1 /*************************************************
2 * Perl-Compatible Regular Expressions *
3 *************************************************/
6 /* PCRE is a library of functions to support regular expressions whose syntax
7 and semantics are as close as possible to those of the Perl 5 language.
9 Written by Philip Hazel
10 Copyright (c) 1997-2006 University of Cambridge
12 -----------------------------------------------------------------------------
13 Redistribution and use in source and binary forms, with or without
14 modification, are permitted provided that the following conditions are met:
16 * Redistributions of source code must retain the above copyright notice,
17 this list of conditions and the following disclaimer.
19 * Redistributions in binary form must reproduce the above copyright
20 notice, this list of conditions and the following disclaimer in the
21 documentation and/or other materials provided with the distribution.
23 * Neither the name of the University of Cambridge nor the names of its
24 contributors may be used to endorse or promote products derived from
25 this software without specific prior written permission.
27 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
28 AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
29 IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
30 ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
31 LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
32 CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
33 SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
34 INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
35 CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
37 POSSIBILITY OF SUCH DAMAGE.
38 -----------------------------------------------------------------------------
41 /* This header contains definitions that are shared between the different
42 modules, but which are not relevant to the exported API. This includes some
43 functions whose names all begin with "_pcre_". */
45 #ifndef PCRE_INTERNAL_H
46 #define PCRE_INTERNAL_H
48 /* Define DEBUG to get debugging output on stdout. */
54 /* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef
55 inline, and there are *still* stupid compilers about that don't like indented
56 pre-processor statements, or at least there were when I first wrote this. After
57 all, it had only been about 10 years then...
59 It turns out that the Mac Debugging.h header also defines the macro DPRINTF, so
60 be absolutely sure we get our version. */
64 #define DPRINTF(p) printf p
66 #define DPRINTF(p) /* Nothing */
70 /* Get the definitions provided by running "configure" */
74 /* Standard C headers plus the external interface definition. The only time
75 setjmp and stdarg are used is when NO_RECURSE is set. */
87 #define PCRE_DEFINITION /* Win32 __declspec(export) trigger for .dll */
90 /* We need to have types that specify unsigned 16-bit and 32-bit integers. We
91 cannot determine these outside the compilation (e.g. by running a program as
92 part of "configure") because PCRE is often cross-compiled for use on other
93 systems. Instead we make use of the maximum sizes that are available at
94 preprocessor time in standard C environments. */
96 #if USHRT_MAX == 65535
97 typedef unsigned short pcre_uint16;
98 #elif UINT_MAX == 65535
99 typedef unsigned int pcre_uint16;
101 #error Cannot determine a type for 16-bit unsigned integers
104 #if UINT_MAX == 4294967295
105 typedef unsigned int pcre_uint32;
106 #elif ULONG_MAX == 4294967295
107 typedef unsigned long int pcre_uint32;
109 #error Cannot determine a type for 32-bit unsigned integers
112 /* All character handling must be done as unsigned characters. Otherwise there
113 are problems with top-bit-set characters and functions such as isspace().
114 However, we leave the interface to the outside world as char *, because that
115 should make things easier for callers. We define a short type for unsigned char
116 to save lots of typing. I tried "uchar", but it causes problems on Digital
117 Unix, where it is defined in sys/types, so use "uschar" instead. */
119 typedef unsigned char uschar;
121 /* This is an unsigned int value that no character can ever have. UTF-8
122 characters only go up to 0x7fffffff (though Unicode doesn't go beyond
125 #define NOTACHAR 0xffffffff
127 /* PCRE is able to support several different kinds of newline (CR, LF, CRLF,
128 and "all" at present). The following macros are used to package up testing for
129 newlines. NLBLOCK, PSSTART, and PSEND are defined in the various modules to
