|
NCBI Home IEB Home C++ Toolkit docs C Toolkit source browser C Toolkit source browser (2) |
NCBI C++ Toolkit Cross ReferenceC++/src/util/regexp/pcre_internal.h |
source navigation diff markup identifier search freetext search file search |
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 # elif defined(__GNUC__) && __GNUC__ >= 4
135 # ifdef __cplusplus
136 # define PCRE_EXP_DECL extern "C" __attribute__((visibility("default")))
137 # else
138 # define PCRE_EXP_DECL extern __attribute__((visibility("default")))
139 # endif
140 # define PCRE_EXP_DEFN
141 # define PCRE_EXP_DATA_DEFN
142 # else
143 # ifdef __cplusplus
144 # define PCRE_EXP_DECL extern "C"
145 # else
146 # define PCRE_EXP_DECL extern
147 # endif
148 # ifndef PCRE_EXP_DEFN
149 # define PCRE_EXP_DEFN PCRE_EXP_DECL
150 # endif
151 # ifndef PCRE_EXP_DATA_DEFN
152 # define PCRE_EXP_DATA_DEFN
153 # endif
154 # endif
155 #endif
156
157 /* When compiling with the MSVC compiler, it is sometimes necessary to include
158 a "calling convention" before exported function names. (This is secondhand
159 information; I know nothing about MSVC myself). For example, something like
160
161 void __cdecl function(....)
162
163 might be needed. In order so make this easy, all the exported functions have
164 PCRE_CALL_CONVENTION just before their names. It is rarely needed; if not
165 set, we ensure here that it has no effect. */
166
167 #ifndef PCRE_CALL_CONVENTION
168 #define PCRE_CALL_CONVENTION
169 #endif
170
171 /* We need to have types that specify unsigned 16-bit and 32-bit integers. We
172 cannot determine these outside the compilation (e.g. by running a program as
173 part of "configure") because PCRE is often cross-compiled for use on other
174 systems. Instead we make use of the maximum sizes that are available at
175 preprocessor time in standard C environments. */
176
177 #if USHRT_MAX == 65535
178 typedef unsigned short pcre_uint16;
179 typedef short pcre_int16;
180 #elif UINT_MAX == 65535
181 typedef unsigned int pcre_uint16;
182 typedef int pcre_int16;
183 #else
184 #error Cannot determine a type for 16-bit unsigned integers
185 #endif
186
187 #if UINT_MAX == 4294967295
188 typedef unsigned int pcre_uint32;
189 typedef int pcre_int32;
190 #elif ULONG_MAX == 4294967295
191 typedef unsigned long int pcre_uint32;
192 typedef long int pcre_int32;
193 #else
194 #error Cannot determine a type for 32-bit unsigned integers
195 #endif
196
197 /* All character handling must be done as unsigned characters. Otherwise there
198 are problems with top-bit-set characters and functions such as isspace().
199 However, we leave the interface to the outside world as char *, because that
200 should make things easier for callers. We define a short type for unsigned char
201 to save lots of typing. I tried "uchar", but it causes problems on Digital
202 Unix, where it is defined in sys/types, so use "uschar" instead. */
203
204 typedef unsigned char uschar;
205
206 /* This is an unsigned int value that no character can ever have. UTF-8
207 characters only go up to 0x7fffffff (though Unicode doesn't go beyond
208 0x0010ffff). */
209
210 #define NOTACHAR 0xffffffff
211
212 /* PCRE is able to support several different kinds of newline (CR, LF, CRLF,
213 "any" and "anycrlf" at present). The following macros are used to package up
214 testing for newlines. NLBLOCK, PSSTART, and PSEND are defined in the various
215 modules to indicate in which datablock the parameters exist, and what the
216 start/end of string field names are. */
217
218 #define NLTYPE_FIXED 0 /* Newline is a fixed length string */
219 #define NLTYPE_ANY 1 /* Newline is any Unicode line ending */
220 #define NLTYPE_ANYCRLF 2 /* Newline is CR, LF, or CRLF */
221
222 /* This macro checks for a newline at the given position */
223
224 #define IS_NEWLINE(p) \
225 ((NLBLOCK->nltype != NLTYPE_FIXED)? \
226 ((p) < NLBLOCK->PSEND && \
