|
|
| version 1.5, 2011/02/01 05:03:42 | version 1.8, 2015/10/13 00:25:40 |
|---|---|
| Line 7 and semantics are as close as possible t | Line 7 and semantics are as close as possible t |
| below for why this module is different). | below for why this module is different). |
| Written by Philip Hazel | Written by Philip Hazel |
| Copyright (c) 1997-2010 University of Cambridge | Copyright (c) 1997-2014 University of Cambridge |
| ----------------------------------------------------------------------------- | ----------------------------------------------------------------------------- |
| Redistribution and use in source and binary forms, with or without | Redistribution and use in source and binary forms, with or without |
| Line 38 POSSIBILITY OF SUCH DAMAGE. | Line 38 POSSIBILITY OF SUCH DAMAGE. |
| ----------------------------------------------------------------------------- | ----------------------------------------------------------------------------- |
| */ | */ |
| /* This module contains the external function pcre_dfa_exec(), which is an | /* This module contains the external function pcre_dfa_exec(), which is an |
| alternative matching function that uses a sort of DFA algorithm (not a true | alternative matching function that uses a sort of DFA algorithm (not a true |
| FSM). This is NOT Perl- compatible, but it has advantages in certain | FSM). This is NOT Perl-compatible, but it has advantages in certain |
| applications. */ | applications. */ |
| Line 113 small value. Non-zero values in the tabl | Line 112 small value. Non-zero values in the tabl |
| the character is to be found. ***NOTE*** If the start of this table is | the character is to be found. ***NOTE*** If the start of this table is |
| modified, the three tables that follow must also be modified. */ | modified, the three tables that follow must also be modified. */ |
| static const uschar coptable[] = { | static const pcre_uint8 coptable[] = { |
| 0, /* End */ | 0, /* End */ |
| 0, 0, 0, 0, 0, /* \A, \G, \K, \B, \b */ | 0, 0, 0, 0, 0, /* \A, \G, \K, \B, \b */ |
| 0, 0, 0, 0, 0, 0, /* \D, \d, \S, \s, \W, \w */ | 0, 0, 0, 0, 0, 0, /* \D, \d, \S, \s, \W, \w */ |
| Line 121 static const uschar coptable[] = { | Line 120 static const uschar coptable[] = { |
| 0, 0, /* \P, \p */ | 0, 0, /* \P, \p */ |
| 0, 0, 0, 0, 0, /* \R, \H, \h, \V, \v */ | 0, 0, 0, 0, 0, /* \R, \H, \h, \V, \v */ |
| 0, /* \X */ | 0, /* \X */ |
| 0, 0, 0, 0, 0, /* \Z, \z, Opt, ^, $ */ | 0, 0, 0, 0, 0, 0, /* \Z, \z, $, $M, ^, ^M */ |
| 1, /* Char */ | 1, /* Char */ |
| 1, /* Charnc */ | 1, /* Chari */ |
| 1, /* not */ | 1, /* not */ |
| 1, /* noti */ | |
| /* Positive single-char repeats */ | /* Positive single-char repeats */ |
| 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ | 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ |
| 3, 3, 3, /* upto, minupto, exact */ | 1+IMM2_SIZE, 1+IMM2_SIZE, /* upto, minupto */ |
| 1, 1, 1, 3, /* *+, ++, ?+, upto+ */ | 1+IMM2_SIZE, /* exact */ |
| 1, 1, 1, 1+IMM2_SIZE, /* *+, ++, ?+, upto+ */ | |
| 1, 1, 1, 1, 1, 1, /* *I, *?I, +I, +?I, ?I, ??I */ | |
| 1+IMM2_SIZE, 1+IMM2_SIZE, /* upto I, minupto I */ | |
| 1+IMM2_SIZE, /* exact I */ | |
| 1, 1, 1, 1+IMM2_SIZE, /* *+I, ++I, ?+I, upto+I */ | |
| /* Negative single-char repeats - only for chars < 256 */ | /* Negative single-char repeats - only for chars < 256 */ |
| 1, 1, 1, 1, 1, 1, /* NOT *, *?, +, +?, ?, ?? */ | 1, 1, 1, 1, 1, 1, /* NOT *, *?, +, +?, ?, ?? */ |
| 3, 3, 3, /* NOT upto, minupto, exact */ | 1+IMM2_SIZE, 1+IMM2_SIZE, /* NOT upto, minupto */ |
| 1, 1, 1, 3, /* NOT *+, ++, ?+, updo+ */ | 1+IMM2_SIZE, /* NOT exact */ |
| 1, 1, 1, 1+IMM2_SIZE, /* NOT *+, ++, ?+, upto+ */ | |
| 1, 1, 1, 1, 1, 1, /* NOT *I, *?I, +I, +?I, ?I, ??I */ | |
| 1+IMM2_SIZE, 1+IMM2_SIZE, /* NOT upto I, minupto I */ | |
| 1+IMM2_SIZE, /* NOT exact I */ | |
| 1, 1, 1, 1+IMM2_SIZE, /* NOT *+I, ++I, ?+I, upto+I */ | |
| /* Positive type repeats */ | /* Positive type repeats */ |
| 1, 1, 1, 1, 1, 1, /* Type *, *?, +, +?, ?, ?? */ | 1, 1, 1, 1, 1, 1, /* Type *, *?, +, +?, ?, ?? */ |
| 3, 3, 3, /* Type upto, minupto, exact */ | 1+IMM2_SIZE, 1+IMM2_SIZE, /* Type upto, minupto */ |
| 1, 1, 1, 3, /* Type *+, ++, ?+, upto+ */ | 1+IMM2_SIZE, /* Type exact */ |
| 1, 1, 1, 1+IMM2_SIZE, /* Type *+, ++, ?+, upto+ */ | |
| /* Character class & ref repeats */ | /* Character class & ref repeats */ |
| 0, 0, 0, 0, 0, 0, /* *, *?, +, +?, ?, ?? */ | 0, 0, 0, 0, 0, 0, /* *, *?, +, +?, ?, ?? */ |
| 0, 0, /* CRRANGE, CRMINRANGE */ | 0, 0, /* CRRANGE, CRMINRANGE */ |
| 0, 0, 0, 0, /* Possessive *+, ++, ?+, CRPOSRANGE */ | |
| 0, /* CLASS */ | 0, /* CLASS */ |
| 0, /* NCLASS */ | 0, /* NCLASS */ |
| 0, /* XCLASS - variable length */ | 0, /* XCLASS - variable length */ |
| 0, /* REF */ | 0, /* REF */ |
| 0, /* REFI */ | |
| 0, /* DNREF */ | |
| 0, /* DNREFI */ | |
| 0, /* RECURSE */ | 0, /* RECURSE */ |
| 0, /* CALLOUT */ | 0, /* CALLOUT */ |
| 0, /* Alt */ | 0, /* Alt */ |
| 0, /* Ket */ | 0, /* Ket */ |
| 0, /* KetRmax */ | 0, /* KetRmax */ |
| 0, /* KetRmin */ | 0, /* KetRmin */ |
| 0, /* KetRpos */ | |
| 0, /* Reverse */ | |
| 0, /* Assert */ | 0, /* Assert */ |
| 0, /* Assert not */ | 0, /* Assert not */ |
| 0, /* Assert behind */ | 0, /* Assert behind */ |
| 0, /* Assert behind not */ | 0, /* Assert behind not */ |
| 0, /* Reverse */ | 0, 0, /* ONCE, ONCE_NC */ |
| 0, 0, 0, 0, /* ONCE, BRA, CBRA, COND */ | 0, 0, 0, 0, 0, /* BRA, BRAPOS, CBRA, CBRAPOS, COND */ |
| 0, 0, 0, /* SBRA, SCBRA, SCOND */ | 0, 0, 0, 0, 0, /* SBRA, SBRAPOS, SCBRA, SCBRAPOS, SCOND */ |
| 0, 0, /* CREF, NCREF */ | 0, 0, /* CREF, DNCREF */ |
| 0, 0, /* RREF, NRREF */ | 0, 0, /* RREF, DNRREF */ |
| 0, /* DEF */ | 0, /* DEF */ |
| 0, 0, /* BRAZERO, BRAMINZERO */ | 0, 0, 0, /* BRAZERO, BRAMINZERO, BRAPOSZERO */ |
| 0, 0, 0, /* MARK, PRUNE, PRUNE_ARG, */ | 0, 0, 0, /* MARK, PRUNE, PRUNE_ARG */ |
| 0, 0, 0, 0, /* SKIP, SKIP_ARG, THEN, THEN_ARG, */ | 0, 0, 0, 0, /* SKIP, SKIP_ARG, THEN, THEN_ARG */ |
| 0, 0, 0, 0, 0 /* COMMIT, FAIL, ACCEPT, CLOSE, SKIPZERO */ | 0, 0, 0, 0, /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT */ |
| 0, 0 /* CLOSE, SKIPZERO */ | |
| }; | }; |
| /* 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 |
| Line 171 remember the fact that a character could | Line 189 remember the fact that a character could |
| the subject is reached. ***NOTE*** If the start of this table is modified, the | the subject is reached. ***NOTE*** If the start of this table is modified, the |
| two tables that follow must also be modified. */ | two tables that follow must also be modified. */ |
| static const uschar poptable[] = { | static const pcre_uint8 poptable[] = { |
| 0, /* End */ | 0, /* End */ |
| 0, 0, 0, 1, 1, /* \A, \G, \K, \B, \b */ | 0, 0, 0, 1, 1, /* \A, \G, \K, \B, \b */ |
| 1, 1, 1, 1, 1, 1, /* \D, \d, \S, \s, \W, \w */ | 1, 1, 1, 1, 1, 1, /* \D, \d, \S, \s, \W, \w */ |
| Line 179 static const uschar poptable[] = { | Line 197 static const uschar poptable[] = { |
| 1, 1, /* \P, \p */ | 1, 1, /* \P, \p */ |
| 1, 1, 1, 1, 1, /* \R, \H, \h, \V, \v */ | 1, 1, 1, 1, 1, /* \R, \H, \h, \V, \v */ |
| 1, /* \X */ | 1, /* \X */ |
| 0, 0, 0, 0, 0, /* \Z, \z, Opt, ^, $ */ | 0, 0, 0, 0, 0, 0, /* \Z, \z, $, $M, ^, ^M */ |
| 1, /* Char */ | 1, /* Char */ |
| 1, /* Charnc */ | 1, /* Chari */ |
| 1, /* not */ | 1, /* not */ |
| 1, /* noti */ | |
| /* Positive single-char repeats */ | /* Positive single-char repeats */ |
| 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ | 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ |
| 1, 1, 1, /* upto, minupto, exact */ | 1, 1, 1, /* upto, minupto, exact */ |
| 1, 1, 1, 1, /* *+, ++, ?+, upto+ */ | 1, 1, 1, 1, /* *+, ++, ?+, upto+ */ |
| 1, 1, 1, 1, 1, 1, /* *I, *?I, +I, +?I, ?I, ??I */ | |
| 1, 1, 1, /* upto I, minupto I, exact I */ | |
| 1, 1, 1, 1, /* *+I, ++I, ?+I, upto+I */ | |
| /* Negative single-char repeats - only for chars < 256 */ | /* Negative single-char repeats - only for chars < 256 */ |
| 1, 1, 1, 1, 1, 1, /* NOT *, *?, +, +?, ?, ?? */ | 1, 1, 1, 1, 1, 1, /* NOT *, *?, +, +?, ?, ?? */ |
| 1, 1, 1, /* NOT upto, minupto, exact */ | 1, 1, 1, /* NOT upto, minupto, exact */ |
| 1, 1, 1, 1, /* NOT *+, ++, ?+, upto+ */ | 1, 1, 1, 1, /* NOT *+, ++, ?+, upto+ */ |
| 1, 1, 1, 1, 1, 1, /* NOT *I, *?I, +I, +?I, ?I, ??I */ | |
| 1, 1, 1, /* NOT upto I, minupto I, exact I */ | |
| 1, 1, 1, 1, /* NOT *+I, ++I, ?+I, upto+I */ | |
| /* Positive type repeats */ | /* Positive type repeats */ |
| 1, 1, 1, 1, 1, 1, /* Type *, *?, +, +?, ?, ?? */ | 1, 1, 1, 1, 1, 1, /* Type *, *?, +, +?, ?, ?? */ |
| 1, 1, 1, /* Type upto, minupto, exact */ | 1, 1, 1, /* Type upto, minupto, exact */ |
| Line 198 static const uschar poptable[] = { | Line 223 static const uschar poptable[] = { |
| /* Character class & ref repeats */ | /* Character class & ref repeats */ |
| 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ | 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ |
| 1, 1, /* CRRANGE, CRMINRANGE */ | 1, 1, /* CRRANGE, CRMINRANGE */ |
| 1, 1, 1, 1, /* Possessive *+, ++, ?+, CRPOSRANGE */ | |
| 1, /* CLASS */ | 1, /* CLASS */ |
| 1, /* NCLASS */ | 1, /* NCLASS */ |
| 1, /* XCLASS - variable length */ | 1, /* XCLASS - variable length */ |
| 0, /* REF */ | 0, /* REF */ |
| 0, /* REFI */ | |
| 0, /* DNREF */ | |
| 0, /* DNREFI */ | |
| 0, /* RECURSE */ | 0, /* RECURSE */ |
| 0, /* CALLOUT */ | 0, /* CALLOUT */ |
| 0, /* Alt */ | 0, /* Alt */ |
| 0, /* Ket */ | 0, /* Ket */ |
| 0, /* KetRmax */ | 0, /* KetRmax */ |
| 0, /* KetRmin */ | 0, /* KetRmin */ |
| 0, /* KetRpos */ | |
| 0, /* Reverse */ | |
| 0, /* Assert */ | 0, /* Assert */ |
| 0, /* Assert not */ | 0, /* Assert not */ |
| 0, /* Assert behind */ | 0, /* Assert behind */ |
| 0, /* Assert behind not */ | 0, /* Assert behind not */ |
| 0, /* Reverse */ | 0, 0, /* ONCE, ONCE_NC */ |
| 0, 0, 0, 0, /* ONCE, BRA, CBRA, COND */ | 0, 0, 0, 0, 0, /* BRA, BRAPOS, CBRA, CBRAPOS, COND */ |
| 0, 0, 0, /* SBRA, SCBRA, SCOND */ | 0, 0, 0, 0, 0, /* SBRA, SBRAPOS, SCBRA, SCBRAPOS, SCOND */ |
| 0, 0, /* CREF, NCREF */ | 0, 0, /* CREF, DNCREF */ |
| 0, 0, /* RREF, NRREF */ | 0, 0, /* RREF, DNRREF */ |
| 0, /* DEF */ | 0, /* DEF */ |
| 0, 0, /* BRAZERO, BRAMINZERO */ | 0, 0, 0, /* BRAZERO, BRAMINZERO, BRAPOSZERO */ |
| 0, 0, 0, /* MARK, PRUNE, PRUNE_ARG, */ | 0, 0, 0, /* MARK, PRUNE, PRUNE_ARG */ |
| 0, 0, 0, 0, /* SKIP, SKIP_ARG, THEN, THEN_ARG, */ | 0, 0, 0, 0, /* SKIP, SKIP_ARG, THEN, THEN_ARG */ |
| 0, 0, 0, 0, 0 /* COMMIT, FAIL, ACCEPT, CLOSE, SKIPZERO */ | 0, 0, 0, 0, /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT */ |
| 0, 0 /* CLOSE, SKIPZERO */ | |
| }; | }; |
| /* 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, |
| and \w */ | and \w */ |
| static const uschar toptable1[] = { | static const pcre_uint8 toptable1[] = { |
| 0, 0, 0, 0, 0, 0, | 0, 0, 0, 0, 0, 0, |
| ctype_digit, ctype_digit, | ctype_digit, ctype_digit, |
| ctype_space, ctype_space, | ctype_space, ctype_space, |
| Line 235 static const uschar toptable1[] = { | Line 267 static const uschar toptable1[] = { |
| 0, 0 /* OP_ANY, OP_ALLANY */ | 0, 0 /* OP_ANY, OP_ALLANY */ |
| }; | }; |
| static const uschar toptable2[] = { | static const pcre_uint8 toptable2[] = { |
| 0, 0, 0, 0, 0, 0, | 0, 0, 0, 0, 0, 0, |
| ctype_digit, 0, | ctype_digit, 0, |
| ctype_space, 0, | ctype_space, 0, |
| Line 252 these structures in, is a vector of ints | Line 284 these structures in, is a vector of ints |
| typedef struct stateblock { | typedef struct stateblock { |
| int offset; /* Offset to opcode */ | int offset; /* Offset to opcode */ |
| int count; /* Count for repeats */ | int count; /* Count for repeats */ |
| int ims; /* ims flag bits */ | |
| int data; /* Some use extra data */ | int data; /* Some use extra data */ |
| } stateblock; | } stateblock; |
| #define INTS_PER_STATEBLOCK (sizeof(stateblock)/sizeof(int)) | #define INTS_PER_STATEBLOCK (int)(sizeof(stateblock)/sizeof(int)) |
| #ifdef PCRE_DEBUG | #ifdef PCRE_DEBUG |
| Line 275 Returns: nothing | Line 306 Returns: nothing |
| */ | */ |
| static void | static void |
| pchars(unsigned char *p, int length, FILE *f) | pchars(const pcre_uchar *p, int length, FILE *f) |
| { | { |
| int c; | pcre_uint32 c; |
| while (length-- > 0) | while (length-- > 0) |
| { | { |
| if (isprint(c = *(p++))) | if (isprint(c = *(p++))) |
| fprintf(f, "%c", c); | fprintf(f, "%c", c); |
| else | else |
| fprintf(f, "\\x%02x", c); | fprintf(f, "\\x{%02x}", c); |
| } | } |
| } | } |
| #endif | #endif |
| Line 308 Arguments: | Line 339 Arguments: |
| offsetcount size of same | offsetcount size of same |
| workspace vector of workspace | workspace vector of workspace |
| wscount size of same | wscount size of same |
| ims the current ims flags | |
| rlevel function call recursion level | rlevel function call recursion level |
| recursing regex recursive call level | |
| Returns: > 0 => number of match offset pairs placed in offsets | Returns: > 0 => number of match offset pairs placed in offsets |
| = 0 => offsets overflowed; longest matches are present | = 0 => offsets overflowed; longest matches are present |
| Line 325 for the current character, one for the f | Line 354 for the current character, one for the f |
| { \ | { \ |
| next_active_state->offset = (x); \ | next_active_state->offset = (x); \ |
| next_active_state->count = (y); \ | next_active_state->count = (y); \ |
| next_active_state->ims = ims; \ | |
| next_active_state++; \ | next_active_state++; \ |
| DPRINTF(("%.*sADD_ACTIVE(%d,%d)\n", rlevel*2-2, SP, (x), (y))); \ | DPRINTF(("%.*sADD_ACTIVE(%d,%d)\n", rlevel*2-2, SP, (x), (y))); \ |
| } \ | } \ |
| Line 336 for the current character, one for the f | Line 364 for the current character, one for the f |
| { \ | { \ |
| next_active_state->offset = (x); \ | next_active_state->offset = (x); \ |
| next_active_state->count = (y); \ | next_active_state->count = (y); \ |
| next_active_state->ims = ims; \ | |
| next_active_state->data = (z); \ | next_active_state->data = (z); \ |
| next_active_state++; \ | next_active_state++; \ |
| DPRINTF(("%.*sADD_ACTIVE_DATA(%d,%d,%d)\n", rlevel*2-2, SP, (x), (y), (z))); \ | DPRINTF(("%.*sADD_ACTIVE_DATA(%d,%d,%d)\n", rlevel*2-2, SP, (x), (y), (z))); \ |
| Line 348 for the current character, one for the f | Line 375 for the current character, one for the f |
| { \ | { \ |
| next_new_state->offset = (x); \ | next_new_state->offset = (x); \ |
| next_new_state->count = (y); \ | next_new_state->count = (y); \ |
| next_new_state->ims = ims; \ | |
| next_new_state++; \ | next_new_state++; \ |
| DPRINTF(("%.*sADD_NEW(%d,%d)\n", rlevel*2-2, SP, (x), (y))); \ | DPRINTF(("%.*sADD_NEW(%d,%d)\n", rlevel*2-2, SP, (x), (y))); \ |
| } \ | } \ |
| Line 359 for the current character, one for the f | Line 385 for the current character, one for the f |
| { \ | { \ |
| next_new_state->offset = (x); \ | next_new_state->offset = (x); \ |
| next_new_state->count = (y); \ | next_new_state->count = (y); \ |
| next_new_state->ims = ims; \ | |
| next_new_state->data = (z); \ | next_new_state->data = (z); \ |
| next_new_state++; \ | next_new_state++; \ |
| DPRINTF(("%.*sADD_NEW_DATA(%d,%d,%d)\n", rlevel*2-2, SP, (x), (y), (z))); \ | DPRINTF(("%.*sADD_NEW_DATA(%d,%d,%d) line %d\n", rlevel*2-2, SP, \ |
| (x), (y), (z), __LINE__)); \ | |
| } \ | } \ |
| else return PCRE_ERROR_DFA_WSSIZE | else return PCRE_ERROR_DFA_WSSIZE |
| Line 371 for the current character, one for the f | Line 397 for the current character, one for the f |
| static int | static int |
| internal_dfa_exec( | internal_dfa_exec( |
| dfa_match_data *md, | dfa_match_data *md, |
| const uschar *this_start_code, | const pcre_uchar *this_start_code, |
| const uschar *current_subject, | const pcre_uchar *current_subject, |
| int start_offset, | int start_offset, |
| int *offsets, | int *offsets, |
| int offsetcount, | int offsetcount, |