130 indicate in which datablock the parameters exist, and what the start/end of
131 string field names are. */
133 #define NLTYPE_FIXED 0 /* Newline is a fixed length string */
134 #define NLTYPE_ANY 1 /* Newline is any Unicode line ending */
136 /* This macro checks for a newline at the given position */
138 #define IS_NEWLINE(p) \
139 ((NLBLOCK->nltype != NLTYPE_FIXED)? \
140 ((p) < NLBLOCK->PSEND && \
141 _pcre_is_newline((p), NLBLOCK->PSEND, &(NLBLOCK->nllen), utf8) \
144 ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \
145 (p)[0] == NLBLOCK->nl[0] && \
146 (NLBLOCK->nllen == 1 || (p)[1] == NLBLOCK->nl[1]) \
150 /* This macro checks for a newline immediately preceding the given position */
152 #define WAS_NEWLINE(p) \
153 ((NLBLOCK->nltype != NLTYPE_FIXED)? \
154 ((p) > NLBLOCK->PSSTART && \
155 _pcre_was_newline((p), NLBLOCK->PSSTART, &(NLBLOCK->nllen), utf8) \
158 ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \
159 (p)[-NLBLOCK->nllen] == NLBLOCK->nl[0] && \
160 (NLBLOCK->nllen == 1 || (p)[-NLBLOCK->nllen+1] == NLBLOCK->nl[1]) \
164 /* When PCRE is compiled as a C++ library, the subject pointer can be replaced
165 with a custom type. This makes it possible, for example, to allow pcre_exec()
166 to process subject strings that are discontinuous by using a smart pointer
167 class. It must always be possible to inspect all of the subject string in
168 pcre_exec() because of the way it backtracks. Two macros are required in the
169 normal case, for sign-unspecified and unsigned char pointers. The former is
170 used for the external interface and appears in pcre.h, which is why its name
171 must begin with PCRE_. */
173 #ifdef CUSTOM_SUBJECT_PTR
174 #define PCRE_SPTR CUSTOM_SUBJECT_PTR
175 #define USPTR CUSTOM_SUBJECT_PTR
177 #define PCRE_SPTR const char *
178 #define USPTR const unsigned char *
181 /* Include the public PCRE header and the definitions of UCP character property
187 /* When compiling for use with the Virtual Pascal compiler, these functions
188 need to have their names changed. PCRE must be compiled with the -DVPCOMPAT
189 option on the command line. */
192 #define strncmp(s1,s2,m) _strncmp(s1,s2,m)
193 #define memcpy(d,s,n) _memcpy(d,s,n)
194 #define memmove(d,s,n) _memmove(d,s,n)
195 #define memset(s,c,n) _memset(s,c,n)
198 /* To cope with SunOS4 and other systems that lack memmove() but have bcopy(),
199 define a macro for memmove() if HAVE_MEMMOVE is false, provided that HAVE_BCOPY
200 is set. Otherwise, include an emulating function for those systems that have
201 neither (there some non-Unix environments where this is the case). This assumes
202 that all calls to memmove are moving strings upwards in store, which is the
206 #undef memmove /* some systems may have a macro */
208 #define memmove(a, b, c) bcopy(b, a, c)
209 #else /* HAVE_BCOPY */
211 pcre_memmove(unsigned char *dest, const unsigned char *src, size_t n)
216 for (i = 0; i < n; ++i) *(--dest) = *(--src);
219 #define memmove(a, b, c) pcre_memmove(a, b, c)
220 #endif /* not HAVE_BCOPY */
221 #endif /* not HAVE_MEMMOVE */
222 #endif /* not VPCOMPAT */
225 /* PCRE keeps offsets in its compiled code as 2-byte quantities (always stored
226 in big-endian order) by default. These are used, for example, to link from the
227 start of a subpattern to its alternatives and its end. The use of 2 bytes per
228 offset limits the size of the compiled regex to around 64K, which is big enough
229 for almost everybody. However, I received a request for an even bigger limit.
230 For this reason, and also to make the code easier to maintain, the storing and
231 loading of offsets from the byte string is now handled by the macros that are
234 The macros are controlled by the value of LINK_SIZE. This defaults to 2 in
235 the config.h file, but can be overridden by using -D on the command line. This
236 is automated on Unix systems via the "configure" command. */
242 (a[(n)+1] = (d) & 255)
245 (((a)[n] << 8) | (a)[(n)+1])
247 #define MAX_PATTERN_SIZE (1 << 16)
253 (a[n] = (d) >> 16), \
254 (a[(n)+1] = (d) >> 8), \
255 (a[(n)+2] = (d) & 255)
258 (((a)[n] << 16) | ((a)[(n)+1] << 8) | (a)[(n)+2])
260 #define MAX_PATTERN_SIZE (1 << 24)
266 (a[n] = (d) >> 24), \
267 (a[(n)+1] = (d) >> 16), \
268 (a[(n)+2] = (d) >> 8), \
269 (a[(n)+3] = (d) & 255)
272 (((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3])
274 #define MAX_PATTERN_SIZE (1 << 30) /* Keep it positive */
278 #error LINK_SIZE must be either 2, 3, or 4
282 /* Convenience macro defined in terms of the others */