227 _pcre_is_newline((p), NLBLOCK->nltype, NLBLOCK->PSEND, &(NLBLOCK->nllen),\
228 utf8)) \
229 : \
230 ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \
231 (p)[0] == NLBLOCK->nl[0] && \
232 (NLBLOCK->nllen == 1 || (p)[1] == NLBLOCK->nl[1]) \
233 ) \
234 )
235
236 /* This macro checks for a newline immediately preceding the given position */
237
238 #define WAS_NEWLINE(p) \
239 ((NLBLOCK->nltype != NLTYPE_FIXED)? \
240 ((p) > NLBLOCK->PSSTART && \
241 _pcre_was_newline((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \
242 &(NLBLOCK->nllen), utf8)) \
243 : \
244 ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \
245 (p)[-NLBLOCK->nllen] == NLBLOCK->nl[0] && \
246 (NLBLOCK->nllen == 1 || (p)[-NLBLOCK->nllen+1] == NLBLOCK->nl[1]) \
247 ) \
248 )
249
250 /* When PCRE is compiled as a C++ library, the subject pointer can be replaced
251 with a custom type. This makes it possible, for example, to allow pcre_exec()
252 to process subject strings that are discontinuous by using a smart pointer
253 class. It must always be possible to inspect all of the subject string in
254 pcre_exec() because of the way it backtracks. Two macros are required in the
255 normal case, for sign-unspecified and unsigned char pointers. The former is
256 used for the external interface and appears in pcre.h, which is why its name
257 must begin with PCRE_. */
258
259 #ifdef CUSTOM_SUBJECT_PTR
260 #define PCRE_SPTR CUSTOM_SUBJECT_PTR
261 #define USPTR CUSTOM_SUBJECT_PTR
262 #else
263 #define PCRE_SPTR const char *
264 #define USPTR const unsigned char *
265 #endif
266
267
268
269 /* Include the public PCRE header and the definitions of UCP character property
270 values. */
271
272 #include "pcre.h"
273 #include "ucp.h"
274
275 /* When compiling for use with the Virtual Pascal compiler, these functions
276 need to have their names changed. PCRE must be compiled with the -DVPCOMPAT
277 option on the command line. */
278
279 #ifdef VPCOMPAT
280 #define strlen(s) _strlen(s)
281 #define strncmp(s1,s2,m) _strncmp(s1,s2,m)
282 #define memcmp(s,c,n) _memcmp(s,c,n)
283 #define memcpy(d,s,n) _memcpy(d,s,n)
284 #define memmove(d,s,n) _memmove(d,s,n)
285 #define memset(s,c,n) _memset(s,c,n)
286 #else /* VPCOMPAT */
287
288 /* To cope with SunOS4 and other systems that lack memmove() but have bcopy(),
289 define a macro for memmove() if HAVE_MEMMOVE is false, provided that HAVE_BCOPY
290 is set. Otherwise, include an emulating function for those systems that have
291 neither (there some non-Unix environments where this is the case). */
292
293 #ifndef HAVE_MEMMOVE
294 #undef memmove /* some systems may have a macro */
295 #ifdef HAVE_BCOPY
296 #define memmove(a, b, c) bcopy(b, a, c)
297 #else /* HAVE_BCOPY */
298 static void *
299 pcre_memmove(void *d, const void *s, size_t n)
300 {
301 size_t i;
302 unsigned char *dest = (unsigned char *)d;
303 const unsigned char *src = (const unsigned char *)s;
304 if (dest > src)
305 {
306 dest += n;
307 src += n;
308 for (i = 0; i < n; ++i) *(--dest) = *(--src);
309 return (void *)dest;
310 }
311 else
312 {
313 for (i = 0; i < n; ++i) *dest++ = *src++;
314 return (void *)(dest - n);
315 }
316 }
317 #define memmove(a, b, c) pcre_memmove(a, b, c)
318 #endif /* not HAVE_BCOPY */
319 #endif /* not HAVE_MEMMOVE */
320 #endif /* not VPCOMPAT */
321
322
323 /* PCRE keeps offsets in its compiled code as 2-byte quantities (always stored
324 in big-endian order) by default. These are used, for example, to link from the
325 start of a subpattern to its alternatives and its end. The use of 2 bytes per
326 offset limits the size of the compiled regex to around 64K, which is big enough
327 for almost everybody. However, I received a request for an even bigger limit.
328 For this reason, and also to make the code easier to maintain, the storing and
329 loading of offsets from the byte string is now handled by the macros that are
330 defined here.