| int *workspace, | int *workspace, |
| int wscount, | int wscount, |
| int ims, | int rlevel) |
| int rlevel, | |
| int recursing) | |
| { | { |
| stateblock *active_states, *new_states, *temp_states; | stateblock *active_states, *new_states, *temp_states; |
| stateblock *next_active_state, *next_new_state; | stateblock *next_active_state, *next_new_state; |
| const uschar *ctypes, *lcc, *fcc; | const pcre_uint8 *ctypes, *lcc, *fcc; |
| const uschar *ptr; | const pcre_uchar *ptr; |
| const uschar *end_code, *first_op; | const pcre_uchar *end_code, *first_op; |
| dfa_recursion_info new_recursive; | |
| int active_count, new_count, match_count; | int active_count, new_count, match_count; |
| /* Some fields in the md block are frequently referenced, so we load them into | /* Some fields in the md block are frequently referenced, so we load them into |
| independent variables in the hope that this will perform better. */ | independent variables in the hope that this will perform better. */ |
| const uschar *start_subject = md->start_subject; | const pcre_uchar *start_subject = md->start_subject; |
| const uschar *end_subject = md->end_subject; | const pcre_uchar *end_subject = md->end_subject; |
| const uschar *start_code = md->start_code; | const pcre_uchar *start_code = md->start_code; |
| #ifdef SUPPORT_UTF8 | #ifdef SUPPORT_UTF |
| BOOL utf8 = (md->poptions & PCRE_UTF8) != 0; | BOOL utf = (md->poptions & PCRE_UTF8) != 0; |
| #else | #else |
| BOOL utf8 = FALSE; | BOOL utf = FALSE; |
| #endif | #endif |
| BOOL reset_could_continue = FALSE; | |
| rlevel++; | rlevel++; |
| offsetcount &= (-2); | offsetcount &= (-2); |
| Line 412 wscount = (wscount - (wscount % (INTS_PE | Line 440 wscount = (wscount - (wscount % (INTS_PE |
| (2 * INTS_PER_STATEBLOCK); | (2 * INTS_PER_STATEBLOCK); |
| DPRINTF(("\n%.*s---------------------\n" | DPRINTF(("\n%.*s---------------------\n" |
| "%.*sCall to internal_dfa_exec f=%d r=%d\n", | "%.*sCall to internal_dfa_exec f=%d\n", |
| rlevel*2-2, SP, rlevel*2-2, SP, rlevel, recursing)); | rlevel*2-2, SP, rlevel*2-2, SP, rlevel)); |
| ctypes = md->tables + ctypes_offset; | ctypes = md->tables + ctypes_offset; |
| lcc = md->tables + lcc_offset; | lcc = md->tables + lcc_offset; |
| Line 426 next_new_state = new_states = active_sta | Line 454 next_new_state = new_states = active_sta |
| new_count = 0; | new_count = 0; |
| first_op = this_start_code + 1 + LINK_SIZE + | first_op = this_start_code + 1 + LINK_SIZE + |
| ((*this_start_code == OP_CBRA || *this_start_code == OP_SCBRA)? 2:0); | ((*this_start_code == OP_CBRA || *this_start_code == OP_SCBRA || |
| *this_start_code == OP_CBRAPOS || *this_start_code == OP_SCBRAPOS) | |
| ? IMM2_SIZE:0); | |
| /* The first thing in any (sub) pattern is a bracket of some sort. Push all | /* The first thing in any (sub) pattern is a bracket of some sort. Push all |
| the alternative states onto the list, and find out where the end is. This | the alternative states onto the list, and find out where the end is. This |
| Line 454 if (*first_op == OP_REVERSE) | Line 484 if (*first_op == OP_REVERSE) |
| /* If we can't go back the amount required for the longest lookbehind | /* If we can't go back the amount required for the longest lookbehind |
| pattern, go back as far as we can; some alternatives may still be viable. */ | pattern, go back as far as we can; some alternatives may still be viable. */ |
| #ifdef SUPPORT_UTF8 | #ifdef SUPPORT_UTF |
| /* In character mode we have to step back character by character */ | /* In character mode we have to step back character by character */ |
| if (utf8) | if (utf) |
| { | { |
| for (gone_back = 0; gone_back < max_back; gone_back++) | for (gone_back = 0; gone_back < max_back; gone_back++) |
| { | { |
| if (current_subject <= start_subject) break; | if (current_subject <= start_subject) break; |
| current_subject--; | current_subject--; |
| while (current_subject > start_subject && | ACROSSCHAR(current_subject > start_subject, *current_subject, current_subject--); |
| (*current_subject & 0xc0) == 0x80) | |
| current_subject--; | |
| } | } |
| } | } |
| else | else |
| Line 525 else | Line 553 else |
| else | else |
| { | { |
| int length = 1 + LINK_SIZE + | int length = 1 + LINK_SIZE + |
| ((*this_start_code == OP_CBRA || *this_start_code == OP_SCBRA)? 2:0); | ((*this_start_code == OP_CBRA || *this_start_code == OP_SCBRA || |
| *this_start_code == OP_CBRAPOS || *this_start_code == OP_SCBRAPOS) | |
| ? IMM2_SIZE:0); | |
| do | do |
| { | { |
| ADD_NEW((int)(end_code - start_code + length), 0); | ADD_NEW((int)(end_code - start_code + length), 0); |
| Line 538 else | Line 568 else |
| workspace[0] = 0; /* Bit indicating which vector is current */ | workspace[0] = 0; /* Bit indicating which vector is current */ |
| DPRINTF(("%.*sEnd state = %d\n", rlevel*2-2, SP, end_code - start_code)); | DPRINTF(("%.*sEnd state = %d\n", rlevel*2-2, SP, (int)(end_code - start_code))); |
| /* Loop for scanning the subject */ | /* Loop for scanning the subject */ |
| Line 547 for (;;) | Line 577 for (;;) |
| { | { |
| int i, j; | int i, j; |
| int clen, dlen; | int clen, dlen; |
| unsigned int c, d; | pcre_uint32 c, d; |
| int forced_fail = 0; | int forced_fail = 0; |
| BOOL could_continue = FALSE; | BOOL partial_newline = FALSE; |
| BOOL could_continue = reset_could_continue; | |
| reset_could_continue = FALSE; | |
| /* Make the new state list into the active state list and empty the | /* Make the new state list into the active state list and empty the |
| new state list. */ | new state list. */ |
| Line 565 for (;;) | Line 597 for (;;) |
| #ifdef PCRE_DEBUG | #ifdef PCRE_DEBUG |
| printf("%.*sNext character: rest of subject = \"", rlevel*2-2, SP); | printf("%.*sNext character: rest of subject = \"", rlevel*2-2, SP); |
| pchars((uschar *)ptr, strlen((char *)ptr), stdout); | pchars(ptr, STRLEN_UC(ptr), stdout); |
| printf("\"\n"); | printf("\"\n"); |
| printf("%.*sActive states: ", rlevel*2-2, SP); | printf("%.*sActive states: ", rlevel*2-2, SP); |
| Line 585 for (;;) | Line 617 for (;;) |
| if (ptr < end_subject) | if (ptr < end_subject) |
| { | { |
| clen = 1; /* Number of bytes in the character */ | clen = 1; /* Number of data items in the character */ |
| #ifdef SUPPORT_UTF8 | #ifdef SUPPORT_UTF |
| if (utf8) { GETCHARLEN(c, ptr, clen); } else | GETCHARLENTEST(c, ptr, clen); |
| #endif /* SUPPORT_UTF8 */ | #else |
| c = *ptr; | c = *ptr; |
| #endif /* SUPPORT_UTF */ | |
| } | } |
| else | else |
| { | { |
| Line 605 for (;;) | Line 638 for (;;) |
| for (i = 0; i < active_count; i++) | for (i = 0; i < active_count; i++) |
| { | { |
| stateblock *current_state = active_states + i; | stateblock *current_state = active_states + i; |
| const uschar *code; | BOOL caseless = FALSE; |
| const pcre_uchar *code; | |
| int state_offset = current_state->offset; | int state_offset = current_state->offset; |
| int count, codevalue, rrc; | int codevalue, rrc; |
| int count; | |
| #ifdef PCRE_DEBUG | #ifdef PCRE_DEBUG |
| printf ("%.*sProcessing state %d c=", rlevel*2-2, SP, state_offset); | printf ("%.*sProcessing state %d c=", rlevel*2-2, SP, state_offset); |
| Line 616 for (;;) | Line 651 for (;;) |
| else printf("0x%02x\n", c); | else printf("0x%02x\n", c); |
| #endif | #endif |
| /* This variable is referred to implicity in the ADD_xxx macros. */ | |
| ims = current_state->ims; | |
| /* A negative offset is a special case meaning "hold off going to this | /* A negative offset is a special case meaning "hold off going to this |
| (negated) state until the number of characters in the data field have | (negated) state until the number of characters in the data field have |
| been skipped". */ | been skipped". If the could_continue flag was passed over from a previous |
| state, arrange for it to passed on. */ | |
| if (state_offset < 0) | if (state_offset < 0) |
| { | { |
| Line 631 for (;;) | Line 663 for (;;) |
| DPRINTF(("%.*sSkipping this character\n", rlevel*2-2, SP)); | DPRINTF(("%.*sSkipping this character\n", rlevel*2-2, SP)); |
| ADD_NEW_DATA(state_offset, current_state->count, | ADD_NEW_DATA(state_offset, current_state->count, |
| current_state->data - 1); | current_state->data - 1); |
| if (could_continue) reset_could_continue = TRUE; | |
| continue; | continue; |
| } | } |
| else | else |
| Line 670 for (;;) | Line 703 for (;;) |
| permitted. | permitted. |
| 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 |
| argument that is not a data character - but is always one byte long. We | argument that is not a data character - but is always one byte long because |
| have to take special action to deal with \P, \p, \H, \h, \V, \v and \X in | the values are small. We have to take special action to deal with \P, \p, |
| this case. To keep the other cases fast, convert these ones to new opcodes. | \H, \h, \V, \v and \X in this case. To keep the other cases fast, convert |
| */ | these ones to new opcodes. */ |
| if (coptable[codevalue] > 0) | if (coptable[codevalue] > 0) |
| { | { |
| dlen = 1; | dlen = 1; |
| #ifdef SUPPORT_UTF8 | #ifdef SUPPORT_UTF |
| if (utf8) { GETCHARLEN(d, (code + coptable[codevalue]), dlen); } else | if (utf) { GETCHARLEN(d, (code + coptable[codevalue]), dlen); } else |
| #endif /* SUPPORT_UTF8 */ | #endif /* SUPPORT_UTF */ |
| d = code[coptable[codevalue]]; | d = code[coptable[codevalue]]; |
| if (codevalue >= OP_TYPESTAR) | if (codevalue >= OP_TYPESTAR) |
| { | { |
| Line 725 for (;;) | Line 758 for (;;) |
| /* ========================================================================== */ | /* ========================================================================== */ |
| /* 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 |
| on with the next opcode. Otherwise, unless we have an empty string and | on with the next opcode. For repeating opcodes, also add the repeat |
| state. Note that KETRPOS will always be encountered at the end of the | |
| subpattern, because the possessive subpattern repeats are always handled | |
| using recursive calls. Thus, it never adds any new states. | |
| At the end of the (sub)pattern, unless we have an empty string and | |
| PCRE_NOTEMPTY is set, or PCRE_NOTEMPTY_ATSTART is set and we are at the | PCRE_NOTEMPTY is set, or PCRE_NOTEMPTY_ATSTART is set and we are at the |
| start of the subject, save the match data, shifting up all previous | start of the subject, save the match data, shifting up all previous |
| matches so we always have the longest first. */ | matches so we always have the longest first. */ |
| Line 733 for (;;) | Line 771 for (;;) |
| case OP_KET: | case OP_KET: |
| case OP_KETRMIN: | case OP_KETRMIN: |
| case OP_KETRMAX: | case OP_KETRMAX: |
| case OP_KETRPOS: | |
| if (code != end_code) | if (code != end_code) |
| { | { |
| ADD_ACTIVE(state_offset + 1 + LINK_SIZE, 0); | ADD_ACTIVE(state_offset + 1 + LINK_SIZE, 0); |
| Line 749 for (;;) | Line 788 for (;;) |
| current_subject > start_subject + md->start_offset))) | current_subject > start_subject + md->start_offset))) |
| { | { |
| if (match_count < 0) match_count = (offsetcount >= 2)? 1 : 0; | if (match_count < 0) match_count = (offsetcount >= 2)? 1 : 0; |
| else if (match_count > 0 && ++match_count * 2 >= offsetcount) | else if (match_count > 0 && ++match_count * 2 > offsetcount) |
| match_count = 0; | match_count = 0; |
| count = ((match_count == 0)? offsetcount : match_count * 2) - 2; | count = ((match_count == 0)? offsetcount : match_count * 2) - 2; |
| if (count > 0) memmove(offsets + 2, offsets, count * sizeof(int)); | if (count > 0) memmove(offsets + 2, offsets, count * sizeof(int)); |
| Line 758 for (;;) | Line 797 for (;;) |
| offsets[0] = (int)(current_subject - start_subject); | offsets[0] = (int)(current_subject - start_subject); |
| offsets[1] = (int)(ptr - start_subject); | offsets[1] = (int)(ptr - start_subject); |
| DPRINTF(("%.*sSet matched string = \"%.*s\"\n", rlevel*2-2, SP, | DPRINTF(("%.*sSet matched string = \"%.*s\"\n", rlevel*2-2, SP, |
| offsets[1] - offsets[0], current_subject)); | offsets[1] - offsets[0], (char *)current_subject)); |
| } | } |
| if ((md->moptions & PCRE_DFA_SHORTEST) != 0) | if ((md->moptions & PCRE_DFA_SHORTEST) != 0) |
| { | { |
| Line 795 for (;;) | Line 834 for (;;) |
| /*-----------------------------------------------------------------*/ | /*-----------------------------------------------------------------*/ |
| case OP_CBRA: | case OP_CBRA: |
| case OP_SCBRA: | case OP_SCBRA: |
| ADD_ACTIVE((int)(code - start_code + 3 + LINK_SIZE), 0); | ADD_ACTIVE((int)(code - start_code + 1 + LINK_SIZE + IMM2_SIZE), 0); |
| code += GET(code, 1); | code += GET(code, 1); |
| while (*code == OP_ALT) | while (*code == OP_ALT) |
| { | { |
| Line 822 for (;;) | Line 861 for (;;) |
| /*-----------------------------------------------------------------*/ | /*-----------------------------------------------------------------*/ |
| case OP_CIRC: | case OP_CIRC: |
| if (ptr == start_subject && (md->moptions & PCRE_NOTBOL) == 0) | |
| { ADD_ACTIVE(state_offset + 1, 0); } | |
| break; | |
| /*-----------------------------------------------------------------*/ | |
| case OP_CIRCM: | |
| if ((ptr == start_subject && (md->moptions & PCRE_NOTBOL) == 0) || | if ((ptr == start_subject && (md->moptions & PCRE_NOTBOL) == 0) || |
| ((ims & PCRE_MULTILINE) != 0 && | (ptr != end_subject && WAS_NEWLINE(ptr))) |
| ptr != end_subject && | |
| WAS_NEWLINE(ptr))) | |
| { ADD_ACTIVE(state_offset + 1, 0); } | { ADD_ACTIVE(state_offset + 1, 0); } |
| break; | break; |
| Line 840 for (;;) | Line 883 for (;;) |
| break; | break; |
| /*-----------------------------------------------------------------*/ | /*-----------------------------------------------------------------*/ |
| case OP_OPT: | |
| ims = code[1]; | |
| ADD_ACTIVE(state_offset + 2, 0); | |
| break; | |
| /*-----------------------------------------------------------------*/ | |
| case OP_SOD: | case OP_SOD: |
| if (ptr == start_subject) { ADD_ACTIVE(state_offset + 1, 0); } | if (ptr == start_subject) { ADD_ACTIVE(state_offset + 1, 0); } |
| break; | break; |
| Line 865 for (;;) | Line 902 for (;;) |
| /*-----------------------------------------------------------------*/ | /*-----------------------------------------------------------------*/ |
| case OP_ANY: | case OP_ANY: |
| if (clen > 0 && !IS_NEWLINE(ptr)) | if (clen > 0 && !IS_NEWLINE(ptr)) |
| { ADD_NEW(state_offset + 1, 0); } | { |
| if (ptr + 1 >= md->end_subject && | |
| (md->moptions & (PCRE_PARTIAL_HARD)) != 0 && | |
| NLBLOCK->nltype == NLTYPE_FIXED && | |
| NLBLOCK->nllen == 2 && | |
| c == NLBLOCK->nl[0]) | |
| { | |
| could_continue = partial_newline = TRUE; | |
| } | |
| else | |
| { | |
| ADD_NEW(state_offset + 1, 0); | |
| } | |
| } | |
| break; | break; |
| /*-----------------------------------------------------------------*/ | /*-----------------------------------------------------------------*/ |
| Line 890 for (;;) | Line 940 for (;;) |
| could_continue = TRUE; | could_continue = TRUE; |
| else if (clen == 0 || | else if (clen == 0 || |
| ((md->poptions & PCRE_DOLLAR_ENDONLY) == 0 && IS_NEWLINE(ptr) && | ((md->poptions & PCRE_DOLLAR_ENDONLY) == 0 && IS_NEWLINE(ptr) && |
| ((ims & PCRE_MULTILINE) != 0 || ptr == end_subject - md->nllen) | (ptr == end_subject - md->nllen) |
| )) | )) |
| { ADD_ACTIVE(state_offset + 1, 0); } | { ADD_ACTIVE(state_offset + 1, 0); } |
| else if (ptr + 1 >= md->end_subject && | |
| (md->moptions & (PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT)) != 0 && | |
| NLBLOCK->nltype == NLTYPE_FIXED && | |
| NLBLOCK->nllen == 2 && | |
| c == NLBLOCK->nl[0]) | |
| { | |
| if ((md->moptions & PCRE_PARTIAL_HARD) != 0) | |
| { | |
| reset_could_continue = TRUE; | |
| ADD_NEW_DATA(-(state_offset + 1), 0, 1); | |
| } | |
| else could_continue = partial_newline = TRUE; | |
| } | |
| } | |
| break; | |
