1 /*************************************************
2 * Perl-Compatible Regular Expressions *
3 *************************************************/
4
5
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.
8
9 Written by Philip Hazel
10 Copyright (c) 1997-2009 University of Cambridge
11
12 -----------------------------------------------------------------------------
13 Redistribution and use in source and binary forms, with or without
14 modification, are permitted provided that the following conditions are met:
15
16 * Redistributions of source code must retain the above copyright notice,
17 this list of conditions and the following disclaimer.
18
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.
22
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.
26
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 -----------------------------------------------------------------------------
39 */
40
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_". */
44
45 #ifndef PCRE_INTERNAL_H
46 #define PCRE_INTERNAL_H
47
48 /* Define DEBUG to get debugging output on stdout. */
49
50 #if 0
51 #define DEBUG
52 #endif
53
54 /* We do not support both EBCDIC and UTF-8 at the same time. The "configure"
55 script prevents both being selected, but not everybody uses "configure". */
56
57 #if defined EBCDIC && defined SUPPORT_UTF8
58 #error The use of both EBCDIC and SUPPORT_UTF8 is not supported.
59 #endif
60
61 /* If SUPPORT_UCP is defined, SUPPORT_UTF8 must also be defined. The
62 "configure" script ensures this, but not everybody uses "configure". */
63
64 #if defined SUPPORT_UCP && !defined SUPPORT_UTF8
65 #define SUPPORT_UTF8 1
66 #endif
67
68 /* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef
69 inline, and there are *still* stupid compilers about that don't like indented
70 pre-processor statements, or at least there were when I first wrote this. After
71 all, it had only been about 10 years then...
72
73 It turns out that the Mac Debugging.h header also defines the macro DPRINTF, so
74 be absolutely sure we get our version. */
75
76 #undef DPRINTF
77 #ifdef DEBUG
78 #define DPRINTF(p) printf p
79 #else
80 #define DPRINTF(p) /* Nothing */
81 #endif
82
83
84 /* Standard C headers plus the external interface definition. The only time
85 setjmp and stdarg are used is when NO_RECURSE is set. */
86
87 #include <ctype.h>
88 #include <limits.h>
89 #include <setjmp.h>
90 #include <stdarg.h>
91 #include <stddef.h>
92 #include <stdio.h>
93 #include <stdlib.h>
94 #include <string.h>
95
96 /* When compiling a DLL for Windows, the exported symbols have to be declared
97 using some MS magic. I found some useful information on this web page:
98 http://msdn2.microsoft.com/en-us/library/y4h7bcy6(VS.80).aspx. According to the
99 information there, using __declspec(dllexport) without "extern" we have a
100 definition; with "extern" we have a declaration. The settings here override the
101 setting in pcre.h (which is included below); it defines only PCRE_EXP_DECL,
102 which is all that is needed for applications (they just import the symbols). We
103 use:
104
105 PCRE_EXP_DECL for declarations
106 PCRE_EXP_DEFN for definitions of exported functions
107 PCRE_EXP_DATA_DEFN for definitions of exported variables
108
109 The reason for the two DEFN macros is that in non-Windows environments, one
110 does not want to have "extern" before variable definitions because it leads to
111 compiler warnings. So we distinguish between functions and variables. In
112 Windows, the two should always be the same.
113
114 The reason for wrapping this in #ifndef PCRE_EXP_DECL is so that pcretest,
115 which is an application, but needs to import this file in order to "peek" at
116 internals, can #include pcre.h first to get an application's-eye view.
117
118 In principle, people compiling for non-Windows, non-Unix-like (i.e. uncommon,
119 special-purpose environments) might want to stick other stuff in front of
120 exported symbols. That's why, in the non-Windows case, we set PCRE_EXP_DEFN and
121 PCRE_EXP_DATA_DEFN only if they are not already set. */
122
123 #ifndef PCRE_EXP_DECL
124 # ifdef _WIN32
125 # ifndef PCRE_STATIC
126 # define PCRE_EXP_DECL extern __declspec(dllexport)
127 # define PCRE_EXP_DEFN __declspec(dllexport)
128 # define PCRE_EXP_DATA_DEFN __declspec(dllexport)
129 # else
130 # define PCRE_EXP_DECL extern
131 # define PCRE_EXP_DEFN
132 # define PCRE_EXP_DATA_DEFN
133 # endif
134 # else
135 # ifdef __cplusplus
136 # define PCRE_EXP_DECL extern "C"
137 # else
138 # define PCRE_EXP_DECL extern
139 # endif
140 # ifndef PCRE_EXP_DEFN
141 # define PCRE_EXP_DEFN PCRE_EXP_DECL
142 # endif
143 # ifndef PCRE_EXP_DATA_DEFN
144 # define PCRE_EXP_DATA_DEFN
145 # endif
146 # endif
147 #endif
148
149 /* When compiling with the MSVC compiler, it is sometimes necessary to include
150 a "calling convention" before exported function names. (This is secondhand
151 information; I know nothing about MSVC myself). For example, something like
152
153 void __cdecl function(....)
154
155 might be needed. In order so make this easy, all the exported functions have
156 PCRE_CALL_CONVENTION just before their names. It is rarely needed; if not
157 set, we ensure here that it has no effect. */
158
159 #ifndef PCRE_CALL_CONVENTION
160 #define PCRE_CALL_CONVENTION
161 #endif
162
163 /* We need to have types that specify unsigned 16-bit and 32-bit integers. We
164 cannot determine these outside the compilation (e.g. by running a program as
165 part of "configure") because PCRE is often cross-compiled for use on other
166 systems. Instead we make use of the maximum sizes that are available at
167 preprocessor time in standard C environments. */
168
169 #if USHRT_MAX == 65535
170 typedef unsigned short pcre_uint16;
171 typedef short pcre_int16;
172 #elif UINT_MAX == 65535
173 typedef unsigned int pcre_uint16;
174 typedef int pcre_int16;
175 #else
176 #error Cannot determine a type for 16-bit unsigned integers
177 #endif
178
179 #if UINT_MAX == 4294967295
180 typedef unsigned int pcre_uint32;
181 typedef int pcre_int32;
182 #elif ULONG_MAX == 4294967295
183 typedef unsigned long int pcre_uint32;
184 typedef long int pcre_int32;
185 #else
186 #error Cannot determine a type for 32-bit unsigned integers
187 #endif
188
189 /* All character handling must be done as unsigned characters. Otherwise there
190 are problems with top-bit-set characters and functions such as isspace().
191 However, we leave the interface to the outside world as char *, because that
192 should make things easier for callers. We define a short type for unsigned char
193 to save lots of typing. I tried "uchar", but it causes problems on Digital
194 Unix, where it is defined in sys/types, so use "uschar" instead. */
195
196 typedef unsigned char uschar;
197
198 /* This is an unsigned int value that no character can ever have. UTF-8
199 characters only go up to 0x7fffffff (though Unicode doesn't go beyond
200 0x0010ffff). */
201
202 #define NOTACHAR 0xffffffff
203
204 /* PCRE is able to support several different kinds of newline (CR, LF, CRLF,
205 "any" and "anycrlf" at present). The following macros are used to package up
206 testing for newlines. NLBLOCK, PSSTART, and PSEND are defined in the various
207 modules to indicate in which datablock the parameters exist, and what the
208 start/end of string field names are. */
209
210 #define NLTYPE_FIXED 0 /* Newline is a fixed length string */
211 #define NLTYPE_ANY 1 /* Newline is any Unicode line ending */
212 #define NLTYPE_ANYCRLF 2 /* Newline is CR, LF, or CRLF */
213
214 /* This macro checks for a newline at the given position */
215
216 #define IS_NEWLINE(p) \
217 ((NLBLOCK->nltype != NLTYPE_FIXED)? \
218 ((p) < NLBLOCK->PSEND && \
219 _pcre_is_newline((p), NLBLOCK->nltype, NLBLOCK->PSEND, &(NLBLOCK->nllen),\
220 utf8)) \
221 : \
222 ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \
223 (p)[0] == NLBLOCK->nl[0] && \
224 (NLBLOCK->nllen == 1 || (p)[1] == NLBLOCK->nl[1]) \
225 ) \
226 )
227
228 /* This macro checks for a newline immediately preceding the given position */
229
230 #define WAS_NEWLINE(p) \
231 ((NLBLOCK->nltype != NLTYPE_FIXED)? \
232 ((p) > NLBLOCK->PSSTART && \
233 _pcre_was_newline((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \
234 &(NLBLOCK->nllen), utf8)) \
235 : \
236 ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \
237 (p)[-NLBLOCK->nllen] == NLBLOCK->nl[0] && \
238 (NLBLOCK->nllen == 1 || (p)[-NLBLOCK->nllen+1] == NLBLOCK->nl[1]) \
239 ) \
240 )
241
242 /* When PCRE is compiled as a C++ library, the subject pointer can be replaced
243 with a custom type. This makes it possible, for example, to allow pcre_exec()
244 to process subject strings that are discontinuous by using a smart pointer
245 class. It must always be possible to inspect all of the subject string in
246 pcre_exec() because of the way it backtracks. Two macros are required in the
247 normal case, for sign-unspecified and unsigned char pointers. The former is
248 used for the external interface and appears in pcre.h, which is why its name
249 must begin with PCRE_. */
250
251 #ifdef CUSTOM_SUBJECT_PTR
252 #define PCRE_SPTR CUSTOM_SUBJECT_PTR
253 #define USPTR CUSTOM_SUBJECT_PTR
254 #else
255 #define PCRE_SPTR const char *
256 #define USPTR const unsigned char *
257 #endif
258
259
260
261 /* Include the public PCRE header and the definitions of UCP character property
262 values. */
263
264 /* Rename API to avoid conflict with existing PCRE installations */
265 #include "pcreemboss.h"
266
267 #include "pcre.h"
268 #include "ucp.h"
269
270 /* When compiling for use with the Virtual Pascal compiler, these functions
271 need to have their names changed. PCRE must be compiled with the -DVPCOMPAT
272 option on the command line. */
273
274 #ifdef VPCOMPAT
275 #define strlen(s) _strlen(s)
276 #define strncmp(s1,s2,m) _strncmp(s1,s2,m)
277 #define memcmp(s,c,n) _memcmp(s,c,n)
278 #define memcpy(d,s,n) _memcpy(d,s,n)
279 #define memmove(d,s,n) _memmove(d,s,n)
280 #define memset(s,c,n) _memset(s,c,n)
281 #else /* VPCOMPAT */
282
283 /* To cope with SunOS4 and other systems that lack memmove() but have bcopy(),
284 define a macro for memmove() if HAVE_MEMMOVE is false, provided that HAVE_BCOPY
285 is set. Otherwise, include an emulating function for those systems that have
286 neither (there some non-Unix environments where this is the case). */
287
288 #ifndef HAVE_MEMMOVE
289 #undef memmove /* some systems may have a macro */
290 #ifdef HAVE_BCOPY
291 #define memmove(a, b, c) bcopy(b, a, c)
292 #else /* HAVE_BCOPY */
293 static void *
pcre_memmove(void * d,const void * s,size_t n)294 pcre_memmove(void *d, const void *s, size_t n)
295 {
296 size_t i;
297 unsigned char *dest = (unsigned char *)d;
298 const unsigned char *src = (const unsigned char *)s;
299 if (dest > src)
300 {
301 dest += n;
302 src += n;
303 for (i = 0; i < n; ++i) *(--dest) = *(--src);
304 return (void *)dest;
305 }
306 else
307 {
308 for (i = 0; i < n; ++i) *dest++ = *src++;
309 return (void *)(dest - n);
310 }
311 }
312 #define memmove(a, b, c) pcre_memmove(a, b, c)
313 #endif /* not HAVE_BCOPY */
314 #endif /* not HAVE_MEMMOVE */
315 #endif /* not VPCOMPAT */
316
317
318 /* PCRE keeps offsets in its compiled code as 2-byte quantities (always stored
319 in big-endian order) by default. These are used, for example, to link from the
320 start of a subpattern to its alternatives and its end. The use of 2 bytes per
321 offset limits the size of the compiled regex to around 64K, which is big enough
322 for almost everybody. However, I received a request for an even bigger limit.