331
332 The macros are controlled by the value of LINK_SIZE. This defaults to 2 in
333 the config.h file, but can be overridden by using -D on the command line. This
334 is automated on Unix systems via the "configure" command. */
335
336 #if LINK_SIZE == 2
337
338 #define PUT(a,n,d) \
339 (a[n] = (d) >> 8), \
340 (a[(n)+1] = (d) & 255)
341
342 #define GET(a,n) \
343 (((a)[n] << 8) | (a)[(n)+1])
344
345 #define MAX_PATTERN_SIZE (1 << 16)
346
347
348 #elif LINK_SIZE == 3
349
350 #define PUT(a,n,d) \
351 (a[n] = (d) >> 16), \
352 (a[(n)+1] = (d) >> 8), \
353 (a[(n)+2] = (d) & 255)
354
355 #define GET(a,n) \
356 (((a)[n] << 16) | ((a)[(n)+1] << 8) | (a)[(n)+2])
357
358 #define MAX_PATTERN_SIZE (1 << 24)
359
360
361 #elif LINK_SIZE == 4
362
363 #define PUT(a,n,d) \
364 (a[n] = (d) >> 24), \
365 (a[(n)+1] = (d) >> 16), \
366 (a[(n)+2] = (d) >> 8), \
367 (a[(n)+3] = (d) & 255)
368
369 #define GET(a,n) \
370 (((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3])
371
372 #define MAX_PATTERN_SIZE (1 << 30) /* Keep it positive */
373
374
375 #else
376 #error LINK_SIZE must be either 2, 3, or 4
377 #endif
378
379
380 /* Convenience macro defined in terms of the others */
381
382 #define PUTINC(a,n,d) PUT(a,n,d), a += LINK_SIZE
383
384
385 /* PCRE uses some other 2-byte quantities that do not change when the size of
386 offsets changes. There are used for repeat counts and for other things such as
387 capturing parenthesis numbers in back references. */
388
389 #define PUT2(a,n,d) \
390 a[n] = (d) >> 8; \
391 a[(n)+1] = (d) & 255
392
393 #define GET2(a,n) \
394 (((a)[n] << 8) | (a)[(n)+1])
395
396 #define PUT2INC(a,n,d) PUT2(a,n,d), a += 2
397
398
399 /* When UTF-8 encoding is being used, a character is no longer just a single
400 byte. The macros for character handling generate simple sequences when used in
401 byte-mode, and more complicated ones for UTF-8 characters. BACKCHAR should
402 never be called in byte mode. To make sure it can never even appear when UTF-8
403 support is omitted, we don't even define it. */
404
405 #ifndef SUPPORT_UTF8
406 #define GETCHAR(c, eptr) c = *eptr;
407 #define GETCHARTEST(c, eptr) c = *eptr;
408 #define GETCHARINC(c, eptr) c = *eptr++;
409 #define GETCHARINCTEST(c, eptr) c = *eptr++;
410 #define GETCHARLEN(c, eptr, len) c = *eptr;
411 /* #define BACKCHAR(eptr) */
412
413 #else /* SUPPORT_UTF8 */
414
415 /* Get the next UTF-8 character, not advancing the pointer. This is called when
416 we know we are in UTF-8 mode. */
417
418 #define GETCHAR(c, eptr) \
419 c = *eptr; \
420 if (c >= 0xc0) \
421 { \
422 int gcii; \
423 int gcaa = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */ \
424 int gcss = 6*gcaa; \
425 c = (c & _pcre_utf8_table3[gcaa]) << gcss; \
426 for (gcii = 1; gcii <= gcaa; gcii++) \
427 { \
428 gcss -= 6; \
429 c |= (eptr[gcii] & 0x3f) << gcss; \
430 } \
431 }
432
433 /* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the
434 pointer. */
435
436 #define GETCHARTEST(c, eptr) \
437 c = *eptr; \
438 if (utf8 && c >= 0xc0) \
439 { \
440 int gcii; \
441 int gcaa = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */ \
442 int gcss = 6*gcaa; \
443 c = (c & _pcre_utf8_table3[gcaa]) << gcss; \
444 for (gcii = 1; gcii <= gcaa; gcii++) \
445 { \
446 gcss -= 6; \
447 c |= (eptr[gcii] & 0x3f) << gcss; \
448 } \
449 }
450
451 /* Get the next UTF-8 character, advancing the pointer. This is called when we
452 know we are in UTF-8 mode. */
453
454 #define GETCHARINC(c, eptr) \
455 c = *eptr++; \
456 if (c >= 0xc0) \
457 { \
458 int gcaa = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */ \
459 int gcss = 6*gcaa; \
460 c = (c & _pcre_utf8_table3[gcaa]) << gcss; \
461 while (gcaa-- > 0) \
462 { \
463 gcss -= 6; \
464 c |= (*eptr++ & 0x3f) << gcss; \
465 } \
466 }
467
468 /* Get the next character, testing for UTF-8 mode, and advancing the pointer */
469