| /*-----------------------------------------------------------------*/ | |
| case OP_DOLLM: | |
| if ((md->moptions & PCRE_NOTEOL) == 0) | |
| { | |
| if (clen == 0 && (md->moptions & PCRE_PARTIAL_HARD) != 0) | |
| could_continue = TRUE; | |
| else if (clen == 0 || | |
| ((md->poptions & PCRE_DOLLAR_ENDONLY) == 0 && IS_NEWLINE(ptr))) | |
| { ADD_ACTIVE(state_offset + 1, 0); } | |
| else if (ptr + 1 >= md->end_subject && | |
| (md->moptions & (PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT)) != 0 && | |
| NLBLOCK->nltype == NLTYPE_FIXED && | |
| NLBLOCK->nllen == 2 && | |
| c == NLBLOCK->nl[0]) | |
| { | |
| if ((md->moptions & PCRE_PARTIAL_HARD) != 0) | |
| { | |
| reset_could_continue = TRUE; | |
| ADD_NEW_DATA(-(state_offset + 1), 0, 1); | |
| } | |
| else could_continue = partial_newline = TRUE; | |
| } | |
| } | } |
| else if ((ims & PCRE_MULTILINE) != 0 && IS_NEWLINE(ptr)) | else if (IS_NEWLINE(ptr)) |
| { ADD_ACTIVE(state_offset + 1, 0); } | { ADD_ACTIVE(state_offset + 1, 0); } |
| break; | break; |
| Line 925 for (;;) | Line 1013 for (;;) |
| if (ptr > start_subject) | if (ptr > start_subject) |
| { | { |
| const uschar *temp = ptr - 1; | const pcre_uchar *temp = ptr - 1; |
| if (temp < md->start_used_ptr) md->start_used_ptr = temp; | if (temp < md->start_used_ptr) md->start_used_ptr = temp; |
| #ifdef SUPPORT_UTF8 | #if defined SUPPORT_UTF && !defined COMPILE_PCRE32 |
| if (utf8) BACKCHAR(temp); | if (utf) { BACKCHAR(temp); } |
| #endif | #endif |
| GETCHARTEST(d, temp); | GETCHARTEST(d, temp); |
| #ifdef SUPPORT_UCP | #ifdef SUPPORT_UCP |
| Line 980 for (;;) | Line 1068 for (;;) |
| if (clen > 0) | if (clen > 0) |
| { | { |
| BOOL OK; | BOOL OK; |
| const pcre_uint32 *cp; | |
| const ucd_record * prop = GET_UCD(c); | const ucd_record * prop = GET_UCD(c); |
| switch(code[1]) | switch(code[1]) |
| { | { |
| Line 993 for (;;) | Line 1082 for (;;) |
| break; | break; |
| case PT_GC: | case PT_GC: |
| OK = _pcre_ucp_gentype[prop->chartype] == code[2]; | OK = PRIV(ucp_gentype)[prop->chartype] == code[2]; |
| break; | break; |
| case PT_PC: | case PT_PC: |
| Line 1007 for (;;) | Line 1096 for (;;) |
| /* These are specials for combination cases. */ | /* These are specials for combination cases. */ |
| case PT_ALNUM: | case PT_ALNUM: |
| OK = _pcre_ucp_gentype[prop->chartype] == ucp_L || | OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L || |
| _pcre_ucp_gentype[prop->chartype] == ucp_N; | PRIV(ucp_gentype)[prop->chartype] == ucp_N; |
| break; | break; |
| case PT_SPACE: /* Perl space */ | /* Perl space used to exclude VT, but from Perl 5.18 it is included, |
| OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z || | which means that Perl space and POSIX space are now identical. PCRE |
| c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR; | was changed at release 8.34. */ |
| break; | |
| case PT_SPACE: /* Perl space */ | |
| case PT_PXSPACE: /* POSIX space */ | case PT_PXSPACE: /* POSIX space */ |
| OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z || | switch(c) |
| c == CHAR_HT || c == CHAR_NL || c == CHAR_VT || | { |
| c == CHAR_FF || c == CHAR_CR; | HSPACE_CASES: |
| VSPACE_CASES: | |
| OK = TRUE; | |
| break; | |
| default: | |
| OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z; | |
| break; | |
| } | |
| break; | break; |
| case PT_WORD: | case PT_WORD: |
| OK = _pcre_ucp_gentype[prop->chartype] == ucp_L || | OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L || |
| _pcre_ucp_gentype[prop->chartype] == ucp_N || | PRIV(ucp_gentype)[prop->chartype] == ucp_N || |
| c == CHAR_UNDERSCORE; | c == CHAR_UNDERSCORE; |
| break; | break; |
| case PT_CLIST: | |
| cp = PRIV(ucd_caseless_sets) + code[2]; | |
| for (;;) | |
| { | |
| if (c < *cp) { OK = FALSE; break; } | |
| if (c == *cp++) { OK = TRUE; break; } | |
| } | |
| break; | |
| case PT_UCNC: | |
| OK = c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT || | |
| c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) || | |
| c >= 0xe000; | |
| break; | |
| /* Should never occur, but keep compilers from grumbling. */ | /* Should never occur, but keep compilers from grumbling. */ |
| default: | default: |
| Line 1055 for (;;) | Line 1167 for (;;) |
| if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); } | if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); } |
| if (clen > 0) | if (clen > 0) |
| { | { |
| if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || | if (d == OP_ANY && ptr + 1 >= md->end_subject && |
| (md->moptions & (PCRE_PARTIAL_HARD)) != 0 && | |
| NLBLOCK->nltype == NLTYPE_FIXED && | |
| NLBLOCK->nllen == 2 && | |
| c == NLBLOCK->nl[0]) | |
| { | |
| could_continue = partial_newline = TRUE; | |
| } | |
| else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || | |
| (c < 256 && | (c < 256 && |
| (d != OP_ANY || !IS_NEWLINE(ptr)) && | (d != OP_ANY || !IS_NEWLINE(ptr)) && |
| ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) | ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) |
| Line 1078 for (;;) | Line 1198 for (;;) |
| ADD_ACTIVE(state_offset + 2, 0); | ADD_ACTIVE(state_offset + 2, 0); |
| if (clen > 0) | if (clen > 0) |
| { | { |
| if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || | if (d == OP_ANY && ptr + 1 >= md->end_subject && |
| (md->moptions & (PCRE_PARTIAL_HARD)) != 0 && | |
| NLBLOCK->nltype == NLTYPE_FIXED && | |
| NLBLOCK->nllen == 2 && | |
| c == NLBLOCK->nl[0]) | |
| { | |
| could_continue = partial_newline = TRUE; | |
| } | |
| else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || | |
| (c < 256 && | (c < 256 && |
| (d != OP_ANY || !IS_NEWLINE(ptr)) && | (d != OP_ANY || !IS_NEWLINE(ptr)) && |
| ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) | ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) |
| Line 1100 for (;;) | Line 1228 for (;;) |
| ADD_ACTIVE(state_offset + 2, 0); | ADD_ACTIVE(state_offset + 2, 0); |
| if (clen > 0) | if (clen > 0) |
| { | { |
| if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || | if (d == OP_ANY && ptr + 1 >= md->end_subject && |
| (md->moptions & (PCRE_PARTIAL_HARD)) != 0 && | |
| NLBLOCK->nltype == NLTYPE_FIXED && | |
| NLBLOCK->nllen == 2 && | |
| c == NLBLOCK->nl[0]) | |
| { | |
| could_continue = partial_newline = TRUE; | |
| } | |
| else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || | |
| (c < 256 && | (c < 256 && |
| (d != OP_ANY || !IS_NEWLINE(ptr)) && | (d != OP_ANY || !IS_NEWLINE(ptr)) && |
| ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) | ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) |
| Line 1120 for (;;) | Line 1256 for (;;) |
| count = current_state->count; /* Number already matched */ | count = current_state->count; /* Number already matched */ |
| if (clen > 0) | if (clen > 0) |
| { | { |
| if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || | if (d == OP_ANY && ptr + 1 >= md->end_subject && |
| (md->moptions & (PCRE_PARTIAL_HARD)) != 0 && | |
| NLBLOCK->nltype == NLTYPE_FIXED && | |
| NLBLOCK->nllen == 2 && | |
| c == NLBLOCK->nl[0]) | |
| { | |
| could_continue = partial_newline = TRUE; | |
| } | |
| else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || | |
| (c < 256 && | (c < 256 && |
| (d != OP_ANY || !IS_NEWLINE(ptr)) && | (d != OP_ANY || !IS_NEWLINE(ptr)) && |
| ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) | ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) |
| { | { |
| if (++count >= GET2(code, 1)) | if (++count >= (int)GET2(code, 1)) |
| { ADD_NEW(state_offset + 4, 0); } | { ADD_NEW(state_offset + 1 + IMM2_SIZE + 1, 0); } |
| else | else |
| { ADD_NEW(state_offset, count); } | { ADD_NEW(state_offset, count); } |
| } | } |
| Line 1137 for (;;) | Line 1281 for (;;) |
| case OP_TYPEUPTO: | case OP_TYPEUPTO: |
| case OP_TYPEMINUPTO: | case OP_TYPEMINUPTO: |
| case OP_TYPEPOSUPTO: | case OP_TYPEPOSUPTO: |
| ADD_ACTIVE(state_offset + 4, 0); | ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0); |
| count = current_state->count; /* Number already matched */ | count = current_state->count; /* Number already matched */ |
| if (clen > 0) | if (clen > 0) |
| { | { |
| if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || | if (d == OP_ANY && ptr + 1 >= md->end_subject && |
| (md->moptions & (PCRE_PARTIAL_HARD)) != 0 && | |
| NLBLOCK->nltype == NLTYPE_FIXED && | |
| NLBLOCK->nllen == 2 && | |
| c == NLBLOCK->nl[0]) | |
| { | |
| could_continue = partial_newline = TRUE; | |
| } | |
| else if ((c >= 256 && d != OP_DIGIT && d != OP_WHITESPACE && d != OP_WORDCHAR) || | |
| (c < 256 && | (c < 256 && |
| (d != OP_ANY || !IS_NEWLINE(ptr)) && | (d != OP_ANY || !IS_NEWLINE(ptr)) && |
| ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) | ((ctypes[c] & toptable1[d]) ^ toptable2[d]) != 0)) |
| Line 1151 for (;;) | Line 1303 for (;;) |
| active_count--; /* Remove non-match possibility */ | active_count--; /* Remove non-match possibility */ |
| next_active_state--; | next_active_state--; |
| } | } |
| if (++count >= GET2(code, 1)) | if (++count >= (int)GET2(code, 1)) |
| { ADD_NEW(state_offset + 4, 0); } | { ADD_NEW(state_offset + 2 + IMM2_SIZE, 0); } |
| else | else |
| { ADD_NEW(state_offset, count); } | { ADD_NEW(state_offset, count); } |
| } | } |
| Line 1174 for (;;) | Line 1326 for (;;) |
| if (clen > 0) | if (clen > 0) |
| { | { |
| BOOL OK; | BOOL OK; |
| const pcre_uint32 *cp; | |
| const ucd_record * prop = GET_UCD(c); | const ucd_record * prop = GET_UCD(c); |
| switch(code[2]) | switch(code[2]) |
| { | { |
| Line 1187 for (;;) | Line 1340 for (;;) |
| break; | break; |
| case PT_GC: | case PT_GC: |
| OK = _pcre_ucp_gentype[prop->chartype] == code[3]; | OK = PRIV(ucp_gentype)[prop->chartype] == code[3]; |
| break; | break; |
| case PT_PC: | case PT_PC: |
| Line 1201 for (;;) | Line 1354 for (;;) |
| /* These are specials for combination cases. */ | /* These are specials for combination cases. */ |
| case PT_ALNUM: | case PT_ALNUM: |
| OK = _pcre_ucp_gentype[prop->chartype] == ucp_L || | OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L || |
| _pcre_ucp_gentype[prop->chartype] == ucp_N; | PRIV(ucp_gentype)[prop->chartype] == ucp_N; |
| break; | break; |
| case PT_SPACE: /* Perl space */ | /* Perl space used to exclude VT, but from Perl 5.18 it is included, |
| OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z || | which means that Perl space and POSIX space are now identical. PCRE |
| c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR; | was changed at release 8.34. */ |
| break; | |
| case PT_SPACE: /* Perl space */ | |
| case PT_PXSPACE: /* POSIX space */ | case PT_PXSPACE: /* POSIX space */ |
| OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z || | switch(c) |
| c == CHAR_HT || c == CHAR_NL || c == CHAR_VT || | { |
| c == CHAR_FF || c == CHAR_CR; | HSPACE_CASES: |
| VSPACE_CASES: | |
| OK = TRUE; | |
| break; | |
| default: | |
| OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z; | |
| break; | |
| } | |
| break; | break; |
| case PT_WORD: | case PT_WORD: |
| OK = _pcre_ucp_gentype[prop->chartype] == ucp_L || | OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L || |
| _pcre_ucp_gentype[prop->chartype] == ucp_N || | PRIV(ucp_gentype)[prop->chartype] == ucp_N || |
| c == CHAR_UNDERSCORE; | c == CHAR_UNDERSCORE; |
| break; | break; |
| case PT_CLIST: | |
| cp = PRIV(ucd_caseless_sets) + code[3]; | |
| for (;;) | |
| { | |
| if (c < *cp) { OK = FALSE; break; } | |
| if (c == *cp++) { OK = TRUE; break; } | |
| } | |
| break; | |
| case PT_UCNC: | |
| OK = c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT || | |
| c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) || | |
| c >= 0xe000; | |
| break; | |
| /* Should never occur, but keep compilers from grumbling. */ | /* Should never occur, but keep compilers from grumbling. */ |
| default: | default: |
| Line 1248 for (;;) | Line 1424 for (;;) |
| case OP_EXTUNI_EXTRA + OP_TYPEPOSPLUS: | case OP_EXTUNI_EXTRA + OP_TYPEPOSPLUS: |
| count = current_state->count; /* Already matched */ | count = current_state->count; /* Already matched */ |
| if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); } | if (count > 0) { ADD_ACTIVE(state_offset + 2, 0); } |
| if (clen > 0 && UCD_CATEGORY(c) != ucp_M) | if (clen > 0) |
| { | { |
| const uschar *nptr = ptr + clen; | int lgb, rgb; |
| const pcre_uchar *nptr = ptr + clen; | |
| int ncount = 0; | int ncount = 0; |
| if (count > 0 && codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSPLUS) | if (count > 0 && codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSPLUS) |
| { | { |
| active_count--; /* Remove non-match possibility */ | active_count--; /* Remove non-match possibility */ |
| next_active_state--; | next_active_state--; |
| } | } |
| lgb = UCD_GRAPHBREAK(c); | |
| while (nptr < end_subject) | while (nptr < end_subject) |
| { | { |
| int nd; | dlen = 1; |
| int ndlen = 1; | if (!utf) d = *nptr; else { GETCHARLEN(d, nptr, dlen); } |
| GETCHARLEN(nd, nptr, ndlen); | rgb = UCD_GRAPHBREAK(d); |
| if (UCD_CATEGORY(nd) != ucp_M) break; | if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break; |
| ncount++; | ncount++; |
| nptr += ndlen; | lgb = rgb; |
| nptr += dlen; | |
| } | } |
| count++; | count++; |
| ADD_NEW_DATA(-state_offset, count, ncount); | ADD_NEW_DATA(-state_offset, count, ncount); |
| Line 1283 for (;;) | Line 1462 for (;;) |
| int ncount = 0; | int ncount = 0; |
| switch (c) | switch (c) |
| { | { |
| case 0x000b: | case CHAR_VT: |
| case 0x000c: | case CHAR_FF: |
| case 0x0085: | case CHAR_NEL: |
| #ifndef EBCDIC | |
| case 0x2028: | case 0x2028: |
| case 0x2029: | case 0x2029: |
| #endif /* Not EBCDIC */ | |
| if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; | if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; |
| goto ANYNL01; | goto ANYNL01; |
| case 0x000d: | case CHAR_CR: |
| if (ptr + 1 < end_subject && ptr[1] == 0x0a) ncount = 1; | if (ptr + 1 < end_subject && UCHAR21TEST(ptr + 1) == CHAR_LF) ncount = 1; |
| /* Fall through */ | /* Fall through */ |
| ANYNL01: | ANYNL01: |
| case 0x000a: | case CHAR_LF: |
| if (count > 0 && codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSPLUS) | if (count > 0 && codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSPLUS) |
| { | { |
| active_count--; /* Remove non-match possibility */ | active_count--; /* Remove non-match possibility */ |
| Line 1323 for (;;) | Line 1504 for (;;) |
| BOOL OK; | BOOL OK; |
| switch (c) | switch (c) |
| { | { |
| case 0x000a: | VSPACE_CASES: |
| case 0x000b: | |
| case 0x000c: | |
| case 0x000d: | |
| case 0x0085: | |
| case 0x2028: | |
| case 0x2029: | |
| OK = TRUE; | OK = TRUE; |
| break; | break; |
| Line 1362 for (;;) | Line 1537 for (;;) |
| BOOL OK; | BOOL OK; |
| switch (c) | switch (c) |
| { | { |
| case 0x09: /* HT */ | HSPACE_CASES: |
| case 0x20: /* SPACE */ | |
| case 0xa0: /* NBSP */ | |
| case 0x1680: /* OGHAM SPACE MARK */ | |
| case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */ | |
| case 0x2000: /* EN QUAD */ | |
| case 0x2001: /* EM QUAD */ | |
| case 0x2002: /* EN SPACE */ | |
| case 0x2003: /* EM SPACE */ | |
| case 0x2004: /* THREE-PER-EM SPACE */ | |
| case 0x2005: /* FOUR-PER-EM SPACE */ | |
| case 0x2006: /* SIX-PER-EM SPACE */ | |
| case 0x2007: /* FIGURE SPACE */ | |
| case 0x2008: /* PUNCTUATION SPACE */ | |
| case 0x2009: /* THIN SPACE */ | |
| case 0x200A: /* HAIR SPACE */ | |
| case 0x202f: /* NARROW NO-BREAK SPACE */ | |
| case 0x205f: /* MEDIUM MATHEMATICAL SPACE */ | |
| case 0x3000: /* IDEOGRAPHIC SPACE */ | |
| OK = TRUE; | OK = TRUE; |
| break; | break; |
| Line 1421 for (;;) | Line 1578 for (;;) |
| if (clen > 0) | if (clen > 0) |
| { | { |
| BOOL OK; | BOOL OK; |
| const pcre_uint32 *cp; | |
| const ucd_record * prop = GET_UCD(c); | const ucd_record * prop = GET_UCD(c); |
| switch(code[2]) | switch(code[2]) |
| { | { |
| Line 1434 for (;;) | Line 1592 for (;;) |
