
BOOL OK; |
1409 |
|
switch (c) |
1410 |
|
{ |
1411 |
|
case 0x09: /* HT */ |
1412 |
|
case 0x20: /* SPACE */ |
1413 |
|
case 0xa0: /* NBSP */ |
1414 |
|
case 0x1680: /* OGHAM SPACE MARK */ |
1415 |
|
case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */ |
1416 |
|
case 0x2000: /* EN QUAD */ |
1417 |
|
case 0x2001: /* EM QUAD */ |
1418 |
|
case 0x2002: /* EN SPACE */ |
1419 |
|
case 0x2003: /* EM SPACE */ |
1420 |
|
case 0x2004: /* THREE-PER-EM SPACE */ |
1421 |
|
case 0x2005: /* FOUR-PER-EM SPACE */ |
1422 |
|
case 0x2006: /* SIX-PER-EM SPACE */ |
1423 |
|
case 0x2007: /* FIGURE SPACE */ |
1424 |
|
case 0x2008: /* PUNCTUATION SPACE */ |
1425 |
|
case 0x2009: /* THIN SPACE */ |
1426 |
|
case 0x200A: /* HAIR SPACE */ |
1427 |
|
case 0x202f: /* NARROW NO-BREAK SPACE */ |
1428 |
|
case 0x205f: /* MEDIUM MATHEMATICAL SPACE */ |
1429 |
|
case 0x3000: /* IDEOGRAPHIC SPACE */ |
1430 |
|
OK = TRUE; |
1431 |
break; |
break; |
1432 |
|
|
1433 |
default: |
default: |
1434 |
|
OK = FALSE; |
1435 |
break; |
break; |
1436 |
} |
} |
1437 |
|
|
1438 |
|
if (OK == (d == OP_HSPACE)) |
1439 |
|
{ |
1440 |
|
if (codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSSTAR || |
1441 |
|
codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSQUERY) |
1442 |
|
{ |
1443 |
|
active_count--; /* Remove non-match possibility */ |
1444 |
|
next_active_state--; |
1445 |
|
} |
1446 |
|
ADD_NEW_DATA(-(state_offset + count), 0, 0); |
1447 |
|
} |
1448 |
} |
} |
1449 |
break; |
break; |
1450 |
|
|
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 |
{ |
{ |
1555 |
|
case 0x000b: |
1556 |
|
case 0x000c: |
1557 |
|
case 0x0085: |
1558 |
|
case 0x2028: |
1559 |
|
case 0x2029: |
1560 |
|
if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; |
1561 |
|
goto ANYNL03; |
1562 |
|
|
1563 |
case 0x000d: |
case 0x000d: |
1564 |
if (ptr + 1 < end_subject && ptr[1] == 0x0a) ncount = 1; |
if (ptr + 1 < end_subject && ptr[1] == 0x0a) ncount = 1; |
1565 |
/* Fall through */ |
/* Fall through */ |
1566 |
|
|
1567 |
|
ANYNL03: |
1568 |
|
case 0x000a: |
1569 |
|
if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSUPTO) |
1570 |
|
{ |
1571 |
|
active_count--; /* Remove non-match possibility */ |
1572 |
|
next_active_state--; |
1573 |
|
} |
1574 |
|
if (++count >= GET2(code, 1)) |
1575 |
|
{ ADD_NEW_DATA(-(state_offset + 4), 0, ncount); } |
1576 |
|
else |
1577 |
|
{ ADD_NEW_DATA(-state_offset, count, ncount); } |
1578 |
|
break; |
1579 |
|
|
1580 |
|
default: |
1581 |
|
break; |
1582 |
|
} |
1583 |
|
} |
1584 |
|
break; |
1585 |
|
|
1586 |
|
/*-----------------------------------------------------------------*/ |
1587 |
|
case OP_VSPACE_EXTRA + OP_TYPEEXACT: |
1588 |
|
case OP_VSPACE_EXTRA + OP_TYPEUPTO: |
1589 |
|
case OP_VSPACE_EXTRA + OP_TYPEMINUPTO: |
1590 |
|
case OP_VSPACE_EXTRA + OP_TYPEPOSUPTO: |
1591 |
|
if (codevalue != OP_VSPACE_EXTRA + OP_TYPEEXACT) |
1592 |
|
{ ADD_ACTIVE(state_offset + 4, 0); } |
1593 |
|
count = current_state->count; /* Number already matched */ |
1594 |
|
if (clen > 0) |
1595 |
|
{ |
1596 |
|
BOOL OK; |
1597 |
|
switch (c) |
1598 |
|
{ |
1599 |
case 0x000a: |
case 0x000a: |
1600 |
case 0x000b: |
case 0x000b: |
1601 |
case 0x000c: |
case 0x000c: |
1602 |
|
case 0x000d: |
1603 |
case 0x0085: |
case 0x0085: |
1604 |
case 0x2028: |
case 0x2028: |
1605 |
case 0x2029: |
case 0x2029: |
1606 |