470 #define GETCHARINCTEST(c, eptr) \
471 c = *eptr++; \
472 if (utf8 && c >= 0xc0) \
473 { \
474 int gcaa = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */ \
475 int gcss = 6*gcaa; \
476 c = (c & _pcre_utf8_table3[gcaa]) << gcss; \
477 while (gcaa-- > 0) \
478 { \
479 gcss -= 6; \
480 c |= (*eptr++ & 0x3f) << gcss; \
481 } \
482 }
483
484 /* Get the next UTF-8 character, not advancing the pointer, incrementing length
485 if there are extra bytes. This is called when we know we are in UTF-8 mode. */
486
487 #define GETCHARLEN(c, eptr, len) \
488 c = *eptr; \
489 if (c >= 0xc0) \
490 { \
491 int gcii; \
492 int gcaa = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */ \
493 int gcss = 6*gcaa; \
494 c = (c & _pcre_utf8_table3[gcaa]) << gcss; \
495 for (gcii = 1; gcii <= gcaa; gcii++) \
496 { \
497 gcss -= 6; \
498 c |= (eptr[gcii] & 0x3f) << gcss; \
499 } \
500 len += gcaa; \
501 }
502
503 /* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the
504 pointer, incrementing length if there are extra bytes. This is called when we
505 know we are in UTF-8 mode. */
506
507 #define GETCHARLENTEST(c, eptr, len) \
508 c = *eptr; \
509 if (utf8 && c >= 0xc0) \
510 { \
511 int gcii; \
512 int gcaa = _pcre_utf8_table4[c & 0x3f]; /* Number of additional bytes */ \
513 int gcss = 6*gcaa; \
514 c = (c & _pcre_utf8_table3[gcaa]) << gcss; \
515 for (gcii = 1; gcii <= gcaa; gcii++) \
516 { \
517 gcss -= 6; \
518 c |= (eptr[gcii] & 0x3f) << gcss; \
519 } \
520 len += gcaa; \
521 }
522
523 /* If the pointer is not at the start of a character, move it back until
524 it is. This is called only in UTF-8 mode - we don't put a test within the macro
525 because almost all calls are already within a block of UTF-8 only code. */
526
527 #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--
528
529 #endif
530
531
532 /* In case there is no definition of offsetof() provided - though any proper
533 Standard C system should have one. */
534
535 #ifndef offsetof
536 #define offsetof(p_type,field) ((size_t)&(((p_type *)0)->field))
537 #endif
538
539
540 /* These are the public options that can change during matching. */
541
542 #define PCRE_IMS (PCRE_CASELESS|PCRE_MULTILINE|PCRE_DOTALL)
543
544 /* Private flags containing information about the compiled regex. They used to
545 live at the top end of the options word, but that got almost full, so now they
546 are in a 16-bit flags word. */
547
548 #define PCRE_NOPARTIAL 0x0001 /* can't use partial with this regex */
549 #define PCRE_FIRSTSET 0x0002 /* first_byte is set */
550 #define PCRE_REQCHSET 0x0004 /* req_byte is set */
551 #define PCRE_STARTLINE 0x0008 /* start after \n for multiline */
552 #define PCRE_JCHANGED 0x0010 /* j option used in regex */
553 #define PCRE_HASCRORLF 0x0020 /* explicit \r or \n in pattern */
554
555 /* Options for the "extra" block produced by pcre_study(). */
556
557 #define PCRE_STUDY_MAPPED 0x01 /* a map of starting chars exists */
558
559 /* Masks for identifying the public options that are permitted at compile
560 time, run time, or study time, respectively. */
561
562 #define PCRE_NEWLINE_BITS (PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_ANY| \
563 PCRE_NEWLINE_ANYCRLF)
564
565 #define PUBLIC_COMPILE_OPTIONS \
566 (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \
567 PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \
568 PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \
569 PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
570 PCRE_JAVASCRIPT_COMPAT)
571
572 #define PUBLIC_EXEC_OPTIONS \
573 (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \
574 PCRE_PARTIAL|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
575 PCRE_NO_START_OPTIMIZE)
576
577 #define PUBLIC_DFA_EXEC_OPTIONS \
578 (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \
579 PCRE_PARTIAL|PCRE_DFA_SHORTEST|PCRE_DFA_RESTART|PCRE_NEWLINE_BITS| \