284 #define PUTINC(a,n,d) PUT(a,n,d), a += LINK_SIZE
287 /* PCRE uses some other 2-byte quantities that do not change when the size of
288 offsets changes. There are used for repeat counts and for other things such as
289 capturing parenthesis numbers in back references. */
291 #define PUT2(a,n,d) \
296 (((a)[n] << 8) | (a)[(n)+1])
298 #define PUT2INC(a,n,d) PUT2(a,n,d), a += 2
301 /* When UTF-8 encoding is being used, a character is no longer just a single
302 byte. The macros for character handling generate simple sequences when used in
303 byte-mode, and more complicated ones for UTF-8 characters. */
306 #define GETCHAR(c, eptr) c = *eptr;
307 #define GETCHARTEST(c, eptr) c = *eptr;
308 #define GETCHARINC(c, eptr) c = *eptr++;
309 #define GETCHARINCTEST(c, eptr) c = *eptr++;
310 #define GETCHARLEN(c, eptr, len) c = *eptr;
311 #define BACKCHAR(eptr)
313 #else /* SUPPORT_UTF8 */
315 /* Get the next UTF-8 character, not advancing the pointer. This is called when
316 we know we are in UTF-8 mode. */
318 #define GETCHAR(c, eptr) \
323 int gcaa = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */ \
325 c = (c & _pcre_utf8_table3[gcaa]) << gcss; \
326 for (gcii = 1; gcii <= gcaa; gcii++) \
329 c |= (eptr[gcii] & 0x3f) << gcss; \
333 /* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the
336 #define GETCHARTEST(c, eptr) \
338 if (utf8 && c >= 0xc0) \
341 int gcaa = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */ \
343 c = (c & _pcre_utf8_table3[gcaa]) << gcss; \
344 for (gcii = 1; gcii <= gcaa; gcii++) \
347 c |= (eptr[gcii] & 0x3f) << gcss; \
351 /* Get the next UTF-8 character, advancing the pointer. This is called when we
352 know we are in UTF-8 mode. */
354 #define GETCHARINC(c, eptr) \
358 int gcaa = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */ \
360 c = (c & _pcre_utf8_table3[gcaa]) << gcss; \
364 c |= (*eptr++ & 0x3f) << gcss; \
368 /* Get the next character, testing for UTF-8 mode, and advancing the pointer */
370 #define GETCHARINCTEST(c, eptr) \
372 if (utf8 && c >= 0xc0) \
374 int gcaa = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */ \
376 c = (c & _pcre_utf8_table3[gcaa]) << gcss; \
380 c |= (*eptr++ & 0x3f) << gcss; \
384 /* Get the next UTF-8 character, not advancing the pointer, incrementing length
385 if there are extra bytes. This is called when we know we are in UTF-8 mode. */
387 #define GETCHARLEN(c, eptr, len) \
392 int gcaa = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */ \
394 c = (c & _pcre_utf8_table3[gcaa]) << gcss; \
395 for (gcii = 1; gcii <= gcaa; gcii++) \
398 c |= (eptr[gcii] & 0x3f) << gcss; \
403 /* If the pointer is not at the start of a character, move it back until
404 it is. Called only in UTF-8 mode. */
406 #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--;
411 /* In case there is no definition of offsetof() provided - though any proper
412 Standard C system should have one. */
415 #define offsetof(p_type,field) ((size_t)&(((p_type *)0)->field))
419 /* These are the public options that can change during matching. */
421 #define PCRE_IMS (PCRE_CASELESS|PCRE_MULTILINE|PCRE_DOTALL)
423 /* Private options flags start at the most significant end of the four bytes.
424 The public options defined in pcre.h start at the least significant end. Make
425 sure they don't overlap! The bits are getting a bit scarce now -- when we run
426 out, there is a dummy word in the structure that could be used for the private
429 #define PCRE_NOPARTIAL 0x80000000 /* can't use partial with this regex */
430 #define PCRE_FIRSTSET 0x40000000 /* first_byte is set */
431 #define PCRE_REQCHSET 0x20000000 /* req_byte is set */
432 #define PCRE_STARTLINE 0x10000000 /* start after \n for multiline */
433 #define PCRE_JCHANGED 0x08000000 /* j option changes within regex */
435 /* Options for the "extra" block produced by pcre_study(). */
437 #define PCRE_STUDY_MAPPED 0x01 /* a map of starting chars exists */
439 /* Masks for identifying the public options that are permitted at compile
440 time, run time, or study time, respectively. */
442 #define PCRE_NEWLINE_BITS (PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_ANY)
444 #define PUBLIC_OPTIONS \
445 (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \
446 PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \
447 PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \
448 PCRE_DUPNAMES|PCRE_NEWLINE_BITS)
450 #define PUBLIC_EXEC_OPTIONS \
451 (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \
452 PCRE_PARTIAL|PCRE_NEWLINE_BITS)
454 #define PUBLIC_DFA_EXEC_OPTIONS \