| break; | break; |
| case PT_GC: | case PT_GC: |
| OK = _pcre_ucp_gentype[prop->chartype] == code[3]; | OK = PRIV(ucp_gentype)[prop->chartype] == code[3]; |
| break; | break; |
| case PT_PC: | case PT_PC: |
| Line 1448 for (;;) | Line 1606 for (;;) |
| /* These are specials for combination cases. */ | /* These are specials for combination cases. */ |
| case PT_ALNUM: | case PT_ALNUM: |
| OK = _pcre_ucp_gentype[prop->chartype] == ucp_L || | OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L || |
| _pcre_ucp_gentype[prop->chartype] == ucp_N; | PRIV(ucp_gentype)[prop->chartype] == ucp_N; |
| break; | break; |
| case PT_SPACE: /* Perl space */ | /* Perl space used to exclude VT, but from Perl 5.18 it is included, |
| OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z || | which means that Perl space and POSIX space are now identical. PCRE |
| c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR; | was changed at release 8.34. */ |
| break; | |
| case PT_SPACE: /* Perl space */ | |
| case PT_PXSPACE: /* POSIX space */ | case PT_PXSPACE: /* POSIX space */ |
| OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z || | switch(c) |
| c == CHAR_HT || c == CHAR_NL || c == CHAR_VT || | { |
| c == CHAR_FF || c == CHAR_CR; | HSPACE_CASES: |
| VSPACE_CASES: | |
| OK = TRUE; | |
| break; | |
| default: | |
| OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z; | |
| break; | |
| } | |
| break; | break; |
| case PT_WORD: | case PT_WORD: |
| OK = _pcre_ucp_gentype[prop->chartype] == ucp_L || | OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L || |
| _pcre_ucp_gentype[prop->chartype] == ucp_N || | PRIV(ucp_gentype)[prop->chartype] == ucp_N || |
| c == CHAR_UNDERSCORE; | c == CHAR_UNDERSCORE; |
| break; | break; |
| case PT_CLIST: | |
| cp = PRIV(ucd_caseless_sets) + code[3]; | |
| for (;;) | |
| { | |
| if (c < *cp) { OK = FALSE; break; } | |
| if (c == *cp++) { OK = TRUE; break; } | |
| } | |
| break; | |
| case PT_UCNC: | |
| OK = c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT || | |
| c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) || | |
| c >= 0xe000; | |
| break; | |
| /* Should never occur, but keep compilers from grumbling. */ | /* Should never occur, but keep compilers from grumbling. */ |
| default: | default: |
| Line 1504 for (;;) | Line 1685 for (;;) |
| QS2: | QS2: |
| ADD_ACTIVE(state_offset + 2, 0); | ADD_ACTIVE(state_offset + 2, 0); |
| if (clen > 0 && UCD_CATEGORY(c) != ucp_M) | if (clen > 0) |
| { | { |
| const uschar *nptr = ptr + clen; | int lgb, rgb; |
| const pcre_uchar *nptr = ptr + clen; | |
| int ncount = 0; | int ncount = 0; |
| if (codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSSTAR || | if (codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSSTAR || |
| codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSQUERY) | codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSQUERY) |
| Line 1514 for (;;) | Line 1696 for (;;) |
| active_count--; /* Remove non-match possibility */ | active_count--; /* Remove non-match possibility */ |
| next_active_state--; | next_active_state--; |
| } | } |
| lgb = UCD_GRAPHBREAK(c); | |
| while (nptr < end_subject) | while (nptr < end_subject) |
| { | { |
| int nd; | dlen = 1; |
| int ndlen = 1; | if (!utf) d = *nptr; else { GETCHARLEN(d, nptr, dlen); } |
| GETCHARLEN(nd, nptr, ndlen); | rgb = UCD_GRAPHBREAK(d); |
| if (UCD_CATEGORY(nd) != ucp_M) break; | if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break; |
| ncount++; | ncount++; |
| nptr += ndlen; | lgb = rgb; |
| nptr += dlen; | |
| } | } |
| ADD_NEW_DATA(-(state_offset + count), 0, ncount); | ADD_NEW_DATA(-(state_offset + count), 0, ncount); |
| } | } |
| Line 1547 for (;;) | Line 1731 for (;;) |
| int ncount = 0; | int ncount = 0; |
| switch (c) | switch (c) |
| { | { |
| case 0x000b: | case CHAR_VT: |
| case 0x000c: | case CHAR_FF: |
| case 0x0085: | case CHAR_NEL: |
| #ifndef EBCDIC | |
| case 0x2028: | case 0x2028: |
| case 0x2029: | case 0x2029: |
| #endif /* Not EBCDIC */ | |
| if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; | if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; |
| goto ANYNL02; | goto ANYNL02; |
| case 0x000d: | case CHAR_CR: |
| if (ptr + 1 < end_subject && ptr[1] == 0x0a) ncount = 1; | if (ptr + 1 < end_subject && UCHAR21TEST(ptr + 1) == CHAR_LF) ncount = 1; |
| /* Fall through */ | /* Fall through */ |
| ANYNL02: | ANYNL02: |
| case 0x000a: | case CHAR_LF: |
| if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSSTAR || | if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSSTAR || |
| codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSQUERY) | codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSQUERY) |
| { | { |
| active_count--; /* Remove non-match possibility */ | active_count--; /* Remove non-match possibility */ |
| next_active_state--; | next_active_state--; |
| } | } |
| ADD_NEW_DATA(-(state_offset + count), 0, ncount); | ADD_NEW_DATA(-(state_offset + (int)count), 0, ncount); |
| break; | break; |
| default: | default: |
| Line 1595 for (;;) | Line 1781 for (;;) |
| BOOL OK; | BOOL OK; |
| switch (c) | switch (c) |
| { | { |
| case 0x000a: | VSPACE_CASES: |
| case 0x000b: | |
| case 0x000c: | |
| case 0x000d: | |
| case 0x0085: | |
| case 0x2028: | |
| case 0x2029: | |
| OK = TRUE; | OK = TRUE; |
| break; | break; |
| Line 1617 for (;;) | Line 1797 for (;;) |
| active_count--; /* Remove non-match possibility */ | active_count--; /* Remove non-match possibility */ |
| next_active_state--; | next_active_state--; |
| } | } |
| ADD_NEW_DATA(-(state_offset + count), 0, 0); | ADD_NEW_DATA(-(state_offset + (int)count), 0, 0); |
| } | } |
| } | } |
| break; | break; |
| Line 1641 for (;;) | Line 1821 for (;;) |
| BOOL OK; | BOOL OK; |
| switch (c) | switch (c) |
| { | { |
| case 0x09: /* HT */ | HSPACE_CASES: |
| case 0x20: /* SPACE */ | |
| case 0xa0: /* NBSP */ | |
| case 0x1680: /* OGHAM SPACE MARK */ | |
| case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */ | |
| case 0x2000: /* EN QUAD */ | |
| case 0x2001: /* EM QUAD */ | |
| case 0x2002: /* EN SPACE */ | |
| case 0x2003: /* EM SPACE */ | |
| case 0x2004: /* THREE-PER-EM SPACE */ | |
| case 0x2005: /* FOUR-PER-EM SPACE */ | |
| case 0x2006: /* SIX-PER-EM SPACE */ | |
| case 0x2007: /* FIGURE SPACE */ | |
| case 0x2008: /* PUNCTUATION SPACE */ | |
| case 0x2009: /* THIN SPACE */ | |
| case 0x200A: /* HAIR SPACE */ | |
| case 0x202f: /* NARROW NO-BREAK SPACE */ | |
| case 0x205f: /* MEDIUM MATHEMATICAL SPACE */ | |
| case 0x3000: /* IDEOGRAPHIC SPACE */ | |
| OK = TRUE; | OK = TRUE; |
| break; | break; |
| Line 1676 for (;;) | Line 1838 for (;;) |
| active_count--; /* Remove non-match possibility */ | active_count--; /* Remove non-match possibility */ |
| next_active_state--; | next_active_state--; |
| } | } |
| ADD_NEW_DATA(-(state_offset + count), 0, 0); | ADD_NEW_DATA(-(state_offset + (int)count), 0, 0); |
| } | } |
| } | } |
| break; | break; |
| Line 1688 for (;;) | Line 1850 for (;;) |
| case OP_PROP_EXTRA + OP_TYPEMINUPTO: | case OP_PROP_EXTRA + OP_TYPEMINUPTO: |
| case OP_PROP_EXTRA + OP_TYPEPOSUPTO: | case OP_PROP_EXTRA + OP_TYPEPOSUPTO: |
| if (codevalue != OP_PROP_EXTRA + OP_TYPEEXACT) | if (codevalue != OP_PROP_EXTRA + OP_TYPEEXACT) |
| { ADD_ACTIVE(state_offset + 6, 0); } | { ADD_ACTIVE(state_offset + 1 + IMM2_SIZE + 3, 0); } |
| count = current_state->count; /* Number already matched */ | count = current_state->count; /* Number already matched */ |
| if (clen > 0) | if (clen > 0) |
| { | { |
| BOOL OK; | BOOL OK; |
| const pcre_uint32 *cp; | |
| const ucd_record * prop = GET_UCD(c); | const ucd_record * prop = GET_UCD(c); |
| switch(code[4]) | switch(code[1 + IMM2_SIZE + 1]) |
| { | { |
| case PT_ANY: | case PT_ANY: |
| OK = TRUE; | OK = TRUE; |
| Line 1706 for (;;) | Line 1869 for (;;) |
| break; | break; |
| case PT_GC: | case PT_GC: |
| OK = _pcre_ucp_gentype[prop->chartype] == code[5]; | OK = PRIV(ucp_gentype)[prop->chartype] == code[1 + IMM2_SIZE + 2]; |
| break; | break; |
| case PT_PC: | case PT_PC: |
| OK = prop->chartype == code[5]; | OK = prop->chartype == code[1 + IMM2_SIZE + 2]; |
| break; | break; |
| case PT_SC: | case PT_SC: |
| OK = prop->script == code[5]; | OK = prop->script == code[1 + IMM2_SIZE + 2]; |
| break; | break; |
| /* These are specials for combination cases. */ | /* These are specials for combination cases. */ |
| case PT_ALNUM: | case PT_ALNUM: |
| OK = _pcre_ucp_gentype[prop->chartype] == ucp_L || | OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L || |
| _pcre_ucp_gentype[prop->chartype] == ucp_N; | PRIV(ucp_gentype)[prop->chartype] == ucp_N; |
| break; | break; |
| case PT_SPACE: /* Perl space */ | /* Perl space used to exclude VT, but from Perl 5.18 it is included, |
| OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z || | which means that Perl space and POSIX space are now identical. PCRE |
| c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR; | was changed at release 8.34. */ |
| break; | |
| case PT_SPACE: /* Perl space */ | |
| case PT_PXSPACE: /* POSIX space */ | case PT_PXSPACE: /* POSIX space */ |
| OK = _pcre_ucp_gentype[prop->chartype] == ucp_Z || | switch(c) |
| c == CHAR_HT || c == CHAR_NL || c == CHAR_VT || | { |
| c == CHAR_FF || c == CHAR_CR; | HSPACE_CASES: |
| VSPACE_CASES: | |
| OK = TRUE; | |
| break; | |
| default: | |
| OK = PRIV(ucp_gentype)[prop->chartype] == ucp_Z; | |
| break; | |
| } | |
| break; | break; |
| case PT_WORD: | case PT_WORD: |
| OK = _pcre_ucp_gentype[prop->chartype] == ucp_L || | OK = PRIV(ucp_gentype)[prop->chartype] == ucp_L || |
| _pcre_ucp_gentype[prop->chartype] == ucp_N || | PRIV(ucp_gentype)[prop->chartype] == ucp_N || |
| c == CHAR_UNDERSCORE; | c == CHAR_UNDERSCORE; |
| break; | break; |
| case PT_CLIST: | |
| cp = PRIV(ucd_caseless_sets) + code[1 + IMM2_SIZE + 2]; | |
| for (;;) | |
| { | |
| if (c < *cp) { OK = FALSE; break; } | |
| if (c == *cp++) { OK = TRUE; break; } | |
| } | |
| break; | |
| case PT_UCNC: | |
| OK = c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT || | |
| c == CHAR_GRAVE_ACCENT || (c >= 0xa0 && c <= 0xd7ff) || | |
| c >= 0xe000; | |
| break; | |
| /* Should never occur, but keep compilers from grumbling. */ | /* Should never occur, but keep compilers from grumbling. */ |
| default: | default: |
| Line 1755 for (;;) | Line 1941 for (;;) |
| active_count--; /* Remove non-match possibility */ | active_count--; /* Remove non-match possibility */ |
| next_active_state--; | next_active_state--; |
| } | } |
| if (++count >= GET2(code, 1)) | if (++count >= (int)GET2(code, 1)) |
| { ADD_NEW(state_offset + 6, 0); } | { ADD_NEW(state_offset + 1 + IMM2_SIZE + 3, 0); } |
| else | else |
| { ADD_NEW(state_offset, count); } | { ADD_NEW(state_offset, count); } |
| } | } |
| Line 1769 for (;;) | Line 1955 for (;;) |
| case OP_EXTUNI_EXTRA + OP_TYPEMINUPTO: | case OP_EXTUNI_EXTRA + OP_TYPEMINUPTO: |
| case OP_EXTUNI_EXTRA + OP_TYPEPOSUPTO: | case OP_EXTUNI_EXTRA + OP_TYPEPOSUPTO: |
| if (codevalue != OP_EXTUNI_EXTRA + OP_TYPEEXACT) | if (codevalue != OP_EXTUNI_EXTRA + OP_TYPEEXACT) |
| { ADD_ACTIVE(state_offset + 4, 0); } | { ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0); } |
| count = current_state->count; /* Number already matched */ | count = current_state->count; /* Number already matched */ |
| if (clen > 0 && UCD_CATEGORY(c) != ucp_M) | if (clen > 0) |
| { | { |
| const uschar *nptr = ptr + clen; | int lgb, rgb; |
| const pcre_uchar *nptr = ptr + clen; | |
| int ncount = 0; | int ncount = 0; |
| if (codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSUPTO) | if (codevalue == OP_EXTUNI_EXTRA + OP_TYPEPOSUPTO) |
| { | { |
| active_count--; /* Remove non-match possibility */ | active_count--; /* Remove non-match possibility */ |
| next_active_state--; | next_active_state--; |
| } | } |
| lgb = UCD_GRAPHBREAK(c); | |
| while (nptr < end_subject) | while (nptr < end_subject) |
| { | { |
| int nd; | dlen = 1; |
| int ndlen = 1; | if (!utf) d = *nptr; else { GETCHARLEN(d, nptr, dlen); } |
| GETCHARLEN(nd, nptr, ndlen); | rgb = UCD_GRAPHBREAK(d); |
| if (UCD_CATEGORY(nd) != ucp_M) break; | if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break; |
| ncount++; | ncount++; |
| nptr += ndlen; | lgb = rgb; |
| nptr += dlen; | |
| } | } |
| if (++count >= GET2(code, 1)) | if (nptr >= end_subject && (md->moptions & PCRE_PARTIAL_HARD) != 0) |
| { ADD_NEW_DATA(-(state_offset + 4), 0, ncount); } | reset_could_continue = TRUE; |
| if (++count >= (int)GET2(code, 1)) | |
| { ADD_NEW_DATA(-(state_offset + 2 + IMM2_SIZE), 0, ncount); } | |
| else | else |
| { ADD_NEW_DATA(-state_offset, count, ncount); } | { ADD_NEW_DATA(-state_offset, count, ncount); } |
| } | } |
| Line 1803 for (;;) | Line 1994 for (;;) |
| case OP_ANYNL_EXTRA + OP_TYPEMINUPTO: | case OP_ANYNL_EXTRA + OP_TYPEMINUPTO: |
| case OP_ANYNL_EXTRA + OP_TYPEPOSUPTO: | case OP_ANYNL_EXTRA + OP_TYPEPOSUPTO: |
| if (codevalue != OP_ANYNL_EXTRA + OP_TYPEEXACT) | if (codevalue != OP_ANYNL_EXTRA + OP_TYPEEXACT) |
| { ADD_ACTIVE(state_offset + 4, 0); } | { ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0); } |
| count = current_state->count; /* Number already matched */ | count = current_state->count; /* Number already matched */ |
| if (clen > 0) | if (clen > 0) |
| { | { |
| int ncount = 0; | int ncount = 0; |
| switch (c) | switch (c) |
| { | { |
| case 0x000b: | case CHAR_VT: |
| case 0x000c: | case CHAR_FF: |
| case 0x0085: | case CHAR_NEL: |
| #ifndef EBCDIC | |
| case 0x2028: | case 0x2028: |
| case 0x2029: | case 0x2029: |
| #endif /* Not EBCDIC */ | |
| if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; | if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; |
| goto ANYNL03; | goto ANYNL03; |
| case 0x000d: | case CHAR_CR: |
| if (ptr + 1 < end_subject && ptr[1] == 0x0a) ncount = 1; | if (ptr + 1 < end_subject && UCHAR21TEST(ptr + 1) == CHAR_LF) ncount = 1; |
| /* Fall through */ | /* Fall through */ |
| ANYNL03: | ANYNL03: |
| case 0x000a: | case CHAR_LF: |
| if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSUPTO) | if (codevalue == OP_ANYNL_EXTRA + OP_TYPEPOSUPTO) |
| { | { |
| active_count--; /* Remove non-match possibility */ | active_count--; /* Remove non-match possibility */ |
| next_active_state--; | next_active_state--; |
| } | } |
| if (++count >= GET2(code, 1)) | if (++count >= (int)GET2(code, 1)) |
| { ADD_NEW_DATA(-(state_offset + 4), 0, ncount); } | { ADD_NEW_DATA(-(state_offset + 2 + IMM2_SIZE), 0, ncount); } |
| else | else |
| { ADD_NEW_DATA(-state_offset, count, ncount); } | { ADD_NEW_DATA(-state_offset, count, ncount); } |
| break; | break; |
| Line 1847 for (;;) | Line 2040 for (;;) |
| case OP_VSPACE_EXTRA + OP_TYPEMINUPTO: | case OP_VSPACE_EXTRA + OP_TYPEMINUPTO: |
| case OP_VSPACE_EXTRA + OP_TYPEPOSUPTO: | case OP_VSPACE_EXTRA + OP_TYPEPOSUPTO: |
| if (codevalue != OP_VSPACE_EXTRA + OP_TYPEEXACT) | if (codevalue != OP_VSPACE_EXTRA + OP_TYPEEXACT) |
| { ADD_ACTIVE(state_offset + 4, 0); } | { ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0); } |
| count = current_state->count; /* Number already matched */ | count = current_state->count; /* Number already matched */ |
| if (clen > 0) | if (clen > 0) |
| { | { |
| BOOL OK; | BOOL OK; |
| switch (c) | switch (c) |
| { | { |
| case 0x000a: | VSPACE_CASES: |
| case 0x000b: | |
| case 0x000c: | |
| case 0x000d: | |
| case 0x0085: | |
| case 0x2028: | |
| case 0x2029: | |
| OK = TRUE; | OK = TRUE; |
| break; | break; |
| Line 1875 for (;;) | Line 2062 for (;;) |
| active_count--; /* Remove non-match possibility */ | active_count--; /* Remove non-match possibility */ |
| next_active_state--; | next_active_state--; |
| } | } |
| if (++count >= GET2(code, 1)) | if (++count >= (int)GET2(code, 1)) |
| { ADD_NEW_DATA(-(state_offset + 4), 0, 0); } | { ADD_NEW_DATA(-(state_offset + 2 + IMM2_SIZE), 0, 0); } |