580 PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE|PCRE_NO_START_OPTIMIZE)
581
582 #define PUBLIC_STUDY_OPTIONS 0 /* None defined */
583
584 /* Magic number to provide a small check against being handed junk. Also used
585 to detect whether a pattern was compiled on a host of different endianness. */
586
587 #define MAGIC_NUMBER 0x50435245UL /* 'PCRE' */
588
589 /* Negative values for the firstchar and reqchar variables */
590
591 #define REQ_UNSET (-2)
592 #define REQ_NONE (-1)
593
594 /* The maximum remaining length of subject we are prepared to search for a
595 req_byte match. */
596
597 #define REQ_BYTE_MAX 1000
598
599 /* Flags added to firstbyte or reqbyte; a "non-literal" item is either a
600 variable-length repeat, or a anything other than literal characters. */
601
602 #define REQ_CASELESS 0x0100 /* indicates caselessness */
603 #define REQ_VARY 0x0200 /* reqbyte followed non-literal item */
604
605 /* Miscellaneous definitions. The #ifndef is to pacify compiler warnings in
606 environments where these macros are defined elsewhere. Unfortunately, there
607 is no way to do the same for the typedef. */
608
609 typedef int BOOL;
610
611 #ifndef FALSE
612 #define FALSE 0
613 #define TRUE 1
614 #endif
615
616 /* If PCRE is to support UTF-8 on EBCDIC platforms, we cannot use normal
617 character constants like '*' because the compiler would emit their EBCDIC code,
618 which is different from their ASCII/UTF-8 code. Instead we define macros for
619 the characters so that they always use the ASCII/UTF-8 code when UTF-8 support
620 is enabled. When UTF-8 support is not enabled, the definitions use character
621 literals. Both character and string versions of each character are needed, and
622 there are some longer strings as well.
623
624 This means that, on EBCDIC platforms, the PCRE library can handle either
625 EBCDIC, or UTF-8, but not both. To support both in the same compiled library
626 would need different lookups depending on whether PCRE_UTF8 was set or not.
627 This would make it impossible to use characters in switch/case statements,
628 which would reduce performance. For a theoretical use (which nobody has asked
629 for) in a minority area (EBCDIC platforms), this is not sensible. Any
630 application that did need both could compile two versions of the library, using
631 macros to give the functions distinct names. */
632
633 #ifndef SUPPORT_UTF8
634
635 /* UTF-8 support is not enabled; use the platform-dependent character literals
636 so that PCRE works on both ASCII and EBCDIC platforms, in non-UTF-mode only. */
637
638 #define CHAR_HT '\t'
639 #define CHAR_VT '\v'
640 #define CHAR_FF '\f'
641 #define CHAR_CR '\r'
642 #define CHAR_NL '\n'
643 #define CHAR_BS '\b'
644 #define CHAR_BEL '\a'
645 #ifdef EBCDIC
646 #define CHAR_ESC '\047'
647 #define CHAR_DEL '\007'
648 #else
649 #define CHAR_ESC '\033'
650 #define CHAR_DEL '\177'
651 #endif
652
653 #define CHAR_SPACE ' '
654 #define CHAR_EXCLAMATION_MARK '!'
655 #define CHAR_QUOTATION_MARK '"'
656 #define CHAR_NUMBER_SIGN '#'
657 #define CHAR_DOLLAR_SIGN '$'
658 #define CHAR_PERCENT_SIGN '%'
659 #define CHAR_AMPERSAND '&'
660 #define CHAR_APOSTROPHE '\''
661 #define CHAR_LEFT_PARENTHESIS '('
662 #define CHAR_RIGHT_PARENTHESIS ')'
663 #define CHAR_ASTERISK '*'
664 #define CHAR_PLUS '+'
665 #define CHAR_COMMA ','
666 #define CHAR_MINUS '-'
667 #define CHAR_DOT '.'
668 #define CHAR_SLASH '/'
669 #define CHAR_0 '0'
670 #define CHAR_1 '1'
671 #define CHAR_2 '2'
672 #define CHAR_3 '3'
673 #define CHAR_4 '4'
674 #define CHAR_5 '5'
675 #define CHAR_6 '6'
676 #define CHAR_7 '7'
677 #define CHAR_8 '8'
678 #define CHAR_9 '9'
679 #define CHAR_COLON ':'
680 #define CHAR_SEMICOLON ';'
681 #define CHAR_LESS_THAN_SIGN '<'
682 #define CHAR_EQUALS_SIGN '='
683 #define CHAR_GREATER_THAN_SIGN '>'
684 #define CHAR_QUESTION_MARK '?'