323 For this reason, and also to make the code easier to maintain, the storing and
324 loading of offsets from the byte string is now handled by the macros that are
325 defined here.
326
327 The macros are controlled by the value of LINK_SIZE. This defaults to 2 in
328 the config.h file, but can be overridden by using -D on the command line. This
329 is automated on Unix systems via the "configure" command. */
330
331 #if LINK_SIZE == 2
332
333 #define PUT(a,n,d) \
334 (a[n] = (d) >> 8), \
335 (a[(n)+1] = (d) & 255)
336
337 #define GET(a,n) \
338 (((a)[n] << 8) | (a)[(n)+1])
339
340 #define MAX_PATTERN_SIZE (1 << 16)
341
342
343 #elif LINK_SIZE == 3
344
345 #define PUT(a,n,d) \
346 (a[n] = (d) >> 16), \
347 (a[(n)+1] = (d) >> 8), \
348 (a[(n)+2] = (d) & 255)
349
350 #define GET(a,n) \
351 (((a)[n] << 16) | ((a)[(n)+1] << 8) | (a)[(n)+2])
352
353 #define MAX_PATTERN_SIZE (1 << 24)
354
355
356 #elif LINK_SIZE == 4
357
358 #define PUT(a,n,d) \
359 (a[n] = (d) >> 24), \
360 (a[(n)+1] = (d) >> 16), \
361 (a[(n)+2] = (d) >> 8), \
362 (a[(n)+3] = (d) & 255)
363
364 #define GET(a,n) \
365 (((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3])
366
367 #define MAX_PATTERN_SIZE (1 << 30) /* Keep it positive */
368
369
370 #else
371 #error LINK_SIZE must be either 2, 3, or 4
372 #endif
373
374
375 /* Convenience macro defined in terms of the others */
376
377 #define PUTINC(a,n,d) PUT(a,n,d), a += LINK_SIZE
378
379
380 /* PCRE uses some other 2-byte quantities that do not change when the size of
381 offsets changes. There are used for repeat counts and for other things such as
382 capturing parenthesis numbers in back references. */
383
384 #define PUT2(a,n,d) \
385 a[n] = (d) >> 8; \
386 a[(n)+1] = (d) & 255
387
388 #define GET2(a,n) \
389 (((a)[n] << 8) | (a)[(n)+1])
390
391 #define PUT2INC(a,n,d) PUT2(a,n,d), a += 2
392
393
394 /* When UTF-8 encoding is being used, a character is no longer just a single
395 byte. The macros for character handling generate simple sequences when used in
396 byte-mode, and more complicated ones for UTF-8 characters. BACKCHAR should
397 never be called in byte mode. To make sure it can never even appear when UTF-8
398 support is omitted, we don't even define it. */
399
400 #ifndef SUPPORT_UTF8
401 #define GETCHAR(c, eptr) c = *eptr;
402 #define GETCHARTEST(c, eptr) c = *eptr;
403 #define GETCHARINC(c, eptr) c = *eptr++;
404 #define GETCHARINCTEST(c, eptr) c = *eptr++;
405 #define GETCHARLEN(c, eptr, len) c = *eptr;
406 /* #define BACKCHAR(eptr) */
407
408 #else /* SUPPORT_UTF8 */
409
410 /* Get the next UTF-8 character, not advancing the pointer. This is called when
411 we know we are in UTF-8 mode. */
412
413 #define GETCHAR(c, eptr) \
414 c = *eptr; \
415 if (c >= 0xc0) \
416 { \
417 int gcii; \
418 int gcaa = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */ \
419 int gcss = 6*gcaa; \
420 c = (c & _pcre_utf8_table3[gcaa]) << gcss; \
421 for (gcii = 1; gcii <= gcaa; gcii++) \
422 { \
423 gcss -= 6; \
424 c |= (eptr[gcii] & 0x3f) << gcss; \
425 } \
426 }
427
428 /* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the
429 pointer. */
430
431 #define GETCHARTEST(c, eptr) \
432 c = *eptr; \
433 if (utf8 && c >= 0xc0) \
434 { \
435 int gcii; \
436 int gcaa = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */ \
437 int gcss = 6*gcaa; \
438 c = (c & _pcre_utf8_table3[gcaa]) << gcss; \
439 for (gcii = 1; gcii <= gcaa; gcii++) \
440 { \
441 gcss -= 6; \
442 c |= (eptr[gcii] & 0x3f) << gcss; \
443 } \
444 }
445
446 /* Get the next UTF-8 character, advancing the pointer. This is called when we
447 know we are in UTF-8 mode. */
448
449 #define GETCHARINC(c, eptr) \
450 c = *eptr++; \
451 if (c >= 0xc0) \
452 { \
453 int gcaa = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */ \
454 int gcss = 6*gcaa; \
455 c = (c & _pcre_utf8_table3[gcaa]) << gcss; \
456 while (gcaa-- > 0) \
457 { \
458 gcss -= 6; \
459 c |= (*eptr++ & 0x3f) << gcss; \
460 } \
461 }
462
463 /* Get the next character, testing for UTF-8 mode, and advancing the pointer */
464
465 #define GETCHARINCTEST(c, eptr) \
466 c = *eptr++; \
467 if (utf8 && c >= 0xc0) \
468 { \
469 int gcaa = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */ \
470 int gcss = 6*gcaa; \
471 c = (c & _pcre_utf8_table3[gcaa]) << gcss; \
472 while (gcaa-- > 0) \
473 { \
474 gcss -= 6; \
475 c |= (*eptr++ & 0x3f) << gcss; \
476 } \
477 }
478
479 /* Get the next UTF-8 character, not advancing the pointer, incrementing length
480 if there are extra bytes. This is called when we know we are in UTF-8 mode. */
481
482 #define GETCHARLEN(c, eptr, len) \
483 c = *eptr; \
484 if (c >= 0xc0) \
485 { \
486 int gcii; \
487 int gcaa = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */ \
488 int gcss = 6*gcaa; \
489 c = (c & _pcre_utf8_table3[gcaa]) << gcss; \
490 for (gcii = 1; gcii <= gcaa; gcii++) \
491 { \
492 gcss -= 6; \
493 c |= (eptr[gcii] & 0x3f) << gcss; \
494 } \
495 len += gcaa; \
496 }
497
498 /* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the
499 pointer, incrementing length if there are extra bytes. This is called when we
500 know we are in UTF-8 mode. */
501
502 #define GETCHARLENTEST(c, eptr, len) \
503 c = *eptr; \
504 if (utf8 && c >= 0xc0) \
505 { \
506 int gcii; \
507 int gcaa = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */ \
508 int gcss = 6*gcaa; \
509 c = (c & _pcre_utf8_table3[gcaa]) << gcss; \
510 for (gcii = 1; gcii <= gcaa; gcii++) \
511 { \
512 gcss -= 6; \
513 c |= (eptr[gcii] & 0x3f) << gcss; \
514 } \
515 len += gcaa; \
516 }
517
518 /* If the pointer is not at the start of a character, move it back until
519 it is. This is called only in UTF-8 mode - we don't put a test within the macro
520 because almost all calls are already within a block of UTF-8 only code. */
521
522 #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--
523
524 #endif
525
526
527 /* In case there is no definition of offsetof() provided - though any proper
528 Standard C system should have one. */
529
530 #ifndef offsetof
531 #define offsetof(p_type,field) ((size_t)&(((p_type *)0)->field))
532 #endif
533
534
535 /* These are the public options that can change during matching. */
536
537 #define PCRE_IMS (PCRE_CASELESS|PCRE_MULTILINE|PCRE_DOTALL)
538
539 /* Private flags containing information about the compiled regex. They used to
540 live at the top end of the options word, but that got almost full, so now they
541 are in a 16-bit flags word. */
542
543 #define PCRE_NOPARTIAL 0x0001 /* can't use partial with this regex */
544 #define PCRE_FIRSTSET 0x0002 /* first_byte is set */
545 #define PCRE_REQCHSET 0x0004 /* req_byte is set */
546 #define PCRE_STARTLINE 0x0008 /* start after \n for multiline */
547 #define PCRE_JCHANGED 0x0010 /* j option used in regex */
548 #define PCRE_HASCRORLF 0x0020 /* explicit \r or \n in pattern */
549
550 /* Options for the "extra" block produced by pcre_study(). */