455 (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \
456 PCRE_PARTIAL|PCRE_DFA_SHORTEST|PCRE_DFA_RESTART|PCRE_NEWLINE_BITS)
458 #define PUBLIC_STUDY_OPTIONS 0 /* None defined */
460 /* Magic number to provide a small check against being handed junk. Also used
461 to detect whether a pattern was compiled on a host of different endianness. */
463 #define MAGIC_NUMBER 0x50435245UL /* 'PCRE' */
465 /* Negative values for the firstchar and reqchar variables */
467 #define REQ_UNSET (-2)
468 #define REQ_NONE (-1)
470 /* The maximum remaining length of subject we are prepared to search for a
473 #define REQ_BYTE_MAX 1000
475 /* Flags added to firstbyte or reqbyte; a "non-literal" item is either a
476 variable-length repeat, or a anything other than literal characters. */
478 #define REQ_CASELESS 0x0100 /* indicates caselessness */
479 #define REQ_VARY 0x0200 /* reqbyte followed non-literal item */
481 /* Miscellaneous definitions */
483 typedef gboolean BOOL;
485 /* Escape items that are just an encoding of a particular data value. */
503 /* We can't officially use ESC_t because it is a POSIX reserved identifier
504 (presumably because of all the others like size_t). */
510 /* Codes for different types of Unicode property */
512 #define PT_ANY 0 /* Any property - matches all chars */
513 #define PT_LAMP 1 /* L& - the union of Lu, Ll, Lt */
514 #define PT_GC 2 /* General characteristic (e.g. L) */
515 #define PT_PC 3 /* Particular characteristic (e.g. Lu) */
516 #define PT_SC 4 /* Script (e.g. Han) */
518 /* Flag bits and data types for the extended class (OP_XCLASS) for classes that
519 contain UTF-8 characters with values greater than 255. */
521 #define XCL_NOT 0x01 /* Flag: this is a negative class */
522 #define XCL_MAP 0x02 /* Flag: a 32-byte map is present */
524 #define XCL_END 0 /* Marks end of individual items */
525 #define XCL_SINGLE 1 /* Single item (one multibyte char) follows */
526 #define XCL_RANGE 2 /* A range (two multibyte chars) follows */
527 #define XCL_PROP 3 /* Unicode property (2-byte property code follows) */
528 #define XCL_NOTPROP 4 /* Unicode inverted property (ditto) */
530 /* These are escaped items that aren't just an encoding of a particular data
531 value such as \n. They must have non-zero values, as check_escape() returns
532 their negation. Also, they must appear in the same order as in the opcode
533 definitions below, up to ESC_z. There's a dummy for OP_ANY because it
534 corresponds to "." rather than an escape sequence. The final one must be
535 ESC_REF as subsequent values are used for backreferences (\1, \2, \3, etc).
536 There are two tests in the code for an escape greater than ESC_b and less than
537 ESC_Z to detect the types that may be repeated. These are the types that
538 consume characters. If any new escapes are put in between that don't consume a
539 character, that code will have to change. */
541 enum { ESC_A = 1, ESC_G, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s, ESC_W,
542 ESC_w, ESC_dum1, ESC_C, ESC_P, ESC_p, ESC_R, ESC_X, ESC_Z, ESC_z,
543 ESC_E, ESC_Q, ESC_k, ESC_REF };
546 /* Opcode table: OP_BRA must be last, as all values >= it are used for brackets
547 that extract substrings. Starting from 1 (i.e. after OP_END), the values up to
548 OP_EOD must correspond in order to the list of escapes immediately above.
550 To keep stored, compiled patterns compatible, new opcodes should be added
551 immediately before OP_BRA, where (since release 7.0) a gap is left for this
554 *** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions
555 that follow must also be updated to match. There is also a table called
556 "coptable" in pcre_dfa_exec.c that must be updated. */
559 OP_END, /* 0 End of pattern */
561 /* Values corresponding to backslashed metacharacters */
563 OP_SOD, /* 1 Start of data: \A */
564 OP_SOM, /* 2 Start of match (subject + offset): \G */
565 OP_NOT_WORD_BOUNDARY, /* 3 \B */
566 OP_WORD_BOUNDARY, /* 4 \b */
567 OP_NOT_DIGIT, /* 5 \D */
569 OP_NOT_WHITESPACE, /* 7 \S */
570 OP_WHITESPACE, /* 8 \s */
571 OP_NOT_WORDCHAR, /* 9 \W */
572 OP_WORDCHAR, /* 10 \w */
573 OP_ANY, /* 11 Match any character */
574 OP_ANYBYTE, /* 12 Match any byte (\C); different to OP_ANY for UTF-8 */
575 OP_NOTPROP, /* 13 \P (not Unicode property) */
576 OP_PROP, /* 14 \p (Unicode property) */
577 OP_ANYNL, /* 15 \R (any newline sequence) */
578 OP_EXTUNI, /* 16 \X (extended Unicode sequence */
579 OP_EODN, /* 17 End of data or \n at end of data: \Z. */
580 OP_EOD, /* 18 End of data: \z */