685 #define CHAR_COMMERCIAL_AT '@'
686 #define CHAR_A 'A'
687 #define CHAR_B 'B'
688 #define CHAR_C 'C'
689 #define CHAR_D 'D'
690 #define CHAR_E 'E'
691 #define CHAR_F 'F'
692 #define CHAR_G 'G'
693 #define CHAR_H 'H'
694 #define CHAR_I 'I'
695 #define CHAR_J 'J'
696 #define CHAR_K 'K'
697 #define CHAR_L 'L'
698 #define CHAR_M 'M'
699 #define CHAR_N 'N'
700 #define CHAR_O 'O'
701 #define CHAR_P 'P'
702 #define CHAR_Q 'Q'
703 #define CHAR_R 'R'
704 #define CHAR_S 'S'
705 #define CHAR_T 'T'
706 #define CHAR_U 'U'
707 #define CHAR_V 'V'
708 #define CHAR_W 'W'
709 #define CHAR_X 'X'
710 #define CHAR_Y 'Y'
711 #define CHAR_Z 'Z'
712 #define CHAR_LEFT_SQUARE_BRACKET '['
713 #define CHAR_BACKSLASH '\\'
714 #define CHAR_RIGHT_SQUARE_BRACKET ']'
715 #define CHAR_CIRCUMFLEX_ACCENT '^'
716 #define CHAR_UNDERSCORE '_'
717 #define CHAR_GRAVE_ACCENT '`'
718 #define CHAR_a 'a'
719 #define CHAR_b 'b'
720 #define CHAR_c 'c'
721 #define CHAR_d 'd'
722 #define CHAR_e 'e'
723 #define CHAR_f 'f'
724 #define CHAR_g 'g'
725 #define CHAR_h 'h'
726 #define CHAR_i 'i'
727 #define CHAR_j 'j'
728 #define CHAR_k 'k'
729 #define CHAR_l 'l'
730 #define CHAR_m 'm'
731 #define CHAR_n 'n'
732 #define CHAR_o 'o'
733 #define CHAR_p 'p'
734 #define CHAR_q 'q'
735 #define CHAR_r 'r'
736 #define CHAR_s 's'
737 #define CHAR_t 't'
738 #define CHAR_u 'u'
739 #define CHAR_v 'v'
740 #define CHAR_w 'w'
741 #define CHAR_x 'x'
742 #define CHAR_y 'y'
743 #define CHAR_z 'z'
744 #define CHAR_LEFT_CURLY_BRACKET '{'
745 #define CHAR_VERTICAL_LINE '|'
746 #define CHAR_RIGHT_CURLY_BRACKET '}'
747 #define CHAR_TILDE '~'
748
749 #define STR_HT "\t"
750 #define STR_VT "\v"
751 #define STR_FF "\f"
752 #define STR_CR "\r"
753 #define STR_NL "\n"
754 #define STR_BS "\b"
755 #define STR_BEL "\a"
756 #ifdef EBCDIC
757 #define STR_ESC "\047"
758 #define STR_DEL "\007"
759 #else
760 #define STR_ESC "\033"
761 #define STR_DEL "\177"
762 #endif
763
764 #define STR_SPACE " "
765 #define STR_EXCLAMATION_MARK "!"
766 #define STR_QUOTATION_MARK "\""
767 #define STR_NUMBER_SIGN "#"
768 #define STR_DOLLAR_SIGN "$"
769 #define STR_PERCENT_SIGN "%"
770 #define STR_AMPERSAND "&"
771 #define STR_APOSTROPHE "'"
772 #define STR_LEFT_PARENTHESIS "("
773 #define STR_RIGHT_PARENTHESIS ")"
774 #define STR_ASTERISK "*"
775 #define STR_PLUS "+"
776 #define STR_COMMA ","
777 #define STR_MINUS "-"
778 #define STR_DOT "."
779 #define STR_SLASH "/"
780 #define STR_0 "0"
781 #define STR_1 "1"
782 #define STR_2 "2"
783 #define STR_3 "3"
784 #define STR_4 "4"
785 #define STR_5 "5"
786 #define STR_6 "6"
787 #define STR_7 "7"
788 #define STR_8 "8"
789 #define STR_9 "9"
790 #define STR_COLON ":"
791 #define STR_SEMICOLON ";"
792 #define STR_LESS_THAN_SIGN "<"
793 #define STR_EQUALS_SIGN "="
794 #define STR_GREATER_THAN_SIGN ">"
795 #define STR_QUESTION_MARK "?"