551
552 #define PCRE_STUDY_MAPPED 0x01 /* a map of starting chars exists */
553
554 /* Masks for identifying the public options that are permitted at compile
555 time, run time, or study time, respectively. */
556
557 #define PCRE_NEWLINE_BITS (PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_ANY| \
558 PCRE_NEWLINE_ANYCRLF)
559
560 #define PUBLIC_COMPILE_OPTIONS \
561 (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \
562 PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \
563 PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \
564 PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
565 PCRE_JAVASCRIPT_COMPAT)
566
567 #define PUBLIC_EXEC_OPTIONS \
568 (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \
569 PCRE_PARTIAL|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
570 PCRE_NO_START_OPTIMIZE)
571
572 #define PUBLIC_DFA_EXEC_OPTIONS \
573 (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \
574 PCRE_PARTIAL|PCRE_DFA_SHORTEST|PCRE_DFA_RESTART|PCRE_NEWLINE_BITS| \
575 PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE|PCRE_NO_START_OPTIMIZE)
576
577 #define PUBLIC_STUDY_OPTIONS 0 /* None defined */
578
579 /* Magic number to provide a small check against being handed junk. Also used
580 to detect whether a pattern was compiled on a host of different endianness. */
581
582 #define MAGIC_NUMBER 0x50435245UL /* 'PCRE' */
583
584 /* Negative values for the firstchar and reqchar variables */
585
586 #define REQ_UNSET (-2)
587 #define REQ_NONE (-1)
588
589 /* The maximum remaining length of subject we are prepared to search for a
590 req_byte match. */
591
592 #define REQ_BYTE_MAX 1000
593
594 /* Flags added to firstbyte or reqbyte; a "non-literal" item is either a
595 variable-length repeat, or a anything other than literal characters. */
596
597 #define REQ_CASELESS 0x0100 /* indicates caselessness */
598 #define REQ_VARY 0x0200 /* reqbyte followed non-literal item */
599
600 /* Miscellaneous definitions. The #ifndef is to pacify compiler warnings in
601 environments where these macros are defined elsewhere. Unfortunately, there
602 is no way to do the same for the typedef. */
603
604 typedef int BOOL;
605
606 #ifndef FALSE
607 #define FALSE 0
608 #define TRUE 1
609 #endif
610
611 /* If PCRE is to support UTF-8 on EBCDIC platforms, we cannot use normal
612 character constants like '*' because the compiler would emit their EBCDIC code,
613 which is different from their ASCII/UTF-8 code. Instead we define macros for
614 the characters so that they always use the ASCII/UTF-8 code when UTF-8 support
615 is enabled. When UTF-8 support is not enabled, the definitions use character
616 literals. Both character and string versions of each character are needed, and
617 there are some longer strings as well.
618
619 This means that, on EBCDIC platforms, the PCRE library can handle either
620 EBCDIC, or UTF-8, but not both. To support both in the same compiled library
621 would need different lookups depending on whether PCRE_UTF8 was set or not.
622 This would make it impossible to use characters in switch/case statements,
623 which would reduce performance. For a theoretical use (which nobody has asked
624 for) in a minority area (EBCDIC platforms), this is not sensible. Any
625 application that did need both could compile two versions of the library, using
626 macros to give the functions distinct names. */
627
628 #ifndef SUPPORT_UTF8
629
630 /* UTF-8 support is not enabled; use the platform-dependent character literals
631 so that PCRE works on both ASCII and EBCDIC platforms, in non-UTF-mode only. */
632
633 #define CHAR_HT '\t'
634 #define CHAR_VT '\v'
635 #define CHAR_FF '\f'
636 #define CHAR_CR '\r'
637 #define CHAR_NL '\n'
638 #define CHAR_BS '\b'
639 #define CHAR_BEL '\a'
640 #ifdef EBCDIC
641 #define CHAR_ESC '\047'
642 #define CHAR_DEL '\007'
643 #else
644 #define CHAR_ESC '\033'
645 #define CHAR_DEL '\177'
646 #endif
647
648 #define CHAR_SPACE ' '
649 #define CHAR_EXCLAMATION_MARK '!'
650 #define CHAR_QUOTATION_MARK '"'
651 #define CHAR_NUMBER_SIGN '#'
652 #define CHAR_DOLLAR_SIGN '$'
653 #define CHAR_PERCENT_SIGN '%'
654 #define CHAR_AMPERSAND '&'
655 #define CHAR_APOSTROPHE '\''
656 #define CHAR_LEFT_PARENTHESIS '('
657 #define CHAR_RIGHT_PARENTHESIS ')'
658 #define CHAR_ASTERISK '*'
659 #define CHAR_PLUS '+'
660 #define CHAR_COMMA ','
661 #define CHAR_MINUS '-'
662 #define CHAR_DOT '.'
663 #define CHAR_SLASH '/'
664 #define CHAR_0 '0'
665 #define CHAR_1 '1'
666 #define CHAR_2 '2'
667 #define CHAR_3 '3'
668 #define CHAR_4 '4'
669 #define CHAR_5 '5'
670 #define CHAR_6 '6'
671 #define CHAR_7 '7'
672 #define CHAR_8 '8'
673 #define CHAR_9 '9'
674 #define CHAR_COLON ':'
675 #define CHAR_SEMICOLON ';'
676 #define CHAR_LESS_THAN_SIGN '<'
677 #define CHAR_EQUALS_SIGN '='
678 #define CHAR_GREATER_THAN_SIGN '>'
679 #define CHAR_QUESTION_MARK '?'
680 #define CHAR_COMMERCIAL_AT '@'
681 #define CHAR_A 'A'
682 #define CHAR_B 'B'
683 #define CHAR_C 'C'
684 #define CHAR_D 'D'
685 #define CHAR_E 'E'
686 #define CHAR_F 'F'
687 #define CHAR_G 'G'
688 #define CHAR_H 'H'
689 #define CHAR_I 'I'
690 #define CHAR_J 'J'
691 #define CHAR_K 'K'
692 #define CHAR_L 'L'
693 #define CHAR_M 'M'
694 #define CHAR_N 'N'
695 #define CHAR_O 'O'
696 #define CHAR_P 'P'
697 #define CHAR_Q 'Q'
698 #define CHAR_R 'R'
699 #define CHAR_S 'S'
700 #define CHAR_T 'T'
701 #define CHAR_U 'U'
702 #define CHAR_V 'V'
703 #define CHAR_W 'W'
704 #define CHAR_X 'X'
705 #define CHAR_Y 'Y'
706 #define CHAR_Z 'Z'
707 #define CHAR_LEFT_SQUARE_BRACKET '['
708 #define CHAR_BACKSLASH '\\'
709 #define CHAR_RIGHT_SQUARE_BRACKET ']'
710 #define CHAR_CIRCUMFLEX_ACCENT '^'
711 #define CHAR_UNDERSCORE '_'
712 #define CHAR_GRAVE_ACCENT '`'
713 #define CHAR_a 'a'
714 #define CHAR_b 'b'
715 #define CHAR_c 'c'
716 #define CHAR_d 'd'
717 #define CHAR_e 'e'
718 #define CHAR_f 'f'
719 #define CHAR_g 'g'
720 #define CHAR_h 'h'
721 #define CHAR_i 'i'
722 #define CHAR_j 'j'
723 #define CHAR_k 'k'
724 #define CHAR_l 'l'
725 #define CHAR_m 'm'
726 #define CHAR_n 'n'
727 #define CHAR_o 'o'
728 #define CHAR_p 'p'
729 #define CHAR_q 'q'
730 #define CHAR_r 'r'
731 #define CHAR_s 's'
732 #define CHAR_t 't'
733 #define CHAR_u 'u'
734 #define CHAR_v 'v'
735 #define CHAR_w 'w'
736 #define CHAR_x 'x'
737 #define CHAR_y 'y'
738 #define CHAR_z 'z'
739 #define CHAR_LEFT_CURLY_BRACKET '{'
740 #define CHAR_VERTICAL_LINE '|'
741 #define CHAR_RIGHT_CURLY_BRACKET '}'
742 #define CHAR_TILDE '~'
743
744 #define STR_HT "\t"
745 #define STR_VT "\v"
746 #define STR_FF "\f"
747 #define STR_CR "\r"
748 #define STR_NL "\n"
749 #define STR_BS "\b"
750 #define STR_BEL "\a"
751 #ifdef EBCDIC
752 #define STR_ESC "\047"
753 #define STR_DEL "\007"
754 #else
755 #define STR_ESC "\033"
756 #define STR_DEL "\177"
757 #endif
758
759 #define STR_SPACE " "
760 #define STR_EXCLAMATION_MARK "!"