582 OP_OPT, /* 19 Set runtime options */
583 OP_CIRC, /* 20 Start of line - varies with multiline switch */
584 OP_DOLL, /* 21 End of line - varies with multiline switch */
585 OP_CHAR, /* 22 Match one character, casefully */
586 OP_CHARNC, /* 23 Match one character, caselessly */
587 OP_NOT, /* 24 Match one character, not the following one */
589 OP_STAR, /* 25 The maximizing and minimizing versions of */
590 OP_MINSTAR, /* 26 these six opcodes must come in pairs, with */
591 OP_PLUS, /* 27 the minimizing one second. */
592 OP_MINPLUS, /* 28 This first set applies to single characters.*/
594 OP_MINQUERY, /* 30 */
596 OP_UPTO, /* 31 From 0 to n matches */
598 OP_EXACT, /* 33 Exactly n matches */
600 OP_POSSTAR, /* 34 Possessified star */
601 OP_POSPLUS, /* 35 Possessified plus */
602 OP_POSQUERY, /* 36 Posesssified query */
603 OP_POSUPTO, /* 37 Possessified upto */
605 OP_NOTSTAR, /* 38 The maximizing and minimizing versions of */
606 OP_NOTMINSTAR, /* 39 these six opcodes must come in pairs, with */
607 OP_NOTPLUS, /* 40 the minimizing one second. They must be in */
608 OP_NOTMINPLUS, /* 41 exactly the same order as those above. */
609 OP_NOTQUERY, /* 42 This set applies to "not" single characters. */
610 OP_NOTMINQUERY, /* 43 */
612 OP_NOTUPTO, /* 44 From 0 to n matches */
613 OP_NOTMINUPTO, /* 45 */
614 OP_NOTEXACT, /* 46 Exactly n matches */
616 OP_NOTPOSSTAR, /* 47 Possessified versions */
617 OP_NOTPOSPLUS, /* 48 */
618 OP_NOTPOSQUERY, /* 49 */
619 OP_NOTPOSUPTO, /* 50 */
621 OP_TYPESTAR, /* 51 The maximizing and minimizing versions of */
622 OP_TYPEMINSTAR, /* 52 these six opcodes must come in pairs, with */
623 OP_TYPEPLUS, /* 53 the minimizing one second. These codes must */
624 OP_TYPEMINPLUS, /* 54 be in exactly the same order as those above. */
625 OP_TYPEQUERY, /* 55 This set applies to character types such as \d */
626 OP_TYPEMINQUERY, /* 56 */
628 OP_TYPEUPTO, /* 57 From 0 to n matches */
629 OP_TYPEMINUPTO, /* 58 */
630 OP_TYPEEXACT, /* 59 Exactly n matches */
632 OP_TYPEPOSSTAR, /* 60 Possessified versions */
633 OP_TYPEPOSPLUS, /* 61 */
634 OP_TYPEPOSQUERY, /* 62 */
635 OP_TYPEPOSUPTO, /* 63 */
637 OP_CRSTAR, /* 64 The maximizing and minimizing versions of */
638 OP_CRMINSTAR, /* 65 all these opcodes must come in pairs, with */
639 OP_CRPLUS, /* 66 the minimizing one second. These codes must */
640 OP_CRMINPLUS, /* 67 be in exactly the same order as those above. */
641 OP_CRQUERY, /* 68 These are for character classes and back refs */
642 OP_CRMINQUERY, /* 69 */
643 OP_CRRANGE, /* 70 These are different to the three sets above. */
644 OP_CRMINRANGE, /* 71 */
646 OP_CLASS, /* 72 Match a character class, chars < 256 only */
647 OP_NCLASS, /* 73 Same, but the bitmap was created from a negative
648 class - the difference is relevant only when a UTF-8
649 character > 255 is encountered. */
651 OP_XCLASS, /* 74 Extended class for handling UTF-8 chars within the
652 class. This does both positive and negative. */
654 OP_REF, /* 75 Match a back reference */
655 OP_RECURSE, /* 76 Match a numbered subpattern (possibly recursive) */
656 OP_CALLOUT, /* 77 Call out to external function if provided */
658 OP_ALT, /* 78 Start of alternation */
659 OP_KET, /* 79 End of group that doesn't have an unbounded repeat */
660 OP_KETRMAX, /* 80 These two must remain together and in this */
661 OP_KETRMIN, /* 81 order. They are for groups the repeat for ever. */
663 /* The assertions must come before BRA, CBRA, ONCE, and COND.*/
665 OP_ASSERT, /* 82 Positive lookahead */
666 OP_ASSERT_NOT, /* 83 Negative lookahead */
667 OP_ASSERTBACK, /* 84 Positive lookbehind */
668 OP_ASSERTBACK_NOT, /* 85 Negative lookbehind */
669 OP_REVERSE, /* 86 Move pointer back - used in lookbehind assertions */
671 /* ONCE, BRA, CBRA, and COND must come after the assertions, with ONCE first,
672 as there's a test for >= ONCE for a subpattern that isn't an assertion. */
674 OP_ONCE, /* 87 Atomic group */
675 OP_BRA, /* 88 Start of non-capturing bracket */
676 OP_CBRA, /* 89 Start of capturing bracket */
677 OP_COND, /* 90 Conditional group */
679 /* These three must follow the previous three, in the same order. There's a
680 check for >= SBRA to distinguish the two sets. */
682 OP_SBRA, /* 91 Start of non-capturing bracket, check empty */
683 OP_SCBRA, /* 92 Start of capturing bracket, check empty */
684 OP_SCOND, /* 93 Conditional group, check empty */
686 OP_CREF, /* 94 Used to hold a capture number as condition */
687 OP_RREF, /* 95 Used to hold a recursion number as condition */
688 OP_DEF, /* 96 The DEFINE condition */