| else | else |
| { ADD_NEW_DATA(-state_offset, count, 0); } | { ADD_NEW_DATA(-state_offset, count, 0); } |
| } | } |
| Line 1889 for (;;) | Line 2076 for (;;) |
| case OP_HSPACE_EXTRA + OP_TYPEMINUPTO: | case OP_HSPACE_EXTRA + OP_TYPEMINUPTO: |
| case OP_HSPACE_EXTRA + OP_TYPEPOSUPTO: | case OP_HSPACE_EXTRA + OP_TYPEPOSUPTO: |
| if (codevalue != OP_HSPACE_EXTRA + OP_TYPEEXACT) | if (codevalue != OP_HSPACE_EXTRA + OP_TYPEEXACT) |
| { ADD_ACTIVE(state_offset + 4, 0); } | { ADD_ACTIVE(state_offset + 2 + IMM2_SIZE, 0); } |
| count = current_state->count; /* Number already matched */ | count = current_state->count; /* Number already matched */ |
| if (clen > 0) | if (clen > 0) |
| { | { |
| BOOL OK; | BOOL OK; |
| switch (c) | switch (c) |
| { | { |
| case 0x09: /* HT */ | HSPACE_CASES: |
| case 0x20: /* SPACE */ | |
| case 0xa0: /* NBSP */ | |
| case 0x1680: /* OGHAM SPACE MARK */ | |
| case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */ | |
| case 0x2000: /* EN QUAD */ | |
| case 0x2001: /* EM QUAD */ | |
| case 0x2002: /* EN SPACE */ | |
| case 0x2003: /* EM SPACE */ | |
| case 0x2004: /* THREE-PER-EM SPACE */ | |
| case 0x2005: /* FOUR-PER-EM SPACE */ | |
| case 0x2006: /* SIX-PER-EM SPACE */ | |
| case 0x2007: /* FIGURE SPACE */ | |
| case 0x2008: /* PUNCTUATION SPACE */ | |
| case 0x2009: /* THIN SPACE */ | |
| case 0x200A: /* HAIR SPACE */ | |
| case 0x202f: /* NARROW NO-BREAK SPACE */ | |
| case 0x205f: /* MEDIUM MATHEMATICAL SPACE */ | |
| case 0x3000: /* IDEOGRAPHIC SPACE */ | |
| OK = TRUE; | OK = TRUE; |
| break; | break; |
| Line 1930 for (;;) | Line 2099 for (;;) |
| active_count--; /* Remove non-match possibility */ | active_count--; /* Remove non-match possibility */ |
| next_active_state--; | next_active_state--; |
| } | } |
| if (++count >= GET2(code, 1)) | if (++count >= (int)GET2(code, 1)) |
| { ADD_NEW_DATA(-(state_offset + 4), 0, 0); } | { ADD_NEW_DATA(-(state_offset + 2 + IMM2_SIZE), 0, 0); } |
| else | else |
| { ADD_NEW_DATA(-state_offset, count, 0); } | { ADD_NEW_DATA(-state_offset, count, 0); } |
| } | } |
| Line 1950 for (;;) | Line 2119 for (;;) |
| break; | break; |
| /*-----------------------------------------------------------------*/ | /*-----------------------------------------------------------------*/ |
| case OP_CHARNC: | case OP_CHARI: |
| if (clen == 0) break; | if (clen == 0) break; |
| #ifdef SUPPORT_UTF8 | #ifdef SUPPORT_UTF |
| if (utf8) | if (utf) |
| { | { |
| if (c == d) { ADD_NEW(state_offset + dlen + 1, 0); } else | if (c == d) { ADD_NEW(state_offset + dlen + 1, 0); } else |
| { | { |
| unsigned int othercase; | unsigned int othercase; |
| if (c < 128) othercase = fcc[c]; else | if (c < 128) |
| othercase = fcc[c]; | |
| /* If we have Unicode property support, we can use it to test the | else |
| other case of the character. */ | /* If we have Unicode property support, we can use it to test the |
| other case of the character. */ | |
| #ifdef SUPPORT_UCP | #ifdef SUPPORT_UCP |
| othercase = UCD_OTHERCASE(c); | othercase = UCD_OTHERCASE(c); |
| #else | #else |
| othercase = NOTACHAR; | othercase = NOTACHAR; |
| #endif | #endif |
| if (d == othercase) { ADD_NEW(state_offset + dlen + 1, 0); } | if (d == othercase) { ADD_NEW(state_offset + dlen + 1, 0); } |
| } | } |
| } | } |
| else | else |
| #endif /* SUPPORT_UTF8 */ | #endif /* SUPPORT_UTF */ |
| /* Not UTF mode */ | |
| /* Non-UTF-8 mode */ | |
| { | { |
| if (lcc[c] == lcc[d]) { ADD_NEW(state_offset + 2, 0); } | if (TABLE_GET(c, lcc, c) == TABLE_GET(d, lcc, d)) |
| { ADD_NEW(state_offset + 2, 0); } | |
| } | } |
| break; | break; |
| Line 1990 for (;;) | Line 2159 for (;;) |
| to wait for them to pass before continuing. */ | to wait for them to pass before continuing. */ |
| case OP_EXTUNI: | case OP_EXTUNI: |
| if (clen > 0 && UCD_CATEGORY(c) != ucp_M) | if (clen > 0) |
| { | { |
| const uschar *nptr = ptr + clen; | int lgb, rgb; |
| const pcre_uchar *nptr = ptr + clen; | |
| int ncount = 0; | int ncount = 0; |
| lgb = UCD_GRAPHBREAK(c); | |
| while (nptr < end_subject) | while (nptr < end_subject) |
| { | { |
| int nclen = 1; | dlen = 1; |
| GETCHARLEN(c, nptr, nclen); | if (!utf) d = *nptr; else { GETCHARLEN(d, nptr, dlen); } |
| if (UCD_CATEGORY(c) != ucp_M) break; | rgb = UCD_GRAPHBREAK(d); |
| if ((PRIV(ucp_gbtable)[lgb] & (1 << rgb)) == 0) break; | |
| ncount++; | ncount++; |
| nptr += nclen; | lgb = rgb; |
| nptr += dlen; | |
| } | } |
| if (nptr >= end_subject && (md->moptions & PCRE_PARTIAL_HARD) != 0) | |
| reset_could_continue = TRUE; | |
| ADD_NEW_DATA(-(state_offset + 1), 0, ncount); | ADD_NEW_DATA(-(state_offset + 1), 0, ncount); |
| } | } |
| break; | break; |
| Line 2015 for (;;) | Line 2190 for (;;) |
| case OP_ANYNL: | case OP_ANYNL: |
| if (clen > 0) switch(c) | if (clen > 0) switch(c) |
| { | { |
| case 0x000b: | case CHAR_VT: |
| case 0x000c: | case CHAR_FF: |
| case 0x0085: | case CHAR_NEL: |
| #ifndef EBCDIC | |
| case 0x2028: | case 0x2028: |
| case 0x2029: | case 0x2029: |
| #endif /* Not EBCDIC */ | |
| if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; | if ((md->moptions & PCRE_BSR_ANYCRLF) != 0) break; |
| case 0x000a: | case CHAR_LF: |
| ADD_NEW(state_offset + 1, 0); | ADD_NEW(state_offset + 1, 0); |
| break; | break; |
| case 0x000d: | case CHAR_CR: |
| if (ptr + 1 < end_subject && ptr[1] == 0x0a) | if (ptr + 1 >= end_subject) |
| { | |
| ADD_NEW(state_offset + 1, 0); | |
| if ((md->moptions & PCRE_PARTIAL_HARD) != 0) | |
| reset_could_continue = TRUE; | |
| } | |
| else if (UCHAR21TEST(ptr + 1) == CHAR_LF) | |
| { | { |
| ADD_NEW_DATA(-(state_offset + 1), 0, 1); | ADD_NEW_DATA(-(state_offset + 1), 0, 1); |
| } | } |
| Line 2043 for (;;) | Line 2226 for (;;) |
| case OP_NOT_VSPACE: | case OP_NOT_VSPACE: |
| if (clen > 0) switch(c) | if (clen > 0) switch(c) |
| { | { |
| case 0x000a: | VSPACE_CASES: |
| case 0x000b: | |
| case 0x000c: | |
| case 0x000d: | |
| case 0x0085: | |
| case 0x2028: | |
| case 0x2029: | |
| break; | break; |
| default: | default: |
| Line 2062 for (;;) | Line 2239 for (;;) |
| case OP_VSPACE: | case OP_VSPACE: |
| if (clen > 0) switch(c) | if (clen > 0) switch(c) |
| { | { |
| case 0x000a: | VSPACE_CASES: |
| case 0x000b: | |
| case 0x000c: | |
| case 0x000d: | |
| case 0x0085: | |
| case 0x2028: | |
| case 0x2029: | |
| ADD_NEW(state_offset + 1, 0); | ADD_NEW(state_offset + 1, 0); |
| break; | break; |
| default: break; | default: |
| break; | |
| } | } |
| break; | break; |
| Line 2080 for (;;) | Line 2252 for (;;) |
| case OP_NOT_HSPACE: | case OP_NOT_HSPACE: |
| if (clen > 0) switch(c) | if (clen > 0) switch(c) |
| { | { |
| case 0x09: /* HT */ | HSPACE_CASES: |
| case 0x20: /* SPACE */ | |
| case 0xa0: /* NBSP */ | |
| case 0x1680: /* OGHAM SPACE MARK */ | |
| case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */ | |
| case 0x2000: /* EN QUAD */ | |
| case 0x2001: /* EM QUAD */ | |
| case 0x2002: /* EN SPACE */ | |
| case 0x2003: /* EM SPACE */ | |
| case 0x2004: /* THREE-PER-EM SPACE */ | |
| case 0x2005: /* FOUR-PER-EM SPACE */ | |
| case 0x2006: /* SIX-PER-EM SPACE */ | |
| case 0x2007: /* FIGURE SPACE */ | |
| case 0x2008: /* PUNCTUATION SPACE */ | |
| case 0x2009: /* THIN SPACE */ | |
| case 0x200A: /* HAIR SPACE */ | |
| case 0x202f: /* NARROW NO-BREAK SPACE */ | |
| case 0x205f: /* MEDIUM MATHEMATICAL SPACE */ | |
| case 0x3000: /* IDEOGRAPHIC SPACE */ | |
| break; | break; |
| default: | default: |
| Line 2111 for (;;) | Line 2265 for (;;) |
| case OP_HSPACE: | case OP_HSPACE: |
| if (clen > 0) switch(c) | if (clen > 0) switch(c) |
| { | { |
| case 0x09: /* HT */ | HSPACE_CASES: |
| case 0x20: /* SPACE */ | |
| case 0xa0: /* NBSP */ | |
| case 0x1680: /* OGHAM SPACE MARK */ | |
| case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */ | |
| case 0x2000: /* EN QUAD */ | |
| case 0x2001: /* EM QUAD */ | |
| case 0x2002: /* EN SPACE */ | |
| case 0x2003: /* EM SPACE */ | |
| case 0x2004: /* THREE-PER-EM SPACE */ | |
| case 0x2005: /* FOUR-PER-EM SPACE */ | |
| case 0x2006: /* SIX-PER-EM SPACE */ | |
| case 0x2007: /* FIGURE SPACE */ | |
| case 0x2008: /* PUNCTUATION SPACE */ | |
| case 0x2009: /* THIN SPACE */ | |
| case 0x200A: /* HAIR SPACE */ | |
| case 0x202f: /* NARROW NO-BREAK SPACE */ | |
| case 0x205f: /* MEDIUM MATHEMATICAL SPACE */ | |
| case 0x3000: /* IDEOGRAPHIC SPACE */ | |
| ADD_NEW(state_offset + 1, 0); | ADD_NEW(state_offset + 1, 0); |
| break; | break; |
| default: | |
| break; | |
| } | } |
| break; | break; |
| /*-----------------------------------------------------------------*/ | /*-----------------------------------------------------------------*/ |
| /* Match a negated single character. This is only used for one-byte | /* Match a negated single character casefully. */ |
| characters, that is, we know that d < 256. The character we are | |
| checking (c) can be multibyte. */ | |
| case OP_NOT: | case OP_NOT: |
| if (clen > 0 && c != d) { ADD_NEW(state_offset + dlen + 1, 0); } | |
| break; | |
| /*-----------------------------------------------------------------*/ | |
| /* Match a negated single character caselessly. */ | |
| case OP_NOTI: | |
| if (clen > 0) | if (clen > 0) |
| { | { |
| unsigned int otherd = ((ims & PCRE_CASELESS) != 0)? fcc[d] : d; | unsigned int otherd; |
| if (c != d && c != otherd) { ADD_NEW(state_offset + dlen + 1, 0); } | #ifdef SUPPORT_UTF |
| if (utf && d >= 128) | |
| { | |
| #ifdef SUPPORT_UCP | |
| otherd = UCD_OTHERCASE(d); | |
| #endif /* SUPPORT_UCP */ | |
| } | |
| else | |
| #endif /* SUPPORT_UTF */ | |
| otherd = TABLE_GET(d, fcc, d); | |
| if (c != d && c != otherd) | |
| { ADD_NEW(state_offset + dlen + 1, 0); } | |
| } | } |
| break; | break; |
| /*-----------------------------------------------------------------*/ | /*-----------------------------------------------------------------*/ |
| case OP_PLUSI: | |
| case OP_MINPLUSI: | |
| case OP_POSPLUSI: | |
| case OP_NOTPLUSI: | |
| case OP_NOTMINPLUSI: | |
| case OP_NOTPOSPLUSI: | |
| caseless = TRUE; | |
| codevalue -= OP_STARI - OP_STAR; | |
| /* Fall through */ | |
| case OP_PLUS: | case OP_PLUS: |
| case OP_MINPLUS: | case OP_MINPLUS: |
| case OP_POSPLUS: | case OP_POSPLUS: |
| Line 2159 for (;;) | Line 2324 for (;;) |
| if (count > 0) { ADD_ACTIVE(state_offset + dlen + 1, 0); } | if (count > 0) { ADD_ACTIVE(state_offset + dlen + 1, 0); } |
| if (clen > 0) | if (clen > 0) |
| { | { |
| unsigned int otherd = NOTACHAR; | pcre_uint32 otherd = NOTACHAR; |
| if ((ims & PCRE_CASELESS) != 0) | if (caseless) |
| { | { |
| #ifdef SUPPORT_UTF8 | #ifdef SUPPORT_UTF |
| if (utf8 && d >= 128) | if (utf && d >= 128) |
| { | { |
| #ifdef SUPPORT_UCP | #ifdef SUPPORT_UCP |
| otherd = UCD_OTHERCASE(d); | otherd = UCD_OTHERCASE(d); |
| #endif /* SUPPORT_UCP */ | #endif /* SUPPORT_UCP */ |
| } | } |
| else | else |
| #endif /* SUPPORT_UTF8 */ | #endif /* SUPPORT_UTF */ |
| otherd = fcc[d]; | otherd = TABLE_GET(d, fcc, d); |
| } | } |
| if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) | if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) |
| { | { |
| Line 2188 for (;;) | Line 2353 for (;;) |
| break; | break; |
| /*-----------------------------------------------------------------*/ | /*-----------------------------------------------------------------*/ |
| case OP_QUERYI: | |
| case OP_MINQUERYI: | |
| case OP_POSQUERYI: | |
| case OP_NOTQUERYI: | |
| case OP_NOTMINQUERYI: | |
| case OP_NOTPOSQUERYI: | |
| caseless = TRUE; | |
| codevalue -= OP_STARI - OP_STAR; | |
| /* Fall through */ | |
| case OP_QUERY: | case OP_QUERY: |
| case OP_MINQUERY: | case OP_MINQUERY: |
| case OP_POSQUERY: | case OP_POSQUERY: |
| Line 2197 for (;;) | Line 2371 for (;;) |
| ADD_ACTIVE(state_offset + dlen + 1, 0); | ADD_ACTIVE(state_offset + dlen + 1, 0); |
| if (clen > 0) | if (clen > 0) |
| { | { |
| unsigned int otherd = NOTACHAR; | pcre_uint32 otherd = NOTACHAR; |
| if ((ims & PCRE_CASELESS) != 0) | if (caseless) |
| { | { |
| #ifdef SUPPORT_UTF8 | #ifdef SUPPORT_UTF |
| if (utf8 && d >= 128) | if (utf && d >= 128) |
| { | { |
| #ifdef SUPPORT_UCP | #ifdef SUPPORT_UCP |
| otherd = UCD_OTHERCASE(d); | otherd = UCD_OTHERCASE(d); |
| #endif /* SUPPORT_UCP */ | #endif /* SUPPORT_UCP */ |
| } | } |
| else | else |
| #endif /* SUPPORT_UTF8 */ | #endif /* SUPPORT_UTF */ |
| otherd = fcc[d]; | otherd = TABLE_GET(d, fcc, d); |
| } | } |
| if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) | if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) |
| { | { |
| Line 2224 for (;;) | Line 2398 for (;;) |
| break; | break; |
| /*-----------------------------------------------------------------*/ | /*-----------------------------------------------------------------*/ |
| case OP_STARI: | |
| case OP_MINSTARI: | |
| case OP_POSSTARI: | |
| case OP_NOTSTARI: | |
| case OP_NOTMINSTARI: | |
| case OP_NOTPOSSTARI: | |
| caseless = TRUE; | |
| codevalue -= OP_STARI - OP_STAR; | |
| /* Fall through */ | |
| case OP_STAR: | case OP_STAR: |
| case OP_MINSTAR: | case OP_MINSTAR: |
| case OP_POSSTAR: | case OP_POSSTAR: |
| Line 2233 for (;;) | Line 2416 for (;;) |
| ADD_ACTIVE(state_offset + dlen + 1, 0); | ADD_ACTIVE(state_offset + dlen + 1, 0); |
| if (clen > 0) | if (clen > 0) |
| { | { |
| unsigned int otherd = NOTACHAR; | pcre_uint32 otherd = NOTACHAR; |
| if ((ims & PCRE_CASELESS) != 0) | if (caseless) |
| { | { |
| #ifdef SUPPORT_UTF8 | #ifdef SUPPORT_UTF |
| if (utf8 && d >= 128) | if (utf && d >= 128) |
| { | { |
| #ifdef SUPPORT_UCP | #ifdef SUPPORT_UCP |
| otherd = UCD_OTHERCASE(d); | otherd = UCD_OTHERCASE(d); |
| #endif /* SUPPORT_UCP */ | #endif /* SUPPORT_UCP */ |
| } | } |
| else | else |
| #endif /* SUPPORT_UTF8 */ | #endif /* SUPPORT_UTF */ |
| otherd = fcc[d]; | otherd = TABLE_GET(d, fcc, d); |
| } | } |
| if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) | if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) |
| { | { |
| Line 2260 for (;;) | Line 2443 for (;;) |
| break; | break; |
| /*-----------------------------------------------------------------*/ | /*-----------------------------------------------------------------*/ |
| case OP_EXACTI: | |
| case OP_NOTEXACTI: | |
| caseless = TRUE; | |
| codevalue -= OP_STARI - OP_STAR; | |
| /* Fall through */ | |
| case OP_EXACT: | case OP_EXACT: |
| case OP_NOTEXACT: | case OP_NOTEXACT: |
| count = current_state->count; /* Number already matched */ | count = current_state->count; /* Number already matched */ |
| if (clen > 0) | if (clen > 0) |
| { | { |
| unsigned int otherd = NOTACHAR; | pcre_uint32 otherd = NOTACHAR; |
| if ((ims & PCRE_CASELESS) != 0) | if (caseless) |
| { | { |
| #ifdef SUPPORT_UTF8 | #ifdef SUPPORT_UTF |
| if (utf8 && d >= 128) | if (utf && d >= 128) |
| { | { |
| #ifdef SUPPORT_UCP | #ifdef SUPPORT_UCP |
| otherd = UCD_OTHERCASE(d); | otherd = UCD_OTHERCASE(d); |
| #endif /* SUPPORT_UCP */ | #endif /* SUPPORT_UCP */ |
| } | } |
| else | else |
| #endif /* SUPPORT_UTF8 */ | #endif /* SUPPORT_UTF */ |
| otherd = fcc[d]; | otherd = TABLE_GET(d, fcc, d); |
| } | } |
| if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) | if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) |
| { | { |
| if (++count >= GET2(code, 1)) | if (++count >= (int)GET2(code, 1)) |
| { ADD_NEW(state_offset + dlen + 3, 0); } | { ADD_NEW(state_offset + dlen + 1 + IMM2_SIZE, 0); } |
| else | else |
| { ADD_NEW(state_offset, count); } | { ADD_NEW(state_offset, count); } |
| } | } |
| Line 2290 for (;;) | Line 2478 for (;;) |
| break; | break; |
| /*-----------------------------------------------------------------*/ | /*-----------------------------------------------------------------*/ |
| case OP_UPTOI: | |
| case OP_MINUPTOI: | |
| case OP_POSUPTOI: | |
| case OP_NOTUPTOI: | |
| case OP_NOTMINUPTOI: | |
| case OP_NOTPOSUPTOI: | |
| caseless = TRUE; | |
| codevalue -= OP_STARI - OP_STAR; | |
| /* Fall through */ | |