796 #define STR_COMMERCIAL_AT "@"
797 #define STR_A "A"
798 #define STR_B "B"
799 #define STR_C "C"
800 #define STR_D "D"
801 #define STR_E "E"
802 #define STR_F "F"
803 #define STR_G "G"
804 #define STR_H "H"
805 #define STR_I "I"
806 #define STR_J "J"
807 #define STR_K "K"
808 #define STR_L "L"
809 #define STR_M "M"
810 #define STR_N "N"
811 #define STR_O "O"
812 #define STR_P "P"
813 #define STR_Q "Q"
814 #define STR_R "R"
815 #define STR_S "S"
816 #define STR_T "T"
817 #define STR_U "U"
818 #define STR_V "V"
819 #define STR_W "W"
820 #define STR_X "X"
821 #define STR_Y "Y"
822 #define STR_Z "Z"
823 #define STR_LEFT_SQUARE_BRACKET "["
824 #define STR_BACKSLASH "\\"
825 #define STR_RIGHT_SQUARE_BRACKET "]"
826 #define STR_CIRCUMFLEX_ACCENT "^"
827 #define STR_UNDERSCORE "_"
828 #define STR_GRAVE_ACCENT "`"
829 #define STR_a "a"
830 #define STR_b "b"
831 #define STR_c "c"
832 #define STR_d "d"
833 #define STR_e "e"
834 #define STR_f "f"
835 #define STR_g "g"
836 #define STR_h "h"
837 #define STR_i "i"
838 #define STR_j "j"
839 #define STR_k "k"
840 #define STR_l "l"
841 #define STR_m "m"
842 #define STR_n "n"
843 #define STR_o "o"
844 #define STR_p "p"
845 #define STR_q "q"
846 #define STR_r "r"
847 #define STR_s "s"
848 #define STR_t "t"
849 #define STR_u "u"
850 #define STR_v "v"
851 #define STR_w "w"
852 #define STR_x "x"
853 #define STR_y "y"
854 #define STR_z "z"
855 #define STR_LEFT_CURLY_BRACKET "{"
856 #define STR_VERTICAL_LINE "|"
857 #define STR_RIGHT_CURLY_BRACKET "}"
858 #define STR_TILDE "~"
859
860 #define STRING_ACCEPT0 "ACCEPT\0"
861 #define STRING_COMMIT0 "COMMIT\0"
862 #define STRING_F0 "F\0"
863 #define STRING_FAIL0 "FAIL\0"
864 #define STRING_PRUNE0 "PRUNE\0"
865 #define STRING_SKIP0 "SKIP\0"
866 #define STRING_THEN "THEN"
867
868 #define STRING_alpha0 "alpha\0"
869 #define STRING_lower0 "lower\0"
870 #define STRING_upper0 "upper\0"
871 #define STRING_alnum0 "alnum\0"
872 #define STRING_ascii0 "ascii\0"
873 #define STRING_blank0 "blank\0"
874 #define STRING_cntrl0 "cntrl\0"
875 #define STRING_digit0 "digit\0"
876 #define STRING_graph0 "graph\0"
877 #define STRING_print0 "print\0"
878 #define STRING_punct0 "punct\0"
879 #define STRING_space0 "space\0"
880 #define STRING_word0 "word\0"
881 #define STRING_xdigit "xdigit"
882
883 #define STRING_DEFINE "DEFINE"
884
885 #define STRING_CR_RIGHTPAR "CR)"
886 #define STRING_LF_RIGHTPAR "LF)"
887 #define STRING_CRLF_RIGHTPAR "CRLF)"
888 #define STRING_ANY_RIGHTPAR "ANY)"
889 #define STRING_ANYCRLF_RIGHTPAR "ANYCRLF)"
890 #define STRING_BSR_ANYCRLF_RIGHTPAR "BSR_ANYCRLF)"
891 #define STRING_BSR_UNICODE_RIGHTPAR "BSR_UNICODE)"
892 #define STRING_UTF8_RIGHTPAR "UTF8)"
893
894 #else /* SUPPORT_UTF8 */
895
896 /* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This
897 works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode
898 only. */
899
900 #define CHAR_HT '\011'
901 #define CHAR_VT '\013'
902 #define CHAR_FF '\014'