690 OP_BRAZERO, /* 97 These two must remain together and in this */
691 OP_BRAMINZERO /* 98 order. */
695 /* This macro defines textual names for all the opcodes. These are used only
696 for debugging. The macro is referenced only in pcre_printint.c. */
698 #define OP_NAME_LIST \
699 "End", "\\A", "\\G", "\\B", "\\b", "\\D", "\\d", \
700 "\\S", "\\s", "\\W", "\\w", "Any", "Anybyte", \
701 "notprop", "prop", "anynl", "extuni", \
703 "Opt", "^", "$", "char", "charnc", "not", \
704 "*", "*?", "+", "+?", "?", "??", "{", "{", "{", \
705 "*+","++", "?+", "{", \
706 "*", "*?", "+", "+?", "?", "??", "{", "{", "{", \
707 "*+","++", "?+", "{", \
708 "*", "*?", "+", "+?", "?", "??", "{", "{", "{", \
709 "*+","++", "?+", "{", \
710 "*", "*?", "+", "+?", "?", "??", "{", "{", \
711 "class", "nclass", "xclass", "Ref", "Recurse", "Callout", \
712 "Alt", "Ket", "KetRmax", "KetRmin", "Assert", "Assert not", \
713 "AssertB", "AssertB not", "Reverse", \
714 "Once", "Bra 0", "Bra", "Cond", "SBra 0", "SBra", "SCond", \
715 "Cond ref", "Cond rec", "Cond def", "Brazero", "Braminzero"
718 /* This macro defines the length of fixed length operations in the compiled
719 regex. The lengths are used when searching for specific things, and also in the
720 debugging printing of a compiled regex. We use a macro so that it can be
721 defined close to the definitions of the opcodes themselves.
723 As things have been extended, some of these are no longer fixed lenths, but are
724 minima instead. For example, the length of a single-character repeat may vary
725 in UTF-8 mode. The code that uses this table must know about such things. */
729 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* \A, \G, \B, \B, \D, \d, \S, \s, \W, \w */ \
730 1, 1, /* Any, Anybyte */ \
731 3, 3, 1, 1, /* NOTPROP, PROP, EXTUNI, ANYNL */ \
732 1, 1, 2, 1, 1, /* \Z, \z, Opt, ^, $ */ \
733 2, /* Char - the minimum length */ \
734 2, /* Charnc - the minimum length */ \
736 /* Positive single-char repeats ** These are */ \
737 2, 2, 2, 2, 2, 2, /* *, *?, +, +?, ?, ?? ** minima in */ \
738 4, 4, 4, /* upto, minupto, exact ** UTF-8 mode */ \
739 2, 2, 2, 4, /* *+, ++, ?+, upto+ */ \
740 /* Negative single-char repeats - only for chars < 256 */ \
741 2, 2, 2, 2, 2, 2, /* NOT *, *?, +, +?, ?, ?? */ \
742 4, 4, 4, /* NOT upto, minupto, exact */ \
743 2, 2, 2, 4, /* Possessive *, +, ?, upto */ \
744 /* Positive type repeats */ \
745 2, 2, 2, 2, 2, 2, /* Type *, *?, +, +?, ?, ?? */ \
746 4, 4, 4, /* Type upto, minupto, exact */ \
747 2, 2, 2, 4, /* Possessive *+, ++, ?+, upto+ */ \
748 /* Character class & ref repeats */ \
749 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ \
750 5, 5, /* CRRANGE, CRMINRANGE */ \
753 0, /* XCLASS - variable length */ \
755 1+LINK_SIZE, /* RECURSE */ \
756 2+2*LINK_SIZE, /* CALLOUT */ \
757 1+LINK_SIZE, /* Alt */ \
758 1+LINK_SIZE, /* Ket */ \
759 1+LINK_SIZE, /* KetRmax */ \
760 1+LINK_SIZE, /* KetRmin */ \
761 1+LINK_SIZE, /* Assert */ \
762 1+LINK_SIZE, /* Assert not */ \
763 1+LINK_SIZE, /* Assert behind */ \
764 1+LINK_SIZE, /* Assert behind not */ \
765 1+LINK_SIZE, /* Reverse */ \
766 1+LINK_SIZE, /* ONCE */ \
767 1+LINK_SIZE, /* BRA */ \
768 3+LINK_SIZE, /* CBRA */ \
769 1+LINK_SIZE, /* COND */ \
770 1+LINK_SIZE, /* SBRA */ \
771 3+LINK_SIZE, /* SCBRA */ \
772 1+LINK_SIZE, /* SCOND */ \
776 1, 1, /* BRAZERO, BRAMINZERO */ \
779 /* A magic value for OP_RREF to indicate the "any recursion" condition. */
781 #define RREF_ANY 0xffff
783 /* Error code numbers. They are given names so that they can more easily be
786 enum { ERR0, ERR1, ERR2, ERR3, ERR4, ERR5, ERR6, ERR7, ERR8, ERR9,
787 ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,
788 ERR20, ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR28, ERR29,
789 ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,
790 ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,
791 ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57 };
793 /* The real format of the start of the pcre block; the index of names and the
794 code vector run on as long as necessary after the end. We store an explicit
795 offset to the name table so that if a regex is compiled on one host, saved, and
796 then run on another where the size of pointers is different, all might still
797 be well. For the case of compiled-on-4 and run-on-8, we include an extra
798 pointer that is always NULL. For future-proofing, a few dummy fields were
799 originally included - even though you can never get this planning right - but
800 there is only one left now.