if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSUPTO) |
OK = TRUE; |
1607 |
|
break; |
1608 |
|
|
1609 |
|
default: |
1610 |
|
OK = FALSE; |
1611 |
|
} |
1612 |
|
|
1613 |
|
if (OK == (d == OP_VSPACE)) |
1614 |
|
{ |
1615 |
|
if (codevalue == OP_VSPACE_EXTRA + OP_TYPEPOSUPTO) |
1616 |
{ |
{ |
1617 |
active_count--; /* Remove non-match possibility */ |
active_count--; /* Remove non-match possibility */ |
1618 |
next_active_state--; |
next_active_state--; |
1619 |
} |
} |
1620 |
if (++count >= GET2(code, 1)) |
if (++count >= GET2(code, 1)) |
1621 |
{ ADD_NEW_DATA(-(state_offset + 4), 0, ncount); } |
{ ADD_NEW_DATA(-(state_offset + 4), 0, 0); } |
1622 |
else |
else |
1623 |
{ ADD_NEW_DATA(-state_offset, count, ncount); } |
{ ADD_NEW_DATA(-state_offset, count, 0); } |
1624 |
|
} |
1625 |
|
} |
1626 |
|
break; |
1627 |
|
|
1628 |
|
/*-----------------------------------------------------------------*/ |
1629 |
|
case OP_HSPACE_EXTRA + OP_TYPEEXACT: |
1630 |
|
case OP_HSPACE_EXTRA + OP_TYPEUPTO: |
1631 |
|
case OP_HSPACE_EXTRA + OP_TYPEMINUPTO: |
1632 |
|
case OP_HSPACE_EXTRA + OP_TYPEPOSUPTO: |
1633 |
|
if (codevalue != OP_HSPACE_EXTRA + OP_TYPEEXACT) |
1634 |
|
{ ADD_ACTIVE(state_offset + 4, 0); } |
1635 |
|
count = current_state->count; /* Number already matched */ |
1636 |
|
if (clen > 0) |
1637 |
|
{ |
1638 |
|
BOOL OK; |
1639 |
|
switch (c) |
1640 |
|
{ |
1641 |
|
case 0x09: /* HT */ |
1642 |
|
case 0x20: /* SPACE */ |
1643 |
|
case 0xa0: /* NBSP */ |
1644 |
|
case 0x1680: /* OGHAM SPACE MARK */ |
1645 |
|
case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */ |
1646 |
|
case 0x2000: /* EN QUAD */ |
1647 |
|
case 0x2001: /* EM QUAD */ |
1648 |
|
case 0x2002: /* EN SPACE */ |
1649 |
|
case 0x2003: /* EM SPACE */ |
1650 |
|
case 0x2004: /* THREE-PER-EM SPACE */ |
1651 |
|
case 0x2005: /* FOUR-PER-EM SPACE */ |
1652 |
|
case 0x2006: /* SIX-PER-EM SPACE */ |
1653 |
|
case 0x2007: /* FIGURE SPACE */ |
1654 |
|
case 0x2008: /* PUNCTUATION SPACE */ |
1655 |
|
case 0x2009: /* THIN SPACE */ |
1656 |
|
case 0x200A: /* HAIR SPACE */ |
1657 |
|
case 0x202f: /* NARROW NO-BREAK SPACE */ |
1658 |
|
case 0x205f: /* MEDIUM MATHEMATICAL SPACE */ |
1659 |
|
case 0x3000: /* IDEOGRAPHIC SPACE */ |
1660 |
|
OK = TRUE; |
1661 |
break; |
break; |
1662 |
|
|
1663 |
default: |
default: |
1664 |
|
OK = FALSE; |
1665 |
break; |
break; |
1666 |
} |
} |
1667 |
|
|
1668 |
|
if (OK == (d == OP_HSPACE)) |
1669 |
|
{ |
1670 |
|
if (codevalue == OP_HSPACE_EXTRA + OP_TYPEPOSUPTO) |
1671 |
|
{ |
1672 |
|
active_count--; /* Remove non-match possibility */ |
1673 |
|
next_active_state--; |
1674 |
|
} |
1675 |
|
if (++count >= GET2(code, 1)) |
1676 |
|
{ ADD_NEW_DATA(-(state_offset + 4), 0, 0); } |
1677 |
|
else |
1678 |
|
{ ADD_NEW_DATA(-state_offset, count, 0); } |
1679 |
|
} |
1680 |
} |
} |
1681 |
break; |
break; |
1682 |
|
|
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 |
{ |
{ |
1782 |
break; |
break; |
1783 |
|
|
1784 |
/*-----------------------------------------------------------------*/ |
/*-----------------------------------------------------------------*/ |
1785 |
|
case OP_NOT_VSPACE: |
1786 |
|
if (clen > 0) switch(c) |
1787 |
|
{ |
1788 |
|
case 0x000a: |
1789 |
|
case 0x000b: |
1790 |
|
case 0x000c: |
1791 |
|
case 0x000d: |
1792 |
|
case 0x0085: |
1793 |
|
case 0x2028: |
1794 |
|
case 0x2029: |
1795 |
|
break; |