761 #define STR_QUOTATION_MARK "\""
762 #define STR_NUMBER_SIGN "#"
763 #define STR_DOLLAR_SIGN "$"
764 #define STR_PERCENT_SIGN "%"
765 #define STR_AMPERSAND "&"
766 #define STR_APOSTROPHE "'"
767 #define STR_LEFT_PARENTHESIS "("
768 #define STR_RIGHT_PARENTHESIS ")"
769 #define STR_ASTERISK "*"
770 #define STR_PLUS "+"
771 #define STR_COMMA ","
772 #define STR_MINUS "-"
773 #define STR_DOT "."
774 #define STR_SLASH "/"
775 #define STR_0 "0"
776 #define STR_1 "1"
777 #define STR_2 "2"
778 #define STR_3 "3"
779 #define STR_4 "4"
780 #define STR_5 "5"
781 #define STR_6 "6"
782 #define STR_7 "7"
783 #define STR_8 "8"
784 #define STR_9 "9"
785 #define STR_COLON ":"
786 #define STR_SEMICOLON ";"
787 #define STR_LESS_THAN_SIGN "<"
788 #define STR_EQUALS_SIGN "="
789 #define STR_GREATER_THAN_SIGN ">"
790 #define STR_QUESTION_MARK "?"
791 #define STR_COMMERCIAL_AT "@"
792 #define STR_A "A"
793 #define STR_B "B"
794 #define STR_C "C"
795 #define STR_D "D"
796 #define STR_E "E"
797 #define STR_F "F"
798 #define STR_G "G"
799 #define STR_H "H"
800 #define STR_I "I"
801 #define STR_J "J"
802 #define STR_K "K"
803 #define STR_L "L"
804 #define STR_M "M"
805 #define STR_N "N"
806 #define STR_O "O"
807 #define STR_P "P"
808 #define STR_Q "Q"
809 #define STR_R "R"
810 #define STR_S "S"
811 #define STR_T "T"
812 #define STR_U "U"
813 #define STR_V "V"
814 #define STR_W "W"
815 #define STR_X "X"
816 #define STR_Y "Y"
817 #define STR_Z "Z"
818 #define STR_LEFT_SQUARE_BRACKET "["
819 #define STR_BACKSLASH "\\"
820 #define STR_RIGHT_SQUARE_BRACKET "]"
821 #define STR_CIRCUMFLEX_ACCENT "^"
822 #define STR_UNDERSCORE "_"
823 #define STR_GRAVE_ACCENT "`"
824 #define STR_a "a"
825 #define STR_b "b"
826 #define STR_c "c"
827 #define STR_d "d"
828 #define STR_e "e"
829 #define STR_f "f"
830 #define STR_g "g"
831 #define STR_h "h"
832 #define STR_i "i"
833 #define STR_j "j"
834 #define STR_k "k"
835 #define STR_l "l"
836 #define STR_m "m"
837 #define STR_n "n"
838 #define STR_o "o"
839 #define STR_p "p"
840 #define STR_q "q"
841 #define STR_r "r"
842 #define STR_s "s"
843 #define STR_t "t"
844 #define STR_u "u"
845 #define STR_v "v"
846 #define STR_w "w"
847 #define STR_x "x"
848 #define STR_y "y"
849 #define STR_z "z"
850 #define STR_LEFT_CURLY_BRACKET "{"
851 #define STR_VERTICAL_LINE "|"
852 #define STR_RIGHT_CURLY_BRACKET "}"
853 #define STR_TILDE "~"
854
855 #define STRING_ACCEPT0 "ACCEPT\0"
856 #define STRING_COMMIT0 "COMMIT\0"
857 #define STRING_F0 "F\0"
858 #define STRING_FAIL0 "FAIL\0"
859 #define STRING_PRUNE0 "PRUNE\0"
860 #define STRING_SKIP0 "SKIP\0"
861 #define STRING_THEN "THEN"
862
863 #define STRING_alpha0 "alpha\0"
864 #define STRING_lower0 "lower\0"
865 #define STRING_upper0 "upper\0"
866 #define STRING_alnum0 "alnum\0"
867 #define STRING_ascii0 "ascii\0"
868 #define STRING_blank0 "blank\0"
869 #define STRING_cntrl0 "cntrl\0"
870 #define STRING_digit0 "digit\0"
871 #define STRING_graph0 "graph\0"
872 #define STRING_print0 "print\0"
873 #define STRING_punct0 "punct\0"
874 #define STRING_space0 "space\0"
875 #define STRING_word0 "word\0"
876 #define STRING_xdigit "xdigit"
877
878 #define STRING_DEFINE "DEFINE"
879
880 #define STRING_CR_RIGHTPAR "CR)"
881 #define STRING_LF_RIGHTPAR "LF)"
882 #define STRING_CRLF_RIGHTPAR "CRLF)"
883 #define STRING_ANY_RIGHTPAR "ANY)"
884 #define STRING_ANYCRLF_RIGHTPAR "ANYCRLF)"
885 #define STRING_BSR_ANYCRLF_RIGHTPAR "BSR_ANYCRLF)"
886 #define STRING_BSR_UNICODE_RIGHTPAR "BSR_UNICODE)"
887 #define STRING_UTF8_RIGHTPAR "UTF8)"
888
889 #else /* SUPPORT_UTF8 */
890
891 /* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This
892 works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode
893 only. */
894
895 #define CHAR_HT '\011'
896 #define CHAR_VT '\013'
897 #define CHAR_FF '\014'
898 #define CHAR_CR '\015'
899 #define CHAR_NL '\012'
900 #define CHAR_BS '\010'
901 #define CHAR_BEL '\007'
902 #define CHAR_ESC '\033'
903 #define CHAR_DEL '\177'
904
905 #define CHAR_SPACE '\040'
906 #define CHAR_EXCLAMATION_MARK '\041'
907 #define CHAR_QUOTATION_MARK '\042'
908 #define CHAR_NUMBER_SIGN '\043'
909 #define CHAR_DOLLAR_SIGN '\044'
910 #define CHAR_PERCENT_SIGN '\045'
911 #define CHAR_AMPERSAND '\046'
912 #define CHAR_APOSTROPHE '\047'
913 #define CHAR_LEFT_PARENTHESIS '\050'
914 #define CHAR_RIGHT_PARENTHESIS '\051'
915 #define CHAR_ASTERISK '\052'
916 #define CHAR_PLUS '\053'
917 #define CHAR_COMMA '\054'
918 #define CHAR_MINUS '\055'
919 #define CHAR_DOT '\056'
920 #define CHAR_SLASH '\057'
921 #define CHAR_0 '\060'
922 #define CHAR_1 '\061'
923 #define CHAR_2 '\062'
924 #define CHAR_3 '\063'
925 #define CHAR_4 '\064'
926 #define CHAR_5 '\065'
927 #define CHAR_6 '\066'
928 #define CHAR_7 '\067'
929 #define CHAR_8 '\070'
930 #define CHAR_9 '\071'
931 #define CHAR_COLON '\072'
932 #define CHAR_SEMICOLON '\073'
933 #define CHAR_LESS_THAN_SIGN '\074'
934 #define CHAR_EQUALS_SIGN '\075'
935 #define CHAR_GREATER_THAN_SIGN '\076'
936 #define CHAR_QUESTION_MARK '\077'
937 #define CHAR_COMMERCIAL_AT '\100'
938 #define CHAR_A '\101'
939 #define CHAR_B '\102'
940 #define CHAR_C '\103'
941 #define CHAR_D '\104'
942 #define CHAR_E '\105'
943 #define CHAR_F '\106'
944 #define CHAR_G '\107'
945 #define CHAR_H '\110'
946 #define CHAR_I '\111'
947 #define CHAR_J '\112'
948 #define CHAR_K '\113'
949 #define CHAR_L '\114'
950 #define CHAR_M '\115'
951 #define CHAR_N '\116'
952 #define CHAR_O '\117'
953 #define CHAR_P '\120'
954 #define CHAR_Q '\121'
955 #define CHAR_R '\122'
956 #define CHAR_S '\123'
957 #define CHAR_T '\124'
958 #define CHAR_U '\125'
959 #define CHAR_V '\126'
960 #define CHAR_W '\127'
961 #define CHAR_X '\130'
962 #define CHAR_Y '\131'
963 #define CHAR_Z '\132'
964 #define CHAR_LEFT_SQUARE_BRACKET '\133'
965 #define CHAR_BACKSLASH '\134'
966 #define CHAR_RIGHT_SQUARE_BRACKET '\135'
967 #define CHAR_CIRCUMFLEX_ACCENT '\136'
968 #define CHAR_UNDERSCORE '\137'
969 #define CHAR_GRAVE_ACCENT '\140'
970 #define CHAR_a '\141'
971 #define CHAR_b '\142'
972 #define CHAR_c '\143'
973 #define CHAR_d '\144'
974 #define CHAR_e '\145'
975 #define CHAR_f '\146'
976 #define CHAR_g '\147'
977 #define CHAR_h '\150'
978 #define CHAR_i '\151'
979 #define CHAR_j '\152'
980 #define CHAR_k '\153'
981 #define CHAR_l '\154'
982 #define CHAR_m '\155'
983 #define CHAR_n '\156'
984 #define CHAR_o '\157'
985 #define CHAR_p '\160'
986 #define CHAR_q '\161'
987 #define CHAR_r '\162'
988 #define CHAR_s '\163'
989 #define CHAR_t '\164'
990 #define CHAR_u '\165'
991 #define CHAR_v '\166'
992 #define CHAR_w '\167'
993 #define CHAR_x '\170'
994 #define CHAR_y '\171'
995 #define CHAR_z '\172'
996 #define CHAR_LEFT_CURLY_BRACKET '\173'
997 #define CHAR_VERTICAL_LINE '\174'
998 #define CHAR_RIGHT_CURLY_BRACKET '\175'
999 #define CHAR_TILDE '\176'
1000
1001 #define STR_HT "\011"
1002 #define STR_VT "\013"
1003 #define STR_FF "\014"
1004 #define STR_CR "\015"
1005 #define STR_NL "\012"
1006 #define STR_BS "\010"
1007 #define STR_BEL "\007"
1008 #define STR_ESC "\033"