803 Because people can now save and re-use compiled patterns, any additions to this
804 structure should be made at the end, and something earlier (e.g. a new
805 flag in the options or one of the dummy fields) should indicate that the new
806 fields are present. Currently PCRE always sets the dummy fields to zero.
810 typedef struct real_pcre {
811 pcre_uint32 magic_number;
812 pcre_uint32 size; /* Total that was malloced */
814 pcre_uint32 dummy1; /* For future use, maybe */
816 pcre_uint16 top_bracket;
817 pcre_uint16 top_backref;
818 pcre_uint16 first_byte;
819 pcre_uint16 req_byte;
820 pcre_uint16 name_table_offset; /* Offset to name table that follows */
821 pcre_uint16 name_entry_size; /* Size of any name items */
822 pcre_uint16 name_count; /* Number of name items */
823 pcre_uint16 ref_count; /* Reference count */
825 const unsigned char *tables; /* Pointer to tables or NULL for std */
826 const unsigned char *nullpad; /* NULL padding */
829 /* The format of the block used to store data from pcre_study(). The same
830 remark (see NOTE above) about extending this structure applies. */
832 typedef struct pcre_study_data {
833 pcre_uint32 size; /* Total that was malloced */
835 uschar start_bits[32];
838 /* Structure for passing "static" information around between the functions
839 doing the compiling, so that they are thread-safe. */
841 typedef struct compile_data {
842 const uschar *lcc; /* Points to lower casing table */
843 const uschar *fcc; /* Points to case-flipping table */
844 const uschar *cbits; /* Points to character type table */
845 const uschar *ctypes; /* Points to table of type maps */
846 const uschar *start_workspace;/* The start of working space */
847 const uschar *start_code; /* The start of the compiled code */
848 const uschar *start_pattern; /* The start of the pattern */
849 const uschar *end_pattern; /* The end of the pattern */
850 uschar *hwm; /* High watermark of workspace */
851 uschar *name_table; /* The name/number table */
852 int names_found; /* Number of entries so far */
853 int name_entry_size; /* Size of each entry */
854 int bracount; /* Count of capturing parens */
855 int top_backref; /* Maximum back reference */
856 unsigned int backref_map; /* Bitmap of low back refs */
857 int external_options; /* External (initial) options */
858 int req_varyopt; /* "After variable item" flag for reqbyte */
859 BOOL nopartial; /* Set TRUE if partial won't work */
860 int nltype; /* Newline type */
861 int nllen; /* Newline string length */
862 uschar nl[4]; /* Newline string when fixed length */
865 /* Structure for maintaining a chain of pointers to the currently incomplete
866 branches, for testing for left recursion. */
868 typedef struct branch_chain {
869 struct branch_chain *outer;
873 /* Structure for items in a linked list that represents an explicit recursive
874 call within the pattern. */
876 typedef struct recursion_info {
877 struct recursion_info *prevrec; /* Previous recursion record (or NULL) */
878 int group_num; /* Number of group that was called */
879 const uschar *after_call; /* "Return value": points after the call in the expr */
880 USPTR save_start; /* Old value of md->start_match */
881 int *offset_save; /* Pointer to start of saved offsets */
882 int saved_max; /* Number of saved offsets */
885 /* When compiling in a mode that doesn't use recursive calls to match(),
886 a structure is used to remember local variables on the heap. It is defined in
887 pcre_exec.c, close to the match() function, so that it is easy to keep it in
888 step with any changes of local variable. However, the pointer to the current
889 frame must be saved in some "static" place over a longjmp(). We declare the
890 structure here so that we can put a pointer in the match_data structure. NOTE:
891 This isn't used for a "normal" compilation of pcre. */
895 /* Structure for building a chain of data for holding the values of the subject
896 pointer at the start of each subpattern, so as to detect when an empty string
897 has been matched by a subpattern - to break infinite loops. */
899 typedef struct eptrblock {
900 struct eptrblock *epb_prev;
901 USPTR epb_saved_eptr;
905 /* Structure for passing "static" information around between the functions
906 doing traditional NFA matching, so that they are thread-safe. */
908 typedef struct match_data {
909 unsigned long int match_call_count; /* As it says */
910 unsigned long int match_limit; /* As it says */
911 unsigned long int match_limit_recursion; /* As it says */
912 int *offset_vector; /* Offset vector */
913 int offset_end; /* One past the end */
914 int offset_max; /* The maximum usable for return data */
915 int nltype; /* Newline type */
916 int nllen; /* Newline string length */