1796 |
|
|
1797 |
|
default: |
1798 |
|
ADD_NEW(state_offset + 1, 0); |
1799 |
|
break; |
1800 |
|
} |
1801 |
|
break; |
1802 |
|
|
1803 |
|
/*-----------------------------------------------------------------*/ |
1804 |
|
case OP_VSPACE: |
1805 |
|
if (clen > 0) switch(c) |
1806 |
|
{ |
1807 |
|
case 0x000a: |
1808 |
|
case 0x000b: |
1809 |
|
case 0x000c: |
1810 |
|
case 0x000d: |
1811 |
|
case 0x0085: |
1812 |
|
case 0x2028: |
1813 |
|
case 0x2029: |
1814 |
|
ADD_NEW(state_offset + 1, 0); |
1815 |
|
break; |
1816 |
|
|
1817 |
|
default: break; |
1818 |
|
} |
1819 |
|
break; |
1820 |
|
|
1821 |
|
/*-----------------------------------------------------------------*/ |
1822 |
|
case OP_NOT_HSPACE: |
1823 |
|
if (clen > 0) switch(c) |
1824 |
|
{ |
1825 |
|
case 0x09: /* HT */ |
1826 |
|
case 0x20: /* SPACE */ |
1827 |
|
case 0xa0: /* NBSP */ |
1828 |
|
case 0x1680: /* OGHAM SPACE MARK */ |
1829 |
|
case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */ |
1830 |
|
case 0x2000: /* EN QUAD */ |
1831 |
|
case 0x2001: /* EM QUAD */ |
1832 |
|
case 0x2002: /* EN SPACE */ |
1833 |
|
case 0x2003: /* EM SPACE */ |
1834 |
|
case 0x2004: /* THREE-PER-EM SPACE */ |
1835 |
|
case 0x2005: /* FOUR-PER-EM SPACE */ |
1836 |
|
case 0x2006: /* SIX-PER-EM SPACE */ |
1837 |
|
case 0x2007: /* FIGURE SPACE */ |
1838 |
|
case 0x2008: /* PUNCTUATION SPACE */ |
1839 |
|
case 0x2009: /* THIN SPACE */ |
1840 |
|
case 0x200A: /* HAIR SPACE */ |
1841 |
|
case 0x202f: /* NARROW NO-BREAK SPACE */ |
1842 |
|
case 0x205f: /* MEDIUM MATHEMATICAL SPACE */ |
1843 |
|
case 0x3000: /* IDEOGRAPHIC SPACE */ |
1844 |
|
break; |
1845 |
|
|
1846 |
|
default: |
1847 |
|
ADD_NEW(state_offset + 1, 0); |
1848 |
|
break; |
1849 |
|
} |
1850 |
|
break; |
1851 |
|
|
1852 |
|
/*-----------------------------------------------------------------*/ |
1853 |
|
case OP_HSPACE: |
1854 |
|
if (clen > 0) switch(c) |
1855 |
|
{ |
1856 |
|
case 0x09: /* HT */ |
1857 |
|
case 0x20: /* SPACE */ |
1858 |
|
case 0xa0: /* NBSP */ |
1859 |
|
case 0x1680: /* OGHAM SPACE MARK */ |
1860 |
|
case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */ |
1861 |
|
case 0x2000: /* EN QUAD */ |
1862 |
|
case 0x2001: /* EM QUAD */ |
1863 |
|
case 0x2002: /* EN SPACE */ |
1864 |
|
case 0x2003: /* EM SPACE */ |
1865 |
|
case 0x2004: /* THREE-PER-EM SPACE */ |
1866 |
|
case 0x2005: /* FOUR-PER-EM SPACE */ |
1867 |
|
case 0x2006: /* SIX-PER-EM SPACE */ |
1868 |
|
case 0x2007: /* FIGURE SPACE */ |
1869 |
|
case 0x2008: /* PUNCTUATION SPACE */ |
1870 |
|
case 0x2009: /* THIN SPACE */ |
1871 |
|
case 0x200A: /* HAIR SPACE */ |
1872 |
|
case 0x202f: /* NARROW NO-BREAK SPACE */ |
1873 |
|
case 0x205f: /* MEDIUM MATHEMATICAL SPACE */ |
1874 |
|
case 0x3000: /* IDEOGRAPHIC SPACE */ |
1875 |
|
ADD_NEW(state_offset + 1, 0); |
1876 |
|
break; |
1877 |
|
} |
1878 |
|
break; |
1879 |
|
|
1880 |
|
/*-----------------------------------------------------------------*/ |
1881 |
/* Match a negated single character. This is only used for one-byte |
/* Match a negated single character. This is only used for one-byte |
1882 |
characters, that is, we know that d < 256. The character we are |
characters, that is, we know that d < 256. The character we are |
1883 |
checking (c) can be multibyte. */ |
checking (c) can be multibyte. */ |
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 */ |