1009 #define STR_DEL "\177"
1010
1011 #define STR_SPACE "\040"
1012 #define STR_EXCLAMATION_MARK "\041"
1013 #define STR_QUOTATION_MARK "\042"
1014 #define STR_NUMBER_SIGN "\043"
1015 #define STR_DOLLAR_SIGN "\044"
1016 #define STR_PERCENT_SIGN "\045"
1017 #define STR_AMPERSAND "\046"
1018 #define STR_APOSTROPHE "\047"
1019 #define STR_LEFT_PARENTHESIS "\050"
1020 #define STR_RIGHT_PARENTHESIS "\051"
1021 #define STR_ASTERISK "\052"
1022 #define STR_PLUS "\053"
1023 #define STR_COMMA "\054"
1024 #define STR_MINUS "\055"
1025 #define STR_DOT "\056"
1026 #define STR_SLASH "\057"
1027 #define STR_0 "\060"
1028 #define STR_1 "\061"
1029 #define STR_2 "\062"
1030 #define STR_3 "\063"
1031 #define STR_4 "\064"
1032 #define STR_5 "\065"
1033 #define STR_6 "\066"
1034 #define STR_7 "\067"
1035 #define STR_8 "\070"
1036 #define STR_9 "\071"
1037 #define STR_COLON "\072"
1038 #define STR_SEMICOLON "\073"
1039 #define STR_LESS_THAN_SIGN "\074"
1040 #define STR_EQUALS_SIGN "\075"
1041 #define STR_GREATER_THAN_SIGN "\076"
1042 #define STR_QUESTION_MARK "\077"
1043 #define STR_COMMERCIAL_AT "\100"
1044 #define STR_A "\101"
1045 #define STR_B "\102"
1046 #define STR_C "\103"
1047 #define STR_D "\104"
1048 #define STR_E "\105"
1049 #define STR_F "\106"
1050 #define STR_G "\107"
1051 #define STR_H "\110"
1052 #define STR_I "\111"
1053 #define STR_J "\112"
1054 #define STR_K "\113"
1055 #define STR_L "\114"
1056 #define STR_M "\115"
1057 #define STR_N "\116"
1058 #define STR_O "\117"
1059 #define STR_P "\120"
1060 #define STR_Q "\121"
1061 #define STR_R "\122"
1062 #define STR_S "\123"
1063 #define STR_T "\124"
1064 #define STR_U "\125"
1065 #define STR_V "\126"
1066 #define STR_W "\127"
1067 #define STR_X "\130"
1068 #define STR_Y "\131"
1069 #define STR_Z "\132"
1070 #define STR_LEFT_SQUARE_BRACKET "\133"
1071 #define STR_BACKSLASH "\134"
1072 #define STR_RIGHT_SQUARE_BRACKET "\135"
1073 #define STR_CIRCUMFLEX_ACCENT "\136"
1074 #define STR_UNDERSCORE "\137"
1075 #define STR_GRAVE_ACCENT "\140"
1076 #define STR_a "\141"
1077 #define STR_b "\142"
1078 #define STR_c "\143"
1079 #define STR_d "\144"
1080 #define STR_e "\145"
1081 #define STR_f "\146"
1082 #define STR_g "\147"
1083 #define STR_h "\150"
1084 #define STR_i "\151"
1085 #define STR_j "\152"
1086 #define STR_k "\153"
1087 #define STR_l "\154"
1088 #define STR_m "\155"
1089 #define STR_n "\156"
1090 #define STR_o "\157"
1091 #define STR_p "\160"
1092 #define STR_q "\161"
1093 #define STR_r "\162"
1094 #define STR_s "\163"
1095 #define STR_t "\164"
1096 #define STR_u "\165"
1097 #define STR_v "\166"
1098 #define STR_w "\167"
1099 #define STR_x "\170"
1100 #define STR_y "\171"
1101 #define STR_z "\172"
1102 #define STR_LEFT_CURLY_BRACKET "\173"
1103 #define STR_VERTICAL_LINE "\174"
1104 #define STR_RIGHT_CURLY_BRACKET "\175"
1105 #define STR_TILDE "\176"
1106
1107 #define STRING_ACCEPT0 STR_A STR_C STR_C STR_E STR_P STR_T "\0"
1108 #define STRING_COMMIT0 STR_C STR_O STR_M STR_M STR_I STR_T "\0"
1109 #define STRING_F0 STR_F "\0"
1110 #define STRING_FAIL0 STR_F STR_A STR_I STR_L "\0"
1111 #define STRING_PRUNE0 STR_P STR_R STR_U STR_N STR_E "\0"
1112 #define STRING_SKIP0 STR_S STR_K STR_I STR_P "\0"
1113 #define STRING_THEN STR_T STR_H STR_E STR_N
1114
1115 #define STRING_alpha0 STR_a STR_l STR_p STR_h STR_a "\0"
1116 #define STRING_lower0 STR_l STR_o STR_w STR_e STR_r "\0"
1117 #define STRING_upper0 STR_u STR_p STR_p STR_e STR_r "\0"
1118 #define STRING_alnum0 STR_a STR_l STR_n STR_u STR_m "\0"
1119 #define STRING_ascii0 STR_a STR_s STR_c STR_i STR_i "\0"
1120 #define STRING_blank0 STR_b STR_l STR_a STR_n STR_k "\0"
1121 #define STRING_cntrl0 STR_c STR_n STR_t STR_r STR_l "\0"
1122 #define STRING_digit0 STR_d STR_i STR_g STR_i STR_t "\0"
1123 #define STRING_graph0 STR_g STR_r STR_a STR_p STR_h "\0"
1124 #define STRING_print0 STR_p STR_r STR_i STR_n STR_t "\0"
1125 #define STRING_punct0 STR_p STR_u STR_n STR_c STR_t "\0"
1126 #define STRING_space0 STR_s STR_p STR_a STR_c STR_e "\0"
1127 #define STRING_word0 STR_w STR_o STR_r STR_d "\0"
1128 #define STRING_xdigit STR_x STR_d STR_i STR_g STR_i STR_t
1129
1130 #define STRING_DEFINE STR_D STR_E STR_F STR_I STR_N STR_E
1131
1132 #define STRING_CR_RIGHTPAR STR_C STR_R STR_RIGHT_PARENTHESIS
1133 #define STRING_LF_RIGHTPAR STR_L STR_F STR_RIGHT_PARENTHESIS
1134 #define STRING_CRLF_RIGHTPAR STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1135 #define STRING_ANY_RIGHTPAR STR_A STR_N STR_Y STR_RIGHT_PARENTHESIS
1136 #define STRING_ANYCRLF_RIGHTPAR STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1137 #define STRING_BSR_ANYCRLF_RIGHTPAR STR_B STR_S STR_R STR_UNDERSCORE STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1138 #define STRING_BSR_UNICODE_RIGHTPAR STR_B STR_S STR_R STR_UNDERSCORE STR_U STR_N STR_I STR_C STR_O STR_D STR_E STR_RIGHT_PARENTHESIS
1139 #define STRING_UTF8_RIGHTPAR STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS
1140
1141 #endif /* SUPPORT_UTF8 */
1142
1143 /* Escape items that are just an encoding of a particular data value. */
1144
1145 #ifndef ESC_e
1146 #define ESC_e CHAR_ESC
1147 #endif
1148
1149 #ifndef ESC_f
1150 #define ESC_f CHAR_FF
1151 #endif
1152
1153 #ifndef ESC_n
1154 #define ESC_n CHAR_NL
1155 #endif
1156
1157 #ifndef ESC_r
1158 #define ESC_r CHAR_CR
1159 #endif
1160
1161 /* We can't officially use ESC_t because it is a POSIX reserved identifier
1162 (presumably because of all the others like size_t). */
1163
1164 #ifndef ESC_tee
1165 #define ESC_tee CHAR_HT
1166 #endif
1167
1168 /* Codes for different types of Unicode property */
1169
1170 #define PT_ANY 0 /* Any property - matches all chars */
1171 #define PT_LAMP 1 /* L& - the union of Lu, Ll, Lt */
1172 #define PT_GC 2 /* General characteristic (e.g. L) */
1173 #define PT_PC 3 /* Particular characteristic (e.g. Lu) */
1174 #define PT_SC 4 /* Script (e.g. Han) */
1175
1176 /* Flag bits and data types for the extended class (OP_XCLASS) for classes that
1177 contain UTF-8 characters with values greater than 255. */
1178
1179 #define XCL_NOT 0x01 /* Flag: this is a negative class */
1180 #define XCL_MAP 0x02 /* Flag: a 32-byte map is present */
1181
1182 #define XCL_END 0 /* Marks end of individual items */
1183 #define XCL_SINGLE 1 /* Single item (one multibyte char) follows */
1184 #define XCL_RANGE 2 /* A range (two multibyte chars) follows */
1185 #define XCL_PROP 3 /* Unicode property (2-byte property code follows) */
1186 #define XCL_NOTPROP 4 /* Unicode inverted property (ditto) */
1187
1188 /* These are escaped items that aren't just an encoding of a particular data
1189 value such as \n. They must have non-zero values, as check_escape() returns
1190 their negation. Also, they must appear in the same order as in the opcode
1191 definitions below, up to ESC_z. There's a dummy for OP_ANY because it
1192 corresponds to "." rather than an escape sequence, and another for OP_ALLANY
1193 (which is used for [^] in JavaScript compatibility mode).