917 uschar nl[4]; /* Newline string when fixed */
918 const uschar *lcc; /* Points to lower casing table */
919 const uschar *ctypes; /* Points to table of type maps */
920 BOOL offset_overflow; /* Set if too many extractions */
921 BOOL notbol; /* NOTBOL flag */
922 BOOL noteol; /* NOTEOL flag */
923 BOOL utf8; /* UTF8 flag */
924 BOOL endonly; /* Dollar not before final \n */
925 BOOL notempty; /* Empty string match not wanted */
926 BOOL partial; /* PARTIAL flag */
927 BOOL hitend; /* Hit the end of the subject at some point */
928 const uschar *start_code; /* For use when recursing */
929 USPTR start_subject; /* Start of the subject string */
930 USPTR end_subject; /* End of the subject string */
931 USPTR start_match; /* Start of this match attempt */
932 USPTR end_match_ptr; /* Subject position at end match */
933 int end_offset_top; /* Highwater mark at end of match */
934 int capture_last; /* Most recent capture number */
935 int start_offset; /* The start offset value */
936 eptrblock *eptrchain; /* Chain of eptrblocks for tail recursions */
937 int eptrn; /* Next free eptrblock */
938 recursion_info *recursive; /* Linked list of recursion data */
939 void *callout_data; /* To pass back to callouts */
940 struct heapframe *thisframe; /* Used only when compiling for no recursion */
943 /* A similar structure is used for the same purpose by the DFA matching
946 typedef struct dfa_match_data {
947 const uschar *start_code; /* Start of the compiled pattern */
948 const uschar *start_subject; /* Start of the subject string */
949 const uschar *end_subject; /* End of subject string */
950 const uschar *tables; /* Character tables */
951 int moptions; /* Match options */
952 int poptions; /* Pattern options */
953 int nltype; /* Newline type */
954 int nllen; /* Newline string length */
955 uschar nl[4]; /* Newline string when fixed */
956 void *callout_data; /* To pass back to callouts */
959 /* Bit definitions for entries in the pcre_ctypes table. */
961 #define ctype_space 0x01
962 #define ctype_letter 0x02
963 #define ctype_digit 0x04
964 #define ctype_xdigit 0x08
965 #define ctype_word 0x10 /* alphameric or '_' */
966 #define ctype_meta 0x80 /* regexp meta char or zero (end pattern) */
968 /* Offsets for the bitmap tables in pcre_cbits. Each table contains a set
969 of bits for a class map. Some classes are built by combining these tables. */
971 #define cbit_space 0 /* [:space:] or \s */
972 #define cbit_xdigit 32 /* [:xdigit:] */
973 #define cbit_digit 64 /* [:digit:] or \d */
974 #define cbit_upper 96 /* [:upper:] */
975 #define cbit_lower 128 /* [:lower:] */
976 #define cbit_word 160 /* [:word:] or \w */
977 #define cbit_graph 192 /* [:graph:] */
978 #define cbit_print 224 /* [:print:] */
979 #define cbit_punct 256 /* [:punct:] */
980 #define cbit_cntrl 288 /* [:cntrl:] */
981 #define cbit_length 320 /* Length of the cbits table */
983 /* Offsets of the various tables from the base tables pointer, and
987 #define fcc_offset 256
988 #define cbits_offset 512
989 #define ctypes_offset (cbits_offset + cbit_length)
990 #define tables_length (ctypes_offset + 256)
992 /* Layout of the UCP type table that translates property names into types and
1002 /* Internal shared data tables. These are tables that are used by more than one
1003 of the exported public functions. They have to be "external" in the C sense,
1004 but are not part of the PCRE public API. The data for these tables is in the
1005 pcre_tables.c module. */
1007 extern const int _pcre_utf8_table1[];
1008 extern const int _pcre_utf8_table2[];
1009 extern const int _pcre_utf8_table3[];
1010 extern const uschar _pcre_utf8_table4[];
1012 extern const int _pcre_utf8_table1_size;
1014 extern const char _pcre_ucp_names[];
1015 extern const ucp_type_table _pcre_utt[];
1016 extern const int _pcre_utt_size;
1018 extern const uschar _pcre_default_tables[];
1020 extern const uschar _pcre_OP_lengths[];
1023 /* Internal shared functions. These are functions that are used by more than
1024 one of the exported public functions. They have to be "external" in the C
1025 sense, but are not part of the PCRE public API. */
1027 extern BOOL _pcre_is_newline(const uschar *, const uschar *, int *,
1029 extern int _pcre_ord2utf8(int, uschar *);
1030 extern real_pcre *_pcre_try_flipped(const real_pcre *, real_pcre *,
1031 const pcre_study_data *, pcre_study_data *);
1032 extern int _pcre_ucp_findprop(const unsigned int, int *, int *);
1033 extern unsigned int _pcre_ucp_othercase(const unsigned int);
1034 extern int _pcre_valid_utf8(const uschar *, int);
1035 extern BOOL _pcre_was_newline(const uschar *, const uschar *, int *,
1037 extern BOOL _pcre_xclass(int, const uschar *);
1041 /* End of pcre_internal.h */