| case OP_UPTO: | case OP_UPTO: |
| case OP_MINUPTO: | case OP_MINUPTO: |
| case OP_POSUPTO: | case OP_POSUPTO: |
| case OP_NOTUPTO: | case OP_NOTUPTO: |
| case OP_NOTMINUPTO: | case OP_NOTMINUPTO: |
| case OP_NOTPOSUPTO: | case OP_NOTPOSUPTO: |
| ADD_ACTIVE(state_offset + dlen + 3, 0); | ADD_ACTIVE(state_offset + dlen + 1 + IMM2_SIZE, 0); |
| count = current_state->count; /* Number already matched */ | count = current_state->count; /* Number already matched */ |
| if (clen > 0) | if (clen > 0) |
| { | { |
| unsigned int otherd = NOTACHAR; | pcre_uint32 otherd = NOTACHAR; |
| if ((ims & PCRE_CASELESS) != 0) | if (caseless) |
| { | { |
| #ifdef SUPPORT_UTF8 | #ifdef SUPPORT_UTF |
| if (utf8 && d >= 128) | if (utf && d >= 128) |
| { | { |
| #ifdef SUPPORT_UCP | #ifdef SUPPORT_UCP |
| otherd = UCD_OTHERCASE(d); | otherd = UCD_OTHERCASE(d); |
| #endif /* SUPPORT_UCP */ | #endif /* SUPPORT_UCP */ |
| } | } |
| else | else |
| #endif /* SUPPORT_UTF8 */ | #endif /* SUPPORT_UTF */ |
| otherd = fcc[d]; | otherd = TABLE_GET(d, fcc, d); |
| } | } |
| if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) | if ((c == d || c == otherd) == (codevalue < OP_NOTSTAR)) |
| { | { |
| Line 2321 for (;;) | Line 2518 for (;;) |
| active_count--; /* Remove non-match possibility */ | active_count--; /* Remove non-match possibility */ |
| next_active_state--; | next_active_state--; |
| } | } |
| if (++count >= GET2(code, 1)) | if (++count >= (int)GET2(code, 1)) |
| { ADD_NEW(state_offset + dlen + 3, 0); } | { ADD_NEW(state_offset + dlen + 1 + IMM2_SIZE, 0); } |
| else | else |
| { ADD_NEW(state_offset, count); } | { ADD_NEW(state_offset, count); } |
| } | } |
| Line 2339 for (;;) | Line 2536 for (;;) |
| { | { |
| BOOL isinclass = FALSE; | BOOL isinclass = FALSE; |
| int next_state_offset; | int next_state_offset; |
| const uschar *ecode; | const pcre_uchar *ecode; |
| /* For a simple class, there is always just a 32-byte table, and we | /* For a simple class, there is always just a 32-byte table, and we |
| can set isinclass from it. */ | can set isinclass from it. */ |
| if (codevalue != OP_XCLASS) | if (codevalue != OP_XCLASS) |
| { | { |
| ecode = code + 33; | ecode = code + 1 + (32 / sizeof(pcre_uchar)); |
| if (clen > 0) | if (clen > 0) |
| { | { |
| isinclass = (c > 255)? (codevalue == OP_NCLASS) : | isinclass = (c > 255)? (codevalue == OP_NCLASS) : |
| ((code[1 + c/8] & (1 << (c&7))) != 0); | ((((pcre_uint8 *)(code + 1))[c/8] & (1 << (c&7))) != 0); |
| } | } |
| } | } |
| Line 2361 for (;;) | Line 2558 for (;;) |
| else | else |
| { | { |
| ecode = code + GET(code, 1); | ecode = code + GET(code, 1); |
| if (clen > 0) isinclass = _pcre_xclass(c, code + 1 + LINK_SIZE); | if (clen > 0) isinclass = PRIV(xclass)(c, code + 1 + LINK_SIZE, utf); |
| } | } |
| /* At this point, isinclass is set for all kinds of class, and ecode | /* At this point, isinclass is set for all kinds of class, and ecode |
| Line 2374 for (;;) | Line 2571 for (;;) |
| { | { |
| case OP_CRSTAR: | case OP_CRSTAR: |
| case OP_CRMINSTAR: | case OP_CRMINSTAR: |
| case OP_CRPOSSTAR: | |
| ADD_ACTIVE(next_state_offset + 1, 0); | ADD_ACTIVE(next_state_offset + 1, 0); |
| if (isinclass) { ADD_NEW(state_offset, 0); } | if (isinclass) |
| { | |
| if (*ecode == OP_CRPOSSTAR) | |
| { | |
| active_count--; /* Remove non-match possibility */ | |
| next_active_state--; | |
| } | |
| ADD_NEW(state_offset, 0); | |
| } | |
| break; | break; |
| case OP_CRPLUS: | case OP_CRPLUS: |
| case OP_CRMINPLUS: | case OP_CRMINPLUS: |
| case OP_CRPOSPLUS: | |
| count = current_state->count; /* Already matched */ | count = current_state->count; /* Already matched */ |
| if (count > 0) { ADD_ACTIVE(next_state_offset + 1, 0); } | if (count > 0) { ADD_ACTIVE(next_state_offset + 1, 0); } |
| if (isinclass) { count++; ADD_NEW(state_offset, count); } | if (isinclass) |
| { | |
| if (count > 0 && *ecode == OP_CRPOSPLUS) | |
| { | |
| active_count--; /* Remove non-match possibility */ | |
| next_active_state--; | |
| } | |
| count++; | |
| ADD_NEW(state_offset, count); | |
| } | |
| break; | break; |
| case OP_CRQUERY: | case OP_CRQUERY: |
| case OP_CRMINQUERY: | case OP_CRMINQUERY: |
| case OP_CRPOSQUERY: | |
| ADD_ACTIVE(next_state_offset + 1, 0); | ADD_ACTIVE(next_state_offset + 1, 0); |
| if (isinclass) { ADD_NEW(next_state_offset + 1, 0); } | if (isinclass) |
| { | |
| if (*ecode == OP_CRPOSQUERY) | |
| { | |
| active_count--; /* Remove non-match possibility */ | |
| next_active_state--; | |
| } | |
| ADD_NEW(next_state_offset + 1, 0); | |
| } | |
| break; | break; |
| case OP_CRRANGE: | case OP_CRRANGE: |
| case OP_CRMINRANGE: | case OP_CRMINRANGE: |
| case OP_CRPOSRANGE: | |
| count = current_state->count; /* Already matched */ | count = current_state->count; /* Already matched */ |
| if (count >= GET2(ecode, 1)) | if (count >= (int)GET2(ecode, 1)) |
| { ADD_ACTIVE(next_state_offset + 5, 0); } | { ADD_ACTIVE(next_state_offset + 1 + 2 * IMM2_SIZE, 0); } |
| if (isinclass) | if (isinclass) |
| { | { |
| int max = GET2(ecode, 3); | int max = (int)GET2(ecode, 1 + IMM2_SIZE); |
| if (*ecode == OP_CRPOSRANGE) | |
| { | |
| active_count--; /* Remove non-match possibility */ | |
| next_active_state--; | |
| } | |
| if (++count >= max && max != 0) /* Max 0 => no limit */ | if (++count >= max && max != 0) /* Max 0 => no limit */ |
| { ADD_NEW(next_state_offset + 5, 0); } | { ADD_NEW(next_state_offset + 1 + 2 * IMM2_SIZE, 0); } |
| else | else |
| { ADD_NEW(state_offset, count); } | { ADD_NEW(state_offset, count); } |
| } | } |
| Line 2431 for (;;) | Line 2662 for (;;) |
| int rc; | int rc; |
| int local_offsets[2]; | int local_offsets[2]; |
| int local_workspace[1000]; | int local_workspace[1000]; |
| const uschar *endasscode = code + GET(code, 1); | const pcre_uchar *endasscode = code + GET(code, 1); |
| while (*endasscode == OP_ALT) endasscode += GET(endasscode, 1); | while (*endasscode == OP_ALT) endasscode += GET(endasscode, 1); |
| Line 2444 for (;;) | Line 2675 for (;;) |
| sizeof(local_offsets)/sizeof(int), /* size of same */ | sizeof(local_offsets)/sizeof(int), /* size of same */ |
| local_workspace, /* workspace vector */ | local_workspace, /* workspace vector */ |
| sizeof(local_workspace)/sizeof(int), /* size of same */ | sizeof(local_workspace)/sizeof(int), /* size of same */ |
| ims, /* the current ims flags */ | rlevel); /* function recursion level */ |
| rlevel, /* function recursion level */ | |
| recursing); /* pass on regex recursion */ | |
| if (rc == PCRE_ERROR_DFA_UITEM) return rc; | if (rc == PCRE_ERROR_DFA_UITEM) return rc; |
| if ((rc >= 0) == (codevalue == OP_ASSERT || codevalue == OP_ASSERTBACK)) | if ((rc >= 0) == (codevalue == OP_ASSERT || codevalue == OP_ASSERTBACK)) |
| Line 2470 for (;;) | Line 2699 for (;;) |
| if (code[LINK_SIZE+1] == OP_CALLOUT) | if (code[LINK_SIZE+1] == OP_CALLOUT) |
| { | { |
| rrc = 0; | rrc = 0; |
| if (pcre_callout != NULL) | if (PUBL(callout) != NULL) |
| { | { |
| pcre_callout_block cb; | PUBL(callout_block) cb; |
| cb.version = 1; /* Version 1 of the callout block */ | cb.version = 1; /* Version 1 of the callout block */ |
| cb.callout_number = code[LINK_SIZE+2]; | cb.callout_number = code[LINK_SIZE+2]; |
| cb.offset_vector = offsets; | cb.offset_vector = offsets; |
| #if defined COMPILE_PCRE8 | |
| cb.subject = (PCRE_SPTR)start_subject; | cb.subject = (PCRE_SPTR)start_subject; |
| #elif defined COMPILE_PCRE16 | |
| cb.subject = (PCRE_SPTR16)start_subject; | |
| #elif defined COMPILE_PCRE32 | |
| cb.subject = (PCRE_SPTR32)start_subject; | |
| #endif | |
| cb.subject_length = (int)(end_subject - start_subject); | cb.subject_length = (int)(end_subject - start_subject); |
| cb.start_match = (int)(current_subject - start_subject); | cb.start_match = (int)(current_subject - start_subject); |
| cb.current_position = (int)(ptr - start_subject); | cb.current_position = (int)(ptr - start_subject); |
| Line 2485 for (;;) | Line 2720 for (;;) |
| cb.capture_top = 1; | cb.capture_top = 1; |
| cb.capture_last = -1; | cb.capture_last = -1; |
| cb.callout_data = md->callout_data; | cb.callout_data = md->callout_data; |
| if ((rrc = (*pcre_callout)(&cb)) < 0) return rrc; /* Abandon */ | cb.mark = NULL; /* No (*MARK) support */ |
| if ((rrc = (*PUBL(callout))(&cb)) < 0) return rrc; /* Abandon */ | |
| } | } |
| if (rrc > 0) break; /* Fail this thread */ | if (rrc > 0) break; /* Fail this thread */ |
| code += _pcre_OP_lengths[OP_CALLOUT]; /* Skip callout data */ | code += PRIV(OP_lengths)[OP_CALLOUT]; /* Skip callout data */ |
| } | } |
| condcode = code[LINK_SIZE+1]; | condcode = code[LINK_SIZE+1]; |
| /* Back reference conditions are not supported */ | /* Back reference conditions and duplicate named recursion conditions |
| are not supported */ | |
| if (condcode == OP_CREF || condcode == OP_NCREF) | if (condcode == OP_CREF || condcode == OP_DNCREF || |
| condcode == OP_DNRREF) | |
| return PCRE_ERROR_DFA_UCOND; | return PCRE_ERROR_DFA_UCOND; |
| /* The DEFINE condition is always false */ | /* The DEFINE condition is always false, and the assertion (?!) is |
| converted to OP_FAIL. */ | |
| if (condcode == OP_DEF) | if (condcode == OP_DEF || condcode == OP_FAIL) |
| { ADD_ACTIVE(state_offset + codelink + LINK_SIZE + 1, 0); } | { ADD_ACTIVE(state_offset + codelink + LINK_SIZE + 1, 0); } |
| /* The only supported version of OP_RREF is for the value RREF_ANY, | /* The only supported version of OP_RREF is for the value RREF_ANY, |
| which means "test if in any recursion". We can't test for specifically | which means "test if in any recursion". We can't test for specifically |
| recursed groups. */ | recursed groups. */ |
| else if (condcode == OP_RREF || condcode == OP_NRREF) | else if (condcode == OP_RREF) |
| { | { |
| int value = GET2(code, LINK_SIZE+2); | int value = GET2(code, LINK_SIZE + 2); |
| if (value != RREF_ANY) return PCRE_ERROR_DFA_UCOND; | if (value != RREF_ANY) return PCRE_ERROR_DFA_UCOND; |
| if (recursing > 0) | if (md->recursive != NULL) |
| { ADD_ACTIVE(state_offset + LINK_SIZE + 4, 0); } | { ADD_ACTIVE(state_offset + LINK_SIZE + 2 + IMM2_SIZE, 0); } |
| else { ADD_ACTIVE(state_offset + codelink + LINK_SIZE + 1, 0); } | else { ADD_ACTIVE(state_offset + codelink + LINK_SIZE + 1, 0); } |
| } | } |
| Line 2521 for (;;) | Line 2760 for (;;) |
| else | else |
| { | { |
| int rc; | int rc; |
| const uschar *asscode = code + LINK_SIZE + 1; | const pcre_uchar *asscode = code + LINK_SIZE + 1; |
| const uschar *endasscode = asscode + GET(asscode, 1); | const pcre_uchar *endasscode = asscode + GET(asscode, 1); |
| while (*endasscode == OP_ALT) endasscode += GET(endasscode, 1); | while (*endasscode == OP_ALT) endasscode += GET(endasscode, 1); |
| Line 2535 for (;;) | Line 2774 for (;;) |
| sizeof(local_offsets)/sizeof(int), /* size of same */ | sizeof(local_offsets)/sizeof(int), /* size of same */ |
| local_workspace, /* workspace vector */ | local_workspace, /* workspace vector */ |
| sizeof(local_workspace)/sizeof(int), /* size of same */ | sizeof(local_workspace)/sizeof(int), /* size of same */ |
| ims, /* the current ims flags */ | rlevel); /* function recursion level */ |
| rlevel, /* function recursion level */ | |
| recursing); /* pass on regex recursion */ | |
| if (rc == PCRE_ERROR_DFA_UITEM) return rc; | if (rc == PCRE_ERROR_DFA_UITEM) return rc; |
| if ((rc >= 0) == | if ((rc >= 0) == |
| Line 2552 for (;;) | Line 2789 for (;;) |
| /*-----------------------------------------------------------------*/ | /*-----------------------------------------------------------------*/ |
| case OP_RECURSE: | case OP_RECURSE: |
| { | { |
| dfa_recursion_info *ri; | |
| int local_offsets[1000]; | int local_offsets[1000]; |
| int local_workspace[1000]; | int local_workspace[1000]; |
| const pcre_uchar *callpat = start_code + GET(code, 1); | |
| int recno = (callpat == md->start_code)? 0 : | |
| GET2(callpat, 1 + LINK_SIZE); | |
| int rc; | int rc; |
| DPRINTF(("%.*sStarting regex recursion %d\n", rlevel*2-2, SP, | DPRINTF(("%.*sStarting regex recursion\n", rlevel*2-2, SP)); |
| recursing + 1)); | |
| /* Check for repeating a recursion without advancing the subject | |
| pointer. This should catch convoluted mutual recursions. (Some simple | |
| cases are caught at compile time.) */ | |
| for (ri = md->recursive; ri != NULL; ri = ri->prevrec) | |
| if (recno == ri->group_num && ptr == ri->subject_position) | |
| return PCRE_ERROR_RECURSELOOP; | |
| /* Remember this recursion and where we started it so as to | |
| catch infinite loops. */ | |
| new_recursive.group_num = recno; | |
| new_recursive.subject_position = ptr; | |
| new_recursive.prevrec = md->recursive; | |
| md->recursive = &new_recursive; | |
| rc = internal_dfa_exec( | rc = internal_dfa_exec( |
| md, /* fixed match data */ | md, /* fixed match data */ |
| start_code + GET(code, 1), /* this subexpression's code */ | callpat, /* this subexpression's code */ |
| ptr, /* where we currently are */ | ptr, /* where we currently are */ |
| (int)(ptr - start_subject), /* start offset */ | (int)(ptr - start_subject), /* start offset */ |
| local_offsets, /* offset vector */ | local_offsets, /* offset vector */ |
| sizeof(local_offsets)/sizeof(int), /* size of same */ | sizeof(local_offsets)/sizeof(int), /* size of same */ |
| local_workspace, /* workspace vector */ | local_workspace, /* workspace vector */ |
| sizeof(local_workspace)/sizeof(int), /* size of same */ | sizeof(local_workspace)/sizeof(int), /* size of same */ |
| ims, /* the current ims flags */ | rlevel); /* function recursion level */ |
| rlevel, /* function recursion level */ | |
| recursing + 1); /* regex recurse level */ | md->recursive = new_recursive.prevrec; /* Done this recursion */ |
| DPRINTF(("%.*sReturn from regex recursion %d: rc=%d\n", rlevel*2-2, SP, | DPRINTF(("%.*sReturn from regex recursion: rc=%d\n", rlevel*2-2, SP, |
| recursing + 1, rc)); | rc)); |
| /* Ran out of internal offsets */ | /* Ran out of internal offsets */ |
| Line 2587 for (;;) | Line 2843 for (;;) |
| { | { |
| for (rc = rc*2 - 2; rc >= 0; rc -= 2) | for (rc = rc*2 - 2; rc >= 0; rc -= 2) |
| { | { |
| const uschar *p = start_subject + local_offsets[rc]; | |
| const uschar *pp = start_subject + local_offsets[rc+1]; | |
| int charcount = local_offsets[rc+1] - local_offsets[rc]; | int charcount = local_offsets[rc+1] - local_offsets[rc]; |
| while (p < pp) if ((*p++ & 0xc0) == 0x80) charcount--; | #if defined SUPPORT_UTF && !defined COMPILE_PCRE32 |
| if (utf) | |
| { | |
| const pcre_uchar *p = start_subject + local_offsets[rc]; | |
| const pcre_uchar *pp = start_subject + local_offsets[rc+1]; | |
| while (p < pp) if (NOT_FIRSTCHAR(*p++)) charcount--; | |
| } | |
| #endif | |
| if (charcount > 0) | if (charcount > 0) |
| { | { |
| ADD_NEW_DATA(-(state_offset + LINK_SIZE + 1), 0, (charcount - 1)); | ADD_NEW_DATA(-(state_offset + LINK_SIZE + 1), 0, (charcount - 1)); |
| Line 2606 for (;;) | Line 2867 for (;;) |
| break; | break; |
| /*-----------------------------------------------------------------*/ | /*-----------------------------------------------------------------*/ |
| case OP_BRAPOS: | |
| case OP_SBRAPOS: | |
| case OP_CBRAPOS: | |
| case OP_SCBRAPOS: | |
| case OP_BRAPOSZERO: | |
| { | |
| int charcount, matched_count; | |
| const pcre_uchar *local_ptr = ptr; | |
| BOOL allow_zero; | |
| if (codevalue == OP_BRAPOSZERO) | |
| { | |
| allow_zero = TRUE; | |
| codevalue = *(++code); /* Codevalue will be one of above BRAs */ | |
| } | |
| else allow_zero = FALSE; | |
| /* Loop to match the subpattern as many times as possible as if it were | |
| a complete pattern. */ | |
| for (matched_count = 0;; matched_count++) | |
| { | |
| int local_offsets[2]; | |
| int local_workspace[1000]; | |
| int rc = internal_dfa_exec( | |
| md, /* fixed match data */ | |
| code, /* this subexpression's code */ | |