1194
1195 The final escape must be ESC_REF as subsequent values are used for
1196 backreferences (\1, \2, \3, etc). There are two tests in the code for an escape
1197 greater than ESC_b and less than ESC_Z to detect the types that may be
1198 repeated. These are the types that consume characters. If any new escapes are
1199 put in between that don't consume a character, that code will have to change.
1200 */
1201
1202 enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s,
1203 ESC_W, ESC_w, ESC_dum1, ESC_dum2, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H,
1204 ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z, ESC_E, ESC_Q, ESC_g, ESC_k,
1205 ESC_REF };
1206
1207
1208 /* Opcode table: Starting from 1 (i.e. after OP_END), the values up to
1209 OP_EOD must correspond in order to the list of escapes immediately above.
1210
1211 *** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions
1212 that follow must also be updated to match. There is also a table called
1213 "coptable" in pcre_dfa_exec.c that must be updated. */
1214
1215 enum {
1216 OP_END, /* 0 End of pattern */
1217
1218 /* Values corresponding to backslashed metacharacters */
1219
1220 OP_SOD, /* 1 Start of data: \A */
1221 OP_SOM, /* 2 Start of match (subject + offset): \G */
1222 OP_SET_SOM, /* 3 Set start of match (\K) */
1223 OP_NOT_WORD_BOUNDARY, /* 4 \B */
1224 OP_WORD_BOUNDARY, /* 5 \b */
1225 OP_NOT_DIGIT, /* 6 \D */
1226 OP_DIGIT, /* 7 \d */
1227 OP_NOT_WHITESPACE, /* 8 \S */
1228 OP_WHITESPACE, /* 9 \s */
1229 OP_NOT_WORDCHAR, /* 10 \W */
1230 OP_WORDCHAR, /* 11 \w */
1231 OP_ANY, /* 12 Match any character (subject to DOTALL) */
1232 OP_ALLANY, /* 13 Match any character (not subject to DOTALL) */
1233 OP_ANYBYTE, /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */
1234 OP_NOTPROP, /* 15 \P (not Unicode property) */
1235 OP_PROP, /* 16 \p (Unicode property) */
1236 OP_ANYNL, /* 17 \R (any newline sequence) */
1237 OP_NOT_HSPACE, /* 18 \H (not horizontal whitespace) */
1238 OP_HSPACE, /* 19 \h (horizontal whitespace) */
1239 OP_NOT_VSPACE, /* 20 \V (not vertical whitespace) */
1240 OP_VSPACE, /* 21 \v (vertical whitespace) */
1241 OP_EXTUNI, /* 22 \X (extended Unicode sequence */
1242 OP_EODN, /* 23 End of data or \n at end of data: \Z. */
1243 OP_EOD, /* 24 End of data: \z */
1244
1245 OP_OPT, /* 25 Set runtime options */
1246 OP_CIRC, /* 26 Start of line - varies with multiline switch */
1247 OP_DOLL, /* 27 End of line - varies with multiline switch */
1248 OP_CHAR, /* 28 Match one character, casefully */
1249 OP_CHARNC, /* 29 Match one character, caselessly */
1250 OP_NOT, /* 30 Match one character, not the following one */
1251
1252 OP_STAR, /* 31 The maximizing and minimizing versions of */
1253 OP_MINSTAR, /* 32 these six opcodes must come in pairs, with */
1254 OP_PLUS, /* 33 the minimizing one second. */
1255 OP_MINPLUS, /* 34 This first set applies to single characters.*/
1256 OP_QUERY, /* 35 */
1257 OP_MINQUERY, /* 36 */
1258
1259 OP_UPTO, /* 37 From 0 to n matches */
1260 OP_MINUPTO, /* 38 */
1261 OP_EXACT, /* 39 Exactly n matches */
1262
1263 OP_POSSTAR, /* 40 Possessified star */
1264 OP_POSPLUS, /* 41 Possessified plus */
1265 OP_POSQUERY, /* 42 Posesssified query */
1266 OP_POSUPTO, /* 43 Possessified upto */
1267
1268 OP_NOTSTAR, /* 44 The maximizing and minimizing versions of */
1269 OP_NOTMINSTAR, /* 45 these six opcodes must come in pairs, with */
1270 OP_NOTPLUS, /* 46 the minimizing one second. They must be in */
1271 OP_NOTMINPLUS, /* 47 exactly the same order as those above. */
1272 OP_NOTQUERY, /* 48 This set applies to "not" single characters. */
1273 OP_NOTMINQUERY, /* 49 */
1274
1275 OP_NOTUPTO, /* 50 From 0 to n matches */
1276 OP_NOTMINUPTO, /* 51 */
1277 OP_NOTEXACT, /* 52 Exactly n matches */
1278
1279 OP_NOTPOSSTAR, /* 53 Possessified versions */
1280 OP_NOTPOSPLUS, /* 54 */
1281 OP_NOTPOSQUERY, /* 55 */
1282 OP_NOTPOSUPTO, /* 56 */
1283
1284 OP_TYPESTAR, /* 57 The maximizing and minimizing versions of */
1285 OP_TYPEMINSTAR, /* 58 these six opcodes must come in pairs, with */
1286 OP_TYPEPLUS, /* 59 the minimizing one second. These codes must */
1287 OP_TYPEMINPLUS, /* 60 be in exactly the same order as those above. */
1288 OP_TYPEQUERY, /* 61 This set applies to character types such as \d */
1289 OP_TYPEMINQUERY, /* 62 */
1290
1291 OP_TYPEUPTO, /* 63 From 0 to n matches */
1292 OP_TYPEMINUPTO, /* 64 */
1293 OP_TYPEEXACT, /* 65 Exactly n matches */
1294
1295 OP_TYPEPOSSTAR, /* 66 Possessified versions */
1296 OP_TYPEPOSPLUS, /* 67 */
1297 OP_TYPEPOSQUERY, /* 68 */
1298 OP_TYPEPOSUPTO, /* 69 */
1299
1300 OP_CRSTAR, /* 70 The maximizing and minimizing versions of */
1301 OP_CRMINSTAR, /* 71 all these opcodes must come in pairs, with */
1302 OP_CRPLUS, /* 72 the minimizing one second. These codes must */
1303 OP_CRMINPLUS, /* 73 be in exactly the same order as those above. */
1304 OP_CRQUERY, /* 74 These are for character classes and back refs */
1305 OP_CRMINQUERY, /* 75 */
1306 OP_CRRANGE, /* 76 These are different to the three sets above. */
1307 OP_CRMINRANGE, /* 77 */
1308
1309 OP_CLASS, /* 78 Match a character class, chars < 256 only */
1310 OP_NCLASS, /* 79 Same, but the bitmap was created from a negative
1311 class - the difference is relevant only when a UTF-8
1312 character > 255 is encountered. */
1313
1314 OP_XCLASS, /* 80 Extended class for handling UTF-8 chars within the
1315 class. This does both positive and negative. */
1316
1317 OP_REF, /* 81 Match a back reference */
1318 OP_RECURSE, /* 82 Match a numbered subpattern (possibly recursive) */
1319 OP_CALLOUT, /* 83 Call out to external function if provided */
1320
1321 OP_ALT, /* 84 Start of alternation */
1322 OP_KET, /* 85 End of group that doesn't have an unbounded repeat */
1323 OP_KETRMAX, /* 86 These two must remain together and in this */
1324 OP_KETRMIN, /* 87 order. They are for groups the repeat for ever. */
1325
1326 /* The assertions must come before BRA, CBRA, ONCE, and COND.*/
1327
1328 OP_ASSERT, /* 88 Positive lookahead */
1329 OP_ASSERT_NOT, /* 89 Negative lookahead */
1330 OP_ASSERTBACK, /* 90 Positive lookbehind */
1331 OP_ASSERTBACK_NOT, /* 91 Negative lookbehind */
1332 OP_REVERSE, /* 92 Move pointer back - used in lookbehind assertions */
1333
1334 /* ONCE, BRA, CBRA, and COND must come after the assertions, with ONCE first,
1335 as there's a test for >= ONCE for a subpattern that isn't an assertion. */
1336
1337 OP_ONCE, /* 93 Atomic group */
1338 OP_BRA, /* 94 Start of non-capturing bracket */
1339 OP_CBRA, /* 95 Start of capturing bracket */
1340 OP_COND, /* 96 Conditional group */
1341
1342 /* These three must follow the previous three, in the same order. There's a
1343 check for >= SBRA to distinguish the two sets. */
1344
1345 OP_SBRA, /* 97 Start of non-capturing bracket, check empty */
1346 OP_SCBRA, /* 98 Start of capturing bracket, check empty */
1347 OP_SCOND, /* 99 Conditional group, check empty */
1348
1349 OP_CREF, /* 100 Used to hold a capture number as condition */
1350 OP_RREF, /* 101 Used to hold a recursion number as condition */
1351 OP_DEF, /* 102 The DEFINE condition */
1352
1353 OP_BRAZERO, /* 103 These two must remain together and in this */
1354 OP_BRAMINZERO, /* 104 order. */
1355
1356 /* These are backtracking control verbs */
1357
1358 OP_PRUNE, /* 105 */
1359 OP_SKIP, /* 106 */
1360 OP_THEN, /* 107 */
1361 OP_COMMIT, /* 108 */
1362
1363 /* These are forced failure and success verbs */
1364
1365 OP_FAIL, /* 109 */
1366 OP_ACCEPT, /* 110 */
1367
1368 /* This is used to skip a subpattern with a {0} quantifier */
1369
1370 OP_SKIPZERO /* 111 */
1371 };
1372
1373
1374 /* This macro defines textual names for all the opcodes. These are used only
1375 for debugging. The macro is referenced only in pcre_printint.c. */
1376
1377 #define OP_NAME_LIST \
1378 "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d", \
1379 "\\S", "\\s", "\\W", "\\w", "Any", "AllAny", "Anybyte", \
1380 "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v", \
1381 "extuni", "\\Z", "\\z", \
1382 "Opt", "^", "$", "char", "charnc", "not", \
1383 "*", "*?", "+", "+?", "?", "??", "{", "{", "{", \
1384 "*+","++", "?+", "{", \
1385 "*", "*?", "+", "+?", "?", "??", "{", "{", "{", \
1386 "*+","++", "?+", "{", \
1387 "*", "*?", "+", "+?", "?", "??", "{", "{", "{", \
1388 "*+","++", "?+", "{", \
1389 "*", "*?", "+", "+?", "?", "??", "{", "{", \
1390 "class", "nclass", "xclass", "Ref", "Recurse", "Callout", \
1391 "Alt", "Ket", "KetRmax", "KetRmin", "Assert", "Assert not", \
1392 "AssertB", "AssertB not", "Reverse", \
1393 "Once", "Bra", "CBra", "Cond", "SBra", "SCBra", "SCond", \
1394 "Cond ref", "Cond rec", "Cond def", "Brazero", "Braminzero", \
1395 "*PRUNE", "*SKIP", "*THEN", "*COMMIT", "*FAIL", "*ACCEPT", \
1396 "Skip zero"
1397
1398
1399 /* This macro defines the length of fixed length operations in the compiled
1400 regex. The lengths are used when searching for specific things, and also in the
1401 debugging printing of a compiled regex. We use a macro so that it can be
1402 defined close to the definitions of the opcodes themselves.