| local_ptr, /* where we currently are */ | |
| (int)(ptr - start_subject), /* start offset */ | |
| local_offsets, /* offset vector */ | |
| sizeof(local_offsets)/sizeof(int), /* size of same */ | |
| local_workspace, /* workspace vector */ | |
| sizeof(local_workspace)/sizeof(int), /* size of same */ | |
| rlevel); /* function recursion level */ | |
| /* Failed to match */ | |
| if (rc < 0) | |
| { | |
| if (rc != PCRE_ERROR_NOMATCH) return rc; | |
| break; | |
| } | |
| /* Matched: break the loop if zero characters matched. */ | |
| charcount = local_offsets[1] - local_offsets[0]; | |
| if (charcount == 0) break; | |
| local_ptr += charcount; /* Advance temporary position ptr */ | |
| } | |
| /* At this point we have matched the subpattern matched_count | |
| times, and local_ptr is pointing to the character after the end of the | |
| last match. */ | |
| if (matched_count > 0 || allow_zero) | |
| { | |
| const pcre_uchar *end_subpattern = code; | |
| int next_state_offset; | |
| do { end_subpattern += GET(end_subpattern, 1); } | |
| while (*end_subpattern == OP_ALT); | |
| next_state_offset = | |
| (int)(end_subpattern - start_code + LINK_SIZE + 1); | |
| /* Optimization: if there are no more active states, and there | |
| are no new states yet set up, then skip over the subject string | |
| right here, to save looping. Otherwise, set up the new state to swing | |
| into action when the end of the matched substring is reached. */ | |
| if (i + 1 >= active_count && new_count == 0) | |
| { | |
| ptr = local_ptr; | |
| clen = 0; | |
| ADD_NEW(next_state_offset, 0); | |
| } | |
| else | |
| { | |
| const pcre_uchar *p = ptr; | |
| const pcre_uchar *pp = local_ptr; | |
| charcount = (int)(pp - p); | |
| #if defined SUPPORT_UTF && !defined COMPILE_PCRE32 | |
| if (utf) while (p < pp) if (NOT_FIRSTCHAR(*p++)) charcount--; | |
| #endif | |
| ADD_NEW_DATA(-next_state_offset, 0, (charcount - 1)); | |
| } | |
| } | |
| } | |
| break; | |
| /*-----------------------------------------------------------------*/ | |
| case OP_ONCE: | case OP_ONCE: |
| case OP_ONCE_NC: | |
| { | { |
| int local_offsets[2]; | int local_offsets[2]; |
| int local_workspace[1000]; | int local_workspace[1000]; |
| Line 2620 for (;;) | Line 2973 for (;;) |
| sizeof(local_offsets)/sizeof(int), /* size of same */ | sizeof(local_offsets)/sizeof(int), /* size of same */ |
| local_workspace, /* workspace vector */ | local_workspace, /* workspace vector */ |
| sizeof(local_workspace)/sizeof(int), /* size of same */ | sizeof(local_workspace)/sizeof(int), /* size of same */ |
| ims, /* the current ims flags */ | rlevel); /* function recursion level */ |
| rlevel, /* function recursion level */ | |
| recursing); /* pass on regex recursion */ | |
| if (rc >= 0) | if (rc >= 0) |
| { | { |
| const uschar *end_subpattern = code; | const pcre_uchar *end_subpattern = code; |
| int charcount = local_offsets[1] - local_offsets[0]; | int charcount = local_offsets[1] - local_offsets[0]; |
| int next_state_offset, repeat_state_offset; | int next_state_offset, repeat_state_offset; |
| Line 2656 for (;;) | Line 3007 for (;;) |
| /* Optimization: if there are no more active states, and there | /* Optimization: if there are no more active states, and there |
| are no new states yet set up, then skip over the subject string | are no new states yet set up, then skip over the subject string |
| right here, to save looping. Otherwise, set up the new state to swing | right here, to save looping. Otherwise, set up the new state to swing |
| into action when the end of the substring is reached. */ | into action when the end of the matched substring is reached. */ |
| else if (i + 1 >= active_count && new_count == 0) | else if (i + 1 >= active_count && new_count == 0) |
| { | { |
| Line 2679 for (;;) | Line 3030 for (;;) |
| } | } |
| else | else |
| { | { |
| const uschar *p = start_subject + local_offsets[0]; | #if defined SUPPORT_UTF && !defined COMPILE_PCRE32 |
| const uschar *pp = start_subject + local_offsets[1]; | if (utf) |
| while (p < pp) if ((*p++ & 0xc0) == 0x80) charcount--; | { |
| const pcre_uchar *p = start_subject + local_offsets[0]; | |
| const pcre_uchar *pp = start_subject + local_offsets[1]; | |
| while (p < pp) if (NOT_FIRSTCHAR(*p++)) charcount--; | |
| } | |
| #endif | |
| ADD_NEW_DATA(-next_state_offset, 0, (charcount - 1)); | ADD_NEW_DATA(-next_state_offset, 0, (charcount - 1)); |
| if (repeat_state_offset >= 0) | if (repeat_state_offset >= 0) |
| { ADD_NEW_DATA(-repeat_state_offset, 0, (charcount - 1)); } | { ADD_NEW_DATA(-repeat_state_offset, 0, (charcount - 1)); } |
| } | } |
| } | } |
| else if (rc != PCRE_ERROR_NOMATCH) return rc; | else if (rc != PCRE_ERROR_NOMATCH) return rc; |
| } | } |
| Line 2698 for (;;) | Line 3053 for (;;) |
| case OP_CALLOUT: | case OP_CALLOUT: |
| rrc = 0; | rrc = 0; |
| if (pcre_callout != NULL) | if (PUBL(callout) != NULL) |
| { | { |
| pcre_callout_block cb; | PUBL(callout_block) cb; |
| cb.version = 1; /* Version 1 of the callout block */ | cb.version = 1; /* Version 1 of the callout block */ |
| cb.callout_number = code[1]; | cb.callout_number = code[1]; |
| cb.offset_vector = offsets; | cb.offset_vector = offsets; |
| #if defined COMPILE_PCRE8 | |
| cb.subject = (PCRE_SPTR)start_subject; | cb.subject = (PCRE_SPTR)start_subject; |
| #elif defined COMPILE_PCRE16 | |
| cb.subject = (PCRE_SPTR16)start_subject; | |
| #elif defined COMPILE_PCRE32 | |
| cb.subject = (PCRE_SPTR32)start_subject; | |
| #endif | |
| cb.subject_length = (int)(end_subject - start_subject); | cb.subject_length = (int)(end_subject - start_subject); |
| cb.start_match = (int)(current_subject - start_subject); | cb.start_match = (int)(current_subject - start_subject); |
| cb.current_position = (int)(ptr - start_subject); | cb.current_position = (int)(ptr - start_subject); |
| Line 2713 for (;;) | Line 3074 for (;;) |
| cb.capture_top = 1; | cb.capture_top = 1; |
| cb.capture_last = -1; | cb.capture_last = -1; |
| cb.callout_data = md->callout_data; | cb.callout_data = md->callout_data; |
| if ((rrc = (*pcre_callout)(&cb)) < 0) return rrc; /* Abandon */ | cb.mark = NULL; /* No (*MARK) support */ |
| if ((rrc = (*PUBL(callout))(&cb)) < 0) return rrc; /* Abandon */ | |
| } | } |
| if (rrc == 0) | if (rrc == 0) |
| { ADD_ACTIVE(state_offset + _pcre_OP_lengths[OP_CALLOUT], 0); } | { ADD_ACTIVE(state_offset + PRIV(OP_lengths)[OP_CALLOUT], 0); } |
| break; | break; |
| Line 2745 for (;;) | Line 3107 for (;;) |
| if (new_count <= 0) | if (new_count <= 0) |
| { | { |
| if (rlevel == 1 && /* Top level, and */ | if (rlevel == 1 && /* Top level, and */ |
| could_continue && /* Some could go on */ | could_continue && /* Some could go on, and */ |
| forced_fail != workspace[1] && /* Not all forced fail & */ | forced_fail != workspace[1] && /* Not all forced fail & */ |
| ( /* either... */ | ( /* either... */ |
| (md->moptions & PCRE_PARTIAL_HARD) != 0 /* Hard partial */ | (md->moptions & PCRE_PARTIAL_HARD) != 0 /* Hard partial */ |
| Line 2753 for (;;) | Line 3115 for (;;) |
| ((md->moptions & PCRE_PARTIAL_SOFT) != 0 && /* Soft partial and */ | ((md->moptions & PCRE_PARTIAL_SOFT) != 0 && /* Soft partial and */ |
| match_count < 0) /* no matches */ | match_count < 0) /* no matches */ |
| ) && /* And... */ | ) && /* And... */ |
| ptr >= end_subject && /* Reached end of subject */ | ( |
| ptr > md->start_used_ptr) /* Inspected non-empty string */ | partial_newline || /* Either partial NL */ |
| { | ( /* or ... */ |
| if (offsetcount >= 2) | ptr >= end_subject && /* End of subject and */ |
| { | ptr > md->start_used_ptr) /* Inspected non-empty string */ |
| offsets[0] = (int)(md->start_used_ptr - start_subject); | ) |
| offsets[1] = (int)(end_subject - start_subject); | ) |
| } | |
| match_count = PCRE_ERROR_PARTIAL; | match_count = PCRE_ERROR_PARTIAL; |
| } | |
| DPRINTF(("%.*sEnd of internal_dfa_exec %d: returning %d\n" | DPRINTF(("%.*sEnd of internal_dfa_exec %d: returning %d\n" |
| "%.*s---------------------\n\n", rlevel*2-2, SP, rlevel, match_count, | "%.*s---------------------\n\n", rlevel*2-2, SP, rlevel, match_count, |
| rlevel*2-2, SP)); | rlevel*2-2, SP)); |
| Line 2813 Returns: > 0 => number of match | Line 3172 Returns: > 0 => number of match |
| < -1 => some kind of unexpected problem | < -1 => some kind of unexpected problem |
| */ | */ |
| #if defined COMPILE_PCRE8 | |
| PCRE_EXP_DEFN int PCRE_CALL_CONVENTION | PCRE_EXP_DEFN int PCRE_CALL_CONVENTION |
| pcre_dfa_exec(const pcre *argument_re, const pcre_extra *extra_data, | pcre_dfa_exec(const pcre *argument_re, const pcre_extra *extra_data, |
| const char *subject, int length, int start_offset, int options, int *offsets, | const char *subject, int length, int start_offset, int options, int *offsets, |
| int offsetcount, int *workspace, int wscount) | int offsetcount, int *workspace, int wscount) |
| #elif defined COMPILE_PCRE16 | |
| PCRE_EXP_DEFN int PCRE_CALL_CONVENTION | |
| pcre16_dfa_exec(const pcre16 *argument_re, const pcre16_extra *extra_data, | |
| PCRE_SPTR16 subject, int length, int start_offset, int options, int *offsets, | |
| int offsetcount, int *workspace, int wscount) | |
| #elif defined COMPILE_PCRE32 | |
| PCRE_EXP_DEFN int PCRE_CALL_CONVENTION | |
| pcre32_dfa_exec(const pcre32 *argument_re, const pcre32_extra *extra_data, | |
| PCRE_SPTR32 subject, int length, int start_offset, int options, int *offsets, | |
| int offsetcount, int *workspace, int wscount) | |
| #endif | |
| { | { |
| real_pcre *re = (real_pcre *)argument_re; | REAL_PCRE *re = (REAL_PCRE *)argument_re; |
| dfa_match_data match_block; | dfa_match_data match_block; |
| dfa_match_data *md = &match_block; | dfa_match_data *md = &match_block; |
| BOOL utf8, anchored, startline, firstline; | BOOL utf, anchored, startline, firstline; |
| const uschar *current_subject, *end_subject, *lcc; | const pcre_uchar *current_subject, *end_subject; |
| pcre_study_data internal_study; | |
| const pcre_study_data *study = NULL; | const pcre_study_data *study = NULL; |
| real_pcre internal_re; | |
| const uschar *req_byte_ptr; | const pcre_uchar *req_char_ptr; |
| const uschar *start_bits = NULL; | const pcre_uint8 *start_bits = NULL; |
| BOOL first_byte_caseless = FALSE; | BOOL has_first_char = FALSE; |
| BOOL req_byte_caseless = FALSE; | BOOL has_req_char = FALSE; |
| int first_byte = -1; | pcre_uchar first_char = 0; |
| int req_byte = -1; | pcre_uchar first_char2 = 0; |
| int req_byte2 = -1; | pcre_uchar req_char = 0; |
| pcre_uchar req_char2 = 0; | |
| int newline; | int newline; |
| /* Plausibility checks */ | /* Plausibility checks */ |
| Line 2844 if (re == NULL || subject == NULL || wor | Line 3213 if (re == NULL || subject == NULL || wor |
| (offsets == NULL && offsetcount > 0)) return PCRE_ERROR_NULL; | (offsets == NULL && offsetcount > 0)) return PCRE_ERROR_NULL; |
| if (offsetcount < 0) return PCRE_ERROR_BADCOUNT; | if (offsetcount < 0) return PCRE_ERROR_BADCOUNT; |
| if (wscount < 20) return PCRE_ERROR_DFA_WSSIZE; | if (wscount < 20) return PCRE_ERROR_DFA_WSSIZE; |
| if (length < 0) return PCRE_ERROR_BADLENGTH; | |
| if (start_offset < 0 || start_offset > length) return PCRE_ERROR_BADOFFSET; | if (start_offset < 0 || start_offset > length) return PCRE_ERROR_BADOFFSET; |
| /* We need to find the pointer to any study data before we test for byte | /* Check that the first field in the block is the magic number. If it is not, |
| flipping, so we scan the extra_data block first. This may set two fields in the | return with PCRE_ERROR_BADMAGIC. However, if the magic number is equal to |
| match block, so we must initialize them beforehand. However, the other fields | REVERSED_MAGIC_NUMBER we return with PCRE_ERROR_BADENDIANNESS, which |
| in the match block must not be set until after the byte flipping. */ | means that the pattern is likely compiled with different endianness. */ |
| if (re->magic_number != MAGIC_NUMBER) | |
| return re->magic_number == REVERSED_MAGIC_NUMBER? | |
| PCRE_ERROR_BADENDIANNESS:PCRE_ERROR_BADMAGIC; | |
| if ((re->flags & PCRE_MODE) == 0) return PCRE_ERROR_BADMODE; | |
| /* If restarting after a partial match, do some sanity checks on the contents | |
| of the workspace. */ | |
| if ((options & PCRE_DFA_RESTART) != 0) | |
| { | |
| if ((workspace[0] & (-2)) != 0 || workspace[1] < 1 || | |
| workspace[1] > (wscount - 2)/INTS_PER_STATEBLOCK) | |
| return PCRE_ERROR_DFA_BADRESTART; | |
| } | |
| /* Set up study, callout, and table data */ | |
| md->tables = re->tables; | md->tables = re->tables; |
| md->callout_data = NULL; | md->callout_data = NULL; |
| if (extra_data != NULL) | if (extra_data != NULL) |
| { | { |
| unsigned int flags = extra_data->flags; | unsigned long int flags = extra_data->flags; |
| if ((flags & PCRE_EXTRA_STUDY_DATA) != 0) | if ((flags & PCRE_EXTRA_STUDY_DATA) != 0) |
| study = (const pcre_study_data *)extra_data->study_data; | study = (const pcre_study_data *)extra_data->study_data; |
| if ((flags & PCRE_EXTRA_MATCH_LIMIT) != 0) return PCRE_ERROR_DFA_UMLIMIT; | if ((flags & PCRE_EXTRA_MATCH_LIMIT) != 0) return PCRE_ERROR_DFA_UMLIMIT; |
| Line 2868 if (extra_data != NULL) | Line 3255 if (extra_data != NULL) |
| md->tables = extra_data->tables; | md->tables = extra_data->tables; |
| } | } |
| /* Check that the first field in the block is the magic number. If it is not, | |
| test for a regex that was compiled on a host of opposite endianness. If this is | |
| the case, flipped values are put in internal_re and internal_study if there was | |
| study data too. */ | |
| if (re->magic_number != MAGIC_NUMBER) | |
| { | |
| re = _pcre_try_flipped(re, &internal_re, study, &internal_study); | |
| if (re == NULL) return PCRE_ERROR_BADMAGIC; | |
| if (study != NULL) study = &internal_study; | |
| } | |
| /* Set some local values */ | /* Set some local values */ |
| current_subject = (const unsigned char *)subject + start_offset; | current_subject = (const pcre_uchar *)subject + start_offset; |
| end_subject = (const unsigned char *)subject + length; | end_subject = (const pcre_uchar *)subject + length; |
| req_byte_ptr = current_subject - 1; | req_char_ptr = current_subject - 1; |
| #ifdef SUPPORT_UTF8 | #ifdef SUPPORT_UTF |
| utf8 = (re->options & PCRE_UTF8) != 0; | /* PCRE_UTF(16|32) have the same value as PCRE_UTF8. */ |
| utf = (re->options & PCRE_UTF8) != 0; | |
| #else | #else |
| utf8 = FALSE; | utf = FALSE; |
| #endif | #endif |
| anchored = (options & (PCRE_ANCHORED|PCRE_DFA_RESTART)) != 0 || | anchored = (options & (PCRE_ANCHORED|PCRE_DFA_RESTART)) != 0 || |
| Line 2897 anchored = (options & (PCRE_ANCHORED|PCR | Line 3273 anchored = (options & (PCRE_ANCHORED|PCR |
| /* The remaining fixed data for passing around. */ | /* The remaining fixed data for passing around. */ |
| md->start_code = (const uschar *)argument_re + | md->start_code = (const pcre_uchar *)argument_re + |
| re->name_table_offset + re->name_count * re->name_entry_size; | re->name_table_offset + re->name_count * re->name_entry_size; |
| md->start_subject = (const unsigned char *)subject; | md->start_subject = (const pcre_uchar *)subject; |
| md->end_subject = end_subject; | md->end_subject = end_subject; |
| md->start_offset = start_offset; | md->start_offset = start_offset; |
| md->moptions = options; | md->moptions = options; |
| Line 2960 else | Line 3336 else |
| /* Check a UTF-8 string if required. Unfortunately there's no way of passing | /* Check a UTF-8 string if required. Unfortunately there's no way of passing |
| back the character offset. */ | back the character offset. */ |
| #ifdef SUPPORT_UTF8 | #ifdef SUPPORT_UTF |
| if (utf8 && (options & PCRE_NO_UTF8_CHECK) == 0) | if (utf && (options & PCRE_NO_UTF8_CHECK) == 0) |
| { | { |
| int tb; | int erroroffset; |
| if ((tb = _pcre_valid_utf8((uschar *)subject, length)) >= 0) | int errorcode = PRIV(valid_utf)((pcre_uchar *)subject, length, &erroroffset); |
| return (tb == length && (options & PCRE_PARTIAL_HARD) != 0)? | if (errorcode != 0) |
| PCRE_ERROR_SHORTUTF8 : PCRE_ERROR_BADUTF8; | |
| if (start_offset > 0 && start_offset < length) | |
| { | { |
| tb = ((USPTR)subject)[start_offset] & 0xc0; | if (offsetcount >= 2) |
| if (tb == 0x80) return PCRE_ERROR_BADUTF8_OFFSET; | { |
| offsets[0] = erroroffset; | |
| offsets[1] = errorcode; | |
| } | |
| #if defined COMPILE_PCRE8 | |
| return (errorcode <= PCRE_UTF8_ERR5 && (options & PCRE_PARTIAL_HARD) != 0) ? | |
| PCRE_ERROR_SHORTUTF8 : PCRE_ERROR_BADUTF8; | |
| #elif defined COMPILE_PCRE16 | |
| return (errorcode <= PCRE_UTF16_ERR1 && (options & PCRE_PARTIAL_HARD) != 0) ? | |
| PCRE_ERROR_SHORTUTF16 : PCRE_ERROR_BADUTF16; | |
| #elif defined COMPILE_PCRE32 | |
| return PCRE_ERROR_BADUTF32; | |
| #endif | |
| } | } |
| #if defined COMPILE_PCRE8 || defined COMPILE_PCRE16 | |
| if (start_offset > 0 && start_offset < length && | |
| NOT_FIRSTCHAR(((PCRE_PUCHAR)subject)[start_offset])) | |
| return PCRE_ERROR_BADUTF8_OFFSET; | |
| #endif | |
| } | } |
| #endif | #endif |
| Line 2979 if (utf8 && (options & PCRE_NO_UTF8_CHEC | Line 3370 if (utf8 && (options & PCRE_NO_UTF8_CHEC |
| is a feature that makes it possible to save compiled regex and re-use them | is a feature that makes it possible to save compiled regex and re-use them |
| in other programs later. */ | in other programs later. */ |
| if (md->tables == NULL) md->tables = _pcre_default_tables; | if (md->tables == NULL) md->tables = PRIV(default_tables); |
| /* The lower casing table and the "must be at the start of a line" flag are | /* The "must be at the start of a line" flags are used in a loop when finding |
| used in a loop when finding where to start. */ | where to start. */ |
| lcc = md->tables + lcc_offset; | |
| startline = (re->flags & PCRE_STARTLINE) != 0; | startline = (re->flags & PCRE_STARTLINE) != 0; |
| firstline = (re->options & PCRE_FIRSTLINE) != 0; | firstline = (re->options & PCRE_FIRSTLINE) != 0; |
| Line 2998 if (!anchored) | Line 3388 if (!anchored) |
| { | { |
| if ((re->flags & PCRE_FIRSTSET) != 0) | if ((re->flags & PCRE_FIRSTSET) != 0) |
| { | { |
| first_byte = re->first_byte & 255; | has_first_char = TRUE; |
| if ((first_byte_caseless = ((re->first_byte & REQ_CASELESS) != 0)) == TRUE) | first_char = first_char2 = (pcre_uchar)(re->first_char); |
| first_byte = lcc[first_byte]; | if ((re->flags & PCRE_FCH_CASELESS) != 0) |
| { | |
| first_char2 = TABLE_GET(first_char, md->tables + fcc_offset, first_char); | |
| #if defined SUPPORT_UCP && !(defined COMPILE_PCRE8) | |
| if (utf && first_char > 127) | |
| first_char2 = UCD_OTHERCASE(first_char); | |
| #endif | |
| } | |
| } | } |
| else | else |
| { | { |
| Line 3015 character" set. */ | Line 3412 character" set. */ |
| if ((re->flags & PCRE_REQCHSET) != 0) | if ((re->flags & PCRE_REQCHSET) != 0) |
| { | { |
| req_byte = re->req_byte & 255; | has_req_char = TRUE; |
| req_byte_caseless = (re->req_byte & REQ_CASELESS) != 0; | req_char = req_char2 = (pcre_uchar)(re->req_char); |
| req_byte2 = (md->tables + fcc_offset)[req_byte]; /* case flipped */ | if ((re->flags & PCRE_RCH_CASELESS) != 0) |
| { | |
| req_char2 = TABLE_GET(req_char, md->tables + fcc_offset, req_char); | |
| #if defined SUPPORT_UCP && !(defined COMPILE_PCRE8) | |
| if (utf && req_char > 127) | |
| req_char2 = UCD_OTHERCASE(req_char); | |
| #endif | |
| } | |
| } | } |
| /* Call the main matching function, looping for a non-anchored regex after a | /* Call the main matching function, looping for a non-anchored regex after a |
| Line 3030 for (;;) | Line 3434 for (;;) |
| if ((options & PCRE_DFA_RESTART) == 0) | if ((options & PCRE_DFA_RESTART) == 0) |
| { | { |
| const uschar *save_end_subject = end_subject; | const pcre_uchar *save_end_subject = end_subject; |
| /* If firstline is TRUE, the start of the match is constrained to the first | /* If firstline is TRUE, the start of the match is constrained to the first |
| line of a multiline string. Implement this by temporarily adjusting | line of a multiline string. Implement this by temporarily adjusting |
| Line 3039 for (;;) | Line 3443 for (;;) |
| if (firstline) | if (firstline) |
| { | { |
| USPTR t = current_subject; | PCRE_PUCHAR t = current_subject; |
| #ifdef SUPPORT_UTF8 | #ifdef SUPPORT_UTF |
| if (utf8) | if (utf) |
| { | { |
| while (t < md->end_subject && !IS_NEWLINE(t)) | while (t < md->end_subject && !IS_NEWLINE(t)) |
| { | { |
| t++; | t++; |
| while (t < end_subject && (*t & 0xc0) == 0x80) t++; | ACROSSCHAR(t < end_subject, *t, t++); |
| } | } |
| } | } |
| else | else |
| Line 3063 for (;;) | Line 3467 for (;;) |
| if (((options | re->options) & PCRE_NO_START_OPTIMIZE) == 0) | if (((options | re->options) & PCRE_NO_START_OPTIMIZE) == 0) |
| { | { |
| /* Advance to a known first byte. */ | /* Advance to a known first pcre_uchar (i.e. data item) */ |
| if (first_byte >= 0) | if (has_first_char) |
| { | { |
| if (first_byte_caseless) | if (first_char != first_char2) |
| { | |
| pcre_uchar csc; | |
| while (current_subject < end_subject && | while (current_subject < end_subject && |
| lcc[*current_subject] != first_byte) | (csc = UCHAR21TEST(current_subject)) != first_char && csc != first_char2) |
| current_subject++; | current_subject++; |
| } | |
| else | else |
| while (current_subject < end_subject && | while (current_subject < end_subject && |
| *current_subject != first_byte) | UCHAR21TEST(current_subject) != first_char) |
| current_subject++; | current_subject++; |
| } | } |
| Line 3083 for (;;) | Line 3490 for (;;) |
| { | { |
| if (current_subject > md->start_subject + start_offset) | if (current_subject > md->start_subject + start_offset) |
| { | { |
| #ifdef SUPPORT_UTF8 | #ifdef SUPPORT_UTF |
| if (utf8) | if (utf) |
| { | { |
| while (current_subject < end_subject && | while (current_subject < end_subject && |
| !WAS_NEWLINE(current_subject)) | !WAS_NEWLINE(current_subject)) |
| { | { |
| current_subject++; | current_subject++; |
| while(current_subject < end_subject && | ACROSSCHAR(current_subject < end_subject, *current_subject, |
| (*current_subject & 0xc0) == 0x80) | current_subject++); |
| current_subject++; | |
| } | } |
| } | } |
| else | else |
| Line 3104 for (;;) | Line 3510 for (;;) |
| ANYCRLF, and we are now at a LF, advance the match position by one | ANYCRLF, and we are now at a LF, advance the match position by one |
| more character. */ | more character. */ |
| if (current_subject[-1] == CHAR_CR && | if (UCHAR21TEST(current_subject - 1) == CHAR_CR && |
| (md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) && | (md->nltype == NLTYPE_ANY || md->nltype == NLTYPE_ANYCRLF) && |
| current_subject < end_subject && | current_subject < end_subject && |
| *current_subject == CHAR_NL) | UCHAR21TEST(current_subject) == CHAR_NL) |
| current_subject++; | current_subject++; |
| } | } |
| } | } |
| /* Or to a non-unique first char after study */ | /* Advance to a non-unique first pcre_uchar after study */ |
| else if (start_bits != NULL) | else if (start_bits != NULL) |
| { | { |
| while (current_subject < end_subject) | while (current_subject < end_subject) |
| { | { |
| register unsigned int c = *current_subject; | register pcre_uint32 c = UCHAR21TEST(current_subject); |
| if ((start_bits[c/8] & (1 << (c&7))) == 0) | #ifndef COMPILE_PCRE8 |
| { | if (c > 255) c = 255; |
| current_subject++; | |
| #ifdef SUPPORT_UTF8 | |
| if (utf8) | |
| while(current_subject < end_subject && | |
| (*current_subject & 0xc0) == 0x80) current_subject++; | |
| #endif | #endif |
| } | if ((start_bits[c/8] & (1 << (c&7))) != 0) break; |
| else break; | current_subject++; |
| } | } |
| } | } |
| } | } |
| Line 3141 for (;;) | Line 3542 for (;;) |
| disabling is explicitly requested (and of course, by the test above, this | disabling is explicitly requested (and of course, by the test above, this |
| code is not obeyed when restarting after a partial match). */ | code is not obeyed when restarting after a partial match). */ |
| if ((options & PCRE_NO_START_OPTIMIZE) == 0 && | if (((options | re->options) & PCRE_NO_START_OPTIMIZE) == 0 && |
| (options & (PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT)) == 0) | (options & (PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT)) == 0) |
| { | { |
| /* If the pattern was studied, a minimum subject length may be set. This | /* If the pattern was studied, a minimum subject length may be set. This |
| is a lower bound; no actual string of that length may actually match the | is a lower bound; no actual string of that length may actually match the |
| pattern. Although the value is, strictly, in characters, we treat it as | pattern. Although the value is, strictly, in characters, we treat it as |
| bytes to avoid spending too much time in this optimization. */ | in pcre_uchar units to avoid spending too much time in this optimization. |
| */ | |
| if (study != NULL && (study->flags & PCRE_STUDY_MINLEN) != 0 && | if (study != NULL && (study->flags & PCRE_STUDY_MINLEN) != 0 && |
| (pcre_uint32)(end_subject - current_subject) < study->minlength) | (pcre_uint32)(end_subject - current_subject) < study->minlength) |
| return PCRE_ERROR_NOMATCH; | return PCRE_ERROR_NOMATCH; |
| /* If req_byte is set, we know that that character must appear in the | /* If req_char is set, we know that that pcre_uchar must appear in the |
| subject for the match to succeed. If the first character is set, req_byte | subject for the match to succeed. If the first pcre_uchar is set, |
| must be later in the subject; otherwise the test starts at the match | req_char must be later in the subject; otherwise the test starts at the |
| point. This optimization can save a huge amount of work in patterns with | match point. This optimization can save a huge amount of work in patterns |
| nested unlimited repeats that aren't going to match. Writing separate | with nested unlimited repeats that aren't going to match. Writing |
| code for cased/caseless versions makes it go faster, as does using an | separate code for cased/caseless versions makes it go faster, as does |
| autoincrement and backing off on a match. | using an autoincrement and backing off on a match. |
| HOWEVER: when the subject string is very, very long, searching to its end | HOWEVER: when the subject string is very, very long, searching to its end |
| can take a long time, and give bad performance on quite ordinary | can take a long time, and give bad performance on quite ordinary |
| patterns. This showed up when somebody was matching /^C/ on a 32-megabyte | patterns. This showed up when somebody was matching /^C/ on a 32-megabyte |
| string... so we don't do this when the string is sufficiently long. */ | string... so we don't do this when the string is sufficiently long. */ |
| if (req_byte >= 0 && end_subject - current_subject < REQ_BYTE_MAX) | if (has_req_char && end_subject - current_subject < REQ_BYTE_MAX) |
| { | { |
| register const uschar *p = current_subject + ((first_byte >= 0)? 1 : 0); | register PCRE_PUCHAR p = current_subject + (has_first_char? 1:0); |
| /* We don't need to repeat the search if we haven't yet reached the | /* We don't need to repeat the search if we haven't yet reached the |
| place we found it at last time. */ | place we found it at last time. */ |
| if (p > req_byte_ptr) | if (p > req_char_ptr) |
| { | { |
| if (req_byte_caseless) | if (req_char != req_char2) |
| { | { |
| while (p < end_subject) | while (p < end_subject) |
| { | { |
| register int pp = *p++; | register pcre_uint32 pp = UCHAR21INCTEST(p); |
| if (pp == req_byte || pp == req_byte2) { p--; break; } | if (pp == req_char || pp == req_char2) { p--; break; } |
| } | } |
| } | } |
| else | else |
| { | { |
| while (p < end_subject) | while (p < end_subject) |
| { | { |
| if (*p++ == req_byte) { p--; break; } | if (UCHAR21INCTEST(p) == req_char) { p--; break; } |
| } | } |
| } | } |
| /* If we can't find the required character, break the matching loop, | /* If we can't find the required pcre_uchar, break the matching loop, |
| which will cause a return or PCRE_ERROR_NOMATCH. */ | which will cause a return or PCRE_ERROR_NOMATCH. */ |
| if (p >= end_subject) break; | if (p >= end_subject) break; |
| /* If we have found the required character, save the point where we | /* If we have found the required pcre_uchar, save the point where we |
| found it, so that we don't search again next time round the loop if | found it, so that we don't search again next time round the loop if |
| the start hasn't passed this character yet. */ | the start hasn't passed this point yet. */ |
| req_byte_ptr = p; | req_char_ptr = p; |
| } | } |
| } | } |
| } | } |
| Line 3209 for (;;) | Line 3611 for (;;) |
| /* OK, now we can do the business */ | /* OK, now we can do the business */ |
| md->start_used_ptr = current_subject; | md->start_used_ptr = current_subject; |
| md->recursive = NULL; | |
| rc = internal_dfa_exec( | rc = internal_dfa_exec( |
| md, /* fixed match data */ | md, /* fixed match data */ |
| Line 3219 for (;;) | Line 3622 for (;;) |
| offsetcount, /* size of same */ | offsetcount, /* size of same */ |
| workspace, /* workspace vector */ | workspace, /* workspace vector */ |
| wscount, /* size of same */ | wscount, /* size of same */ |
| re->options & (PCRE_CASELESS|PCRE_MULTILINE|PCRE_DOTALL), /* ims flags */ | 0); /* function recurse level */ |
| 0, /* function recurse level */ | |
| 0); /* regex recurse level */ | |
| /* Anything other than "no match" means we are done, always; otherwise, carry | /* Anything other than "no match" means we are done, always; otherwise, carry |
| on only if not anchored. */ | on only if not anchored. */ |
| if (rc != PCRE_ERROR_NOMATCH || anchored) return rc; | if (rc != PCRE_ERROR_NOMATCH || anchored) |
| { | |
| if (rc == PCRE_ERROR_PARTIAL && offsetcount >= 2) | |
| { | |
| offsets[0] = (int)(md->start_used_ptr - (PCRE_PUCHAR)subject); | |
| offsets[1] = (int)(end_subject - (PCRE_PUCHAR)subject); | |
| if (offsetcount > 2) | |
| offsets[2] = (int)(current_subject - (PCRE_PUCHAR)subject); | |
| } | |
| return rc; | |
| } | |
| /* Advance to the next subject character unless we are at the end of a line | /* Advance to the next subject character unless we are at the end of a line |
| and firstline is set. */ | and firstline is set. */ |
| if (firstline && IS_NEWLINE(current_subject)) break; | if (firstline && IS_NEWLINE(current_subject)) break; |
| current_subject++; | current_subject++; |
| if (utf8) | #ifdef SUPPORT_UTF |
| if (utf) | |
| { | { |
| while (current_subject < end_subject && (*current_subject & 0xc0) == 0x80) | ACROSSCHAR(current_subject < end_subject, *current_subject, |
| current_subject++; | current_subject++); |
| } | } |
| #endif | |
| if (current_subject > end_subject) break; | if (current_subject > end_subject) break; |
| /* If we have just passed a CR and we are now at a LF, and the pattern does | /* If we have just passed a CR and we are now at a LF, and the pattern does |
| not contain any explicit matches for \r or \n, and the newline option is CRLF | not contain any explicit matches for \r or \n, and the newline option is CRLF |
| or ANY or ANYCRLF, advance the match position by one more character. */ | or ANY or ANYCRLF, advance the match position by one more character. */ |
| if (current_subject[-1] == CHAR_CR && | if (UCHAR21TEST(current_subject - 1) == CHAR_CR && |
| current_subject < end_subject && | current_subject < end_subject && |
| *current_subject == CHAR_NL && | UCHAR21TEST(current_subject) == CHAR_NL && |
| (re->flags & PCRE_HASCRORLF) == 0 && | (re->flags & PCRE_HASCRORLF) == 0 && |
| (md->nltype == NLTYPE_ANY || | (md->nltype == NLTYPE_ANY || |
| md->nltype == NLTYPE_ANYCRLF || | md->nltype == NLTYPE_ANYCRLF || |