1403
1404 As things have been extended, some of these are no longer fixed lenths, but are
1405 minima instead. For example, the length of a single-character repeat may vary
1406 in UTF-8 mode. The code that uses this table must know about such things. */
1407
1408 #define OP_LENGTHS \
1409 1, /* End */ \
1410 1, 1, 1, 1, 1, /* \A, \G, \K, \B, \b */ \
1411 1, 1, 1, 1, 1, 1, /* \D, \d, \S, \s, \W, \w */ \
1412 1, 1, 1, /* Any, AllAny, Anybyte */ \
1413 3, 3, 1, /* NOTPROP, PROP, EXTUNI */ \
1414 1, 1, 1, 1, 1, /* \R, \H, \h, \V, \v */ \
1415 1, 1, 2, 1, 1, /* \Z, \z, Opt, ^, $ */ \
1416 2, /* Char - the minimum length */ \
1417 2, /* Charnc - the minimum length */ \
1418 2, /* not */ \
1419 /* Positive single-char repeats ** These are */ \
1420 2, 2, 2, 2, 2, 2, /* *, *?, +, +?, ?, ?? ** minima in */ \
1421 4, 4, 4, /* upto, minupto, exact ** UTF-8 mode */ \
1422 2, 2, 2, 4, /* *+, ++, ?+, upto+ */ \
1423 /* Negative single-char repeats - only for chars < 256 */ \
1424 2, 2, 2, 2, 2, 2, /* NOT *, *?, +, +?, ?, ?? */ \
1425 4, 4, 4, /* NOT upto, minupto, exact */ \
1426 2, 2, 2, 4, /* Possessive *, +, ?, upto */ \
1427 /* Positive type repeats */ \
1428 2, 2, 2, 2, 2, 2, /* Type *, *?, +, +?, ?, ?? */ \
1429 4, 4, 4, /* Type upto, minupto, exact */ \
1430 2, 2, 2, 4, /* Possessive *+, ++, ?+, upto+ */ \
1431 /* Character class & ref repeats */ \
1432 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ \
1433 5, 5, /* CRRANGE, CRMINRANGE */ \
1434 33, /* CLASS */ \
1435 33, /* NCLASS */ \
1436 0, /* XCLASS - variable length */ \
1437 3, /* REF */ \
1438 1+LINK_SIZE, /* RECURSE */ \
1439 2+2*LINK_SIZE, /* CALLOUT */ \
1440 1+LINK_SIZE, /* Alt */ \
1441 1+LINK_SIZE, /* Ket */ \
1442 1+LINK_SIZE, /* KetRmax */ \
1443 1+LINK_SIZE, /* KetRmin */ \
1444 1+LINK_SIZE, /* Assert */ \
1445 1+LINK_SIZE, /* Assert not */ \
1446 1+LINK_SIZE, /* Assert behind */ \
1447 1+LINK_SIZE, /* Assert behind not */ \
1448 1+LINK_SIZE, /* Reverse */ \
1449 1+LINK_SIZE, /* ONCE */ \
1450 1+LINK_SIZE, /* BRA */ \
1451 3+LINK_SIZE, /* CBRA */ \
1452 1+LINK_SIZE, /* COND */ \
1453 1+LINK_SIZE, /* SBRA */ \
1454 3+LINK_SIZE, /* SCBRA */ \
1455 1+LINK_SIZE, /* SCOND */ \
1456 3, /* CREF */ \
1457 3, /* RREF */ \
1458 1, /* DEF */ \
1459 1, 1, /* BRAZERO, BRAMINZERO */ \
1460 1, 1, 1, 1, /* PRUNE, SKIP, THEN, COMMIT, */ \
1461 1, 1, 1 /* FAIL, ACCEPT, SKIPZERO */
1462
1463
1464 /* A magic value for OP_RREF to indicate the "any recursion" condition. */
1465
1466 #define RREF_ANY 0xffff
1467
1468 /* Error code numbers. They are given names so that they can more easily be
1469 tracked. */
1470
1471 enum { ERR0, ERR1, ERR2, ERR3, ERR4, ERR5, ERR6, ERR7, ERR8, ERR9,
1472 ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,
1473 ERR20, ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR28, ERR29,
1474 ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,
1475 ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,
1476 ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,
1477 ERR60, ERR61, ERR62, ERR63, ERR64 };
1478
1479 /* The real format of the start of the pcre block; the index of names and the
1480 code vector run on as long as necessary after the end. We store an explicit
1481 offset to the name table so that if a regex is compiled on one host, saved, and
1482 then run on another where the size of pointers is different, all might still
1483 be well. For the case of compiled-on-4 and run-on-8, we include an extra
1484 pointer that is always NULL. For future-proofing, a few dummy fields were
1485 originally included - even though you can never get this planning right - but
1486 there is only one left now.
1487
1488 NOTE NOTE NOTE:
1489 Because people can now save and re-use compiled patterns, any additions to this
1490 structure should be made at the end, and something earlier (e.g. a new
1491 flag in the options or one of the dummy fields) should indicate that the new
1492 fields are present. Currently PCRE always sets the dummy fields to zero.
1493 NOTE NOTE NOTE:
1494 */
1495
1496 typedef struct real_pcre {
1497 pcre_uint32 magic_number;
1498 pcre_uint32 size; /* Total that was malloced */
1499 pcre_uint32 options; /* Public options */
1500 pcre_uint16 flags; /* Private flags */
1501 pcre_uint16 dummy1; /* For future use */
1502 pcre_uint16 top_bracket;
1503 pcre_uint16 top_backref;
1504 pcre_uint16 first_byte;
1505 pcre_uint16 req_byte;
1506 pcre_uint16 name_table_offset; /* Offset to name table that follows */
1507 pcre_uint16 name_entry_size; /* Size of any name items */
1508 pcre_uint16 name_count; /* Number of name items */
1509 pcre_uint16 ref_count; /* Reference count */
1510
1511 const unsigned char *tables; /* Pointer to tables or NULL for std */
1512 const unsigned char *nullpad; /* NULL padding */
1513 } real_pcre;
1514
1515 /* The format of the block used to store data from pcre_study(). The same
1516 remark (see NOTE above) about extending this structure applies. */
1517
1518 typedef struct pcre_study_data {
1519 pcre_uint32 size; /* Total that was malloced */
1520 pcre_uint32 options;
1521 uschar start_bits[32];
1522 } pcre_study_data;
1523
1524 /* Structure for passing "static" information around between the functions
1525 doing the compiling, so that they are thread-safe. */
1526
1527 typedef struct compile_data {
1528 const uschar *lcc; /* Points to lower casing table */
1529 const uschar *fcc; /* Points to case-flipping table */
1530 const uschar *cbits; /* Points to character type table */
1531 const uschar *ctypes; /* Points to table of type maps */
1532 const uschar *start_workspace;/* The start of working space */
1533 const uschar *start_code; /* The start of the compiled code */
1534 const uschar *start_pattern; /* The start of the pattern */
1535 const uschar *end_pattern; /* The end of the pattern */
1536 uschar *hwm; /* High watermark of workspace */
1537 uschar *name_table; /* The name/number table */
1538 int names_found; /* Number of entries so far */
1539 int name_entry_size; /* Size of each entry */
1540 int bracount; /* Count of capturing parens as we compile */
1541 int final_bracount; /* Saved value after first pass */
1542 int top_backref; /* Maximum back reference */
1543 unsigned int backref_map; /* Bitmap of low back refs */
1544 int external_options; /* External (initial) options */
1545 int external_flags; /* External flag bits to be set */
1546 int req_varyopt; /* "After variable item" flag for reqbyte */
1547 BOOL had_accept; /* (*ACCEPT) encountered */
1548 int nltype; /* Newline type */
1549 int nllen; /* Newline string length */
1550 uschar nl[4]; /* Newline string when fixed length */
1551 uschar padding[4];
1552 } compile_data;
1553
1554 /* Structure for maintaining a chain of pointers to the currently incomplete
1555 branches, for testing for left recursion. */
1556
1557 typedef struct branch_chain {
1558 struct branch_chain *outer;
1559 uschar *current;
1560 } branch_chain;
1561
1562 /* Structure for items in a linked list that represents an explicit recursive
1563 call within the pattern. */
1564
1565 typedef struct recursion_info {
1566 struct recursion_info *prevrec; /* Previous recursion record (or NULL) */
1567 const uschar *after_call; /* "Return value": points after the call in the expr */
1568 USPTR save_start; /* Old value of mstart */
1569 int *offset_save; /* Pointer to start of saved offsets */
1570 int group_num; /* Number of group that was called */
1571 int saved_max; /* Number of saved offsets */
1572 } recursion_info;
1573
1574 /* Structure for building a chain of data for holding the values of the subject
1575 pointer at the start of each subpattern, so as to detect when an empty string
1576 has been matched by a subpattern - to break infinite loops. */
1577
1578 typedef struct eptrblock {
1579 struct eptrblock *epb_prev;
1580 USPTR epb_saved_eptr;
1581 } eptrblock;
1582
1583
1584 /* Structure for passing "static" information around between the functions
1585 doing traditional NFA matching, so that they are thread-safe. */
1586
1587 typedef struct match_data {
1588 unsigned long int match_call_count; /* As it says */
1589 unsigned long int match_limit; /* As it says */
1590 unsigned long int match_limit_recursion; /* As it says */
1591 int *offset_vector; /* Offset vector */
1592 int offset_end; /* One past the end */
1593 int offset_max; /* The maximum usable for return data */
1594 int nltype; /* Newline type */
1595 int nllen; /* Newline string length */
1596 const uschar *lcc; /* Points to lower casing table */
1597 const uschar *ctypes; /* Points to table of type maps */
1598 uschar nl[4]; /* Newline string when fixed */
1599 BOOL offset_overflow; /* Set if too many extractions */
1600 BOOL notbol; /* NOTBOL flag */
1601 BOOL noteol; /* NOTEOL flag */
1602 BOOL utf8; /* UTF8 flag */
1603 BOOL jscript_compat; /* JAVASCRIPT_COMPAT flag */
1604 BOOL endonly; /* Dollar not before final \n */
1605 BOOL notempty; /* Empty string match not wanted */
1606 BOOL partial; /* PARTIAL flag */
1607 BOOL hitend; /* Hit the end of the subject at some point */
1608 BOOL bsr_anycrlf; /* \R is just any CRLF, not full Unicode */
1609 int end_offset_top; /* Highwater mark at end of match */
1610 int capture_last; /* Most recent capture number */
1611 int start_offset; /* The start offset value */
1612 const uschar *start_code; /* For use when recursing */
1613 USPTR start_subject; /* Start of the subject string */
1614 USPTR end_subject; /* End of the subject string */
1615 USPTR start_match_ptr; /* Start of matched string */
1616 USPTR end_match_ptr; /* Subject position at end match */
1617 eptrblock *eptrchain; /* Chain of eptrblocks for tail recursions */
1618 recursion_info *recursive; /* Linked list of recursion data */
1619 void *callout_data; /* To pass back to callouts */
1620 int eptrn; /* Next free eptrblock */
1621 uschar padding[4];
1622 } match_data;
1623
1624 /* A similar structure is used for the same purpose by the DFA matching
1625 functions. */
1626
1627 typedef struct dfa_match_data {
1628 const uschar *start_code; /* Start of the compiled pattern */
1629 const uschar *start_subject; /* Start of the subject string */
1630 const uschar *end_subject; /* End of subject string */
1631 const uschar *tables; /* Character tables */
1632 int moptions; /* Match options */
1633 int poptions; /* Pattern options */
1634 int nltype; /* Newline type */
1635 int nllen; /* Newline string length */
1636 void *callout_data; /* To pass back to callouts */
1637 uschar nl[4]; /* Newline string when fixed */
1638 uschar padding[4];
1639 } dfa_match_data;
1640
1641 /* Bit definitions for entries in the pcre_ctypes table. */
1642
1643 #define ctype_space 0x01
1644 #define ctype_letter 0x02
1645 #define ctype_digit 0x04
1646 #define ctype_xdigit 0x08
1647 #define ctype_word 0x10 /* alphanumeric or '_' */
1648 #define ctype_meta 0x80 /* regexp meta char or zero (end pattern) */
1649
1650 /* Offsets for the bitmap tables in pcre_cbits. Each table contains a set
1651 of bits for a class map. Some classes are built by combining these tables. */
1652
1653 #define cbit_space 0 /* [:space:] or \s */
1654 #define cbit_xdigit 32 /* [:xdigit:] */
1655 #define cbit_digit 64 /* [:digit:] or \d */
1656 #define cbit_upper 96 /* [:upper:] */
1657 #define cbit_lower 128 /* [:lower:] */
1658 #define cbit_word 160 /* [:word:] or \w */
1659 #define cbit_graph 192 /* [:graph:] */
1660 #define cbit_print 224 /* [:print:] */
1661 #define cbit_punct 256 /* [:punct:] */
1662 #define cbit_cntrl 288 /* [:cntrl:] */
1663 #define cbit_length 320 /* Length of the cbits table */
1664
1665 /* Offsets of the various tables from the base tables pointer, and
1666 total length. */
1667
1668 #define lcc_offset 0
1669 #define fcc_offset 256
1670 #define cbits_offset 512
1671 #define ctypes_offset (cbits_offset + cbit_length)
1672 #define tables_length (ctypes_offset + 256)
1673
1674 /* Layout of the UCP type table that translates property names into types and
1675 codes. Each entry used to point directly to a name, but to reduce the number of
1676 relocations in shared libraries, it now has an offset into a single string
1677 instead. */
1678
1679 typedef struct {
1680 pcre_uint16 name_offset;
1681 pcre_uint16 type;
1682 pcre_uint16 value;
1683 } ucp_type_table;
1684
1685
1686 /* Internal shared data tables. These are tables that are used by more than one
1687 of the exported public functions. They have to be "external" in the C sense,
1688 but are not part of the PCRE public API. The data for these tables is in the
1689 pcre_tables.c module. */
1690
1691 extern const int _pcre_utf8_table1[];
1692 extern const int _pcre_utf8_table2[];
1693 extern const int _pcre_utf8_table3[];
1694 extern const uschar _pcre_utf8_table4[];
1695
1696 extern const int _pcre_utf8_table1_size;
1697
1698 extern const char _pcre_utt_names[];
1699 extern const ucp_type_table _pcre_utt[];
1700 extern const int _pcre_utt_size;
1701
1702 extern const uschar _pcre_default_tables[];
1703
1704 extern const uschar _pcre_OP_lengths[];
1705
1706
1707 /* Internal shared functions. These are functions that are used by more than
1708 one of the exported public functions. They have to be "external" in the C
1709 sense, but are not part of the PCRE public API. */
1710
1711 extern BOOL _pcre_is_newline(const uschar *, int, const uschar *,
1712 int *, BOOL);
1713 extern int _pcre_ord2utf8(int, uschar *);
1714 extern real_pcre *_pcre_try_flipped(const real_pcre *, real_pcre *,
1715 const pcre_study_data *, pcre_study_data *);
1716 extern int _pcre_valid_utf8(const uschar *, int);
1717 extern BOOL _pcre_was_newline(const uschar *, int, const uschar *,
1718 int *, BOOL);
1719 extern BOOL _pcre_xclass(int, const uschar *);
1720
1721
1722 /* Unicode character database (UCD) */
1723
1724 typedef struct {
1725 pcre_int32 other_case;
1726 uschar script;
1727 uschar chartype;
1728 uschar padding[2];
1729 } ucd_record;
1730
1731 extern const ucd_record _pcre_ucd_records[];
1732 extern const uschar _pcre_ucd_stage1[];
1733 extern const pcre_uint16 _pcre_ucd_stage2[];
1734 extern const int _pcre_ucp_gentype[];
1735
1736
1737 /* UCD access macros */
1738
1739 #define UCD_BLOCK_SIZE 128
1740 #define GET_UCD(ch) (_pcre_ucd_records + \
1741 _pcre_ucd_stage2[_pcre_ucd_stage1[(ch) / UCD_BLOCK_SIZE] * \
1742 UCD_BLOCK_SIZE + ch % UCD_BLOCK_SIZE])
1743
1744 #define UCD_CHARTYPE(ch) GET_UCD(ch)->chartype
1745 #define UCD_SCRIPT(ch) GET_UCD(ch)->script
1746 #define UCD_CATEGORY(ch) _pcre_ucp_gentype[UCD_CHARTYPE(ch)]
1747 #define UCD_OTHERCASE(ch) (ch + GET_UCD(ch)->other_case)
1748
1749 #endif
1750
1751 /* End of pcre_internal.h */
1752