903 #define CHAR_CR '\015'
904 #define CHAR_NL '\012'
905 #define CHAR_BS '\010'
906 #define CHAR_BEL '\007'
907 #define CHAR_ESC '\033'
908 #define CHAR_DEL '\177'
909
910 #define CHAR_SPACE '\040'
911 #define CHAR_EXCLAMATION_MARK '\041'
912 #define CHAR_QUOTATION_MARK '\042'
913 #define CHAR_NUMBER_SIGN '\043'
914 #define CHAR_DOLLAR_SIGN '\044'
915 #define CHAR_PERCENT_SIGN '\045'
916 #define CHAR_AMPERSAND '\046'
917 #define CHAR_APOSTROPHE '\047'
918 #define CHAR_LEFT_PARENTHESIS '\050'
919 #define CHAR_RIGHT_PARENTHESIS '\051'
920 #define CHAR_ASTERISK '\052'
921 #define CHAR_PLUS '\053'
922 #define CHAR_COMMA '\054'
923 #define CHAR_MINUS '\055'
924 #define CHAR_DOT '\056'
925 #define CHAR_SLASH '\057'
926 #define CHAR_0 '\060'
927 #define CHAR_1 '\061'
928 #define CHAR_2 '\062'
929 #define CHAR_3 '\063'
930 #define CHAR_4 '\064'
931 #define CHAR_5 '\065'
932 #define CHAR_6 '\066'
933 #define CHAR_7 '\067'
934 #define CHAR_8 '\070'
935 #define CHAR_9 '\071'
936 #define CHAR_COLON '\072'
937 #define CHAR_SEMICOLON '\073'
938 #define CHAR_LESS_THAN_SIGN '\074'
939 #define CHAR_EQUALS_SIGN '\075'
940 #define CHAR_GREATER_THAN_SIGN '\076'
941 #define CHAR_QUESTION_MARK '\077'
942 #define CHAR_COMMERCIAL_AT '\100'
943 #define CHAR_A '\101'
944 #define CHAR_B '\102'
945 #define CHAR_C '\103'
946 #define CHAR_D '\104'
947 #define CHAR_E '\105'
948 #define CHAR_F '\106'
949 #define CHAR_G '\107'
950 #define CHAR_H '\110'
951 #define CHAR_I '\111'
952 #define CHAR_J '\112'
953 #define CHAR_K '\113'
954 #define CHAR_L '\114'
955 #define CHAR_M '\115'
956 #define CHAR_N '\116'
957 #define CHAR_O '\117'
958 #define CHAR_P '\120'
959 #define CHAR_Q '\121'
960 #define CHAR_R '\122'
961 #define CHAR_S '\123'
962 #define CHAR_T '\124'
963 #define CHAR_U '\125'
964 #define CHAR_V '\126'
965 #define CHAR_W '\127'
966 #define CHAR_X '\130'
967 #define CHAR_Y '\131'
968 #define CHAR_Z '\132'
969 #define CHAR_LEFT_SQUARE_BRACKET '\133'
970 #define CHAR_BACKSLASH '\134'
971 #define CHAR_RIGHT_SQUARE_BRACKET '\135'
972 #define CHAR_CIRCUMFLEX_ACCENT '\136'
973 #define CHAR_UNDERSCORE '\137'
974 #define CHAR_GRAVE_ACCENT '\140'
975 #define CHAR_a '\141'
976 #define CHAR_b '\142'
977 #define CHAR_c '\143'
978 #define CHAR_d '\144'
979 #define CHAR_e '\145'
980 #define CHAR_f '\146'
981 #define CHAR_g '\147'
982 #define CHAR_h '\150'
983 #define CHAR_i '\151'
984 #define CHAR_j '\152'
985 #define CHAR_k '\153'
986 #define CHAR_l '\154'
987 #define CHAR_m '\155'
988 #define CHAR_n '\156'
989 #define CHAR_o '\157'
990 #define CHAR_p '\160'
991 #define CHAR_q '\161'
992 #define CHAR_r '\162'
993 #define CHAR_s '\163'
994 #define CHAR_t '\164'
995 #define CHAR_u '\165'
996 #define CHAR_v '\166'
997 #define CHAR_w '\167'
998 #define CHAR_x '\170'
999 #define CHAR_y '\171'
1000 #define CHAR_z '\172'
1001 #define CHAR_LEFT_CURLY_BRACKET '\173'
1002 #define CHAR_VERTICAL_LINE '\174'
1003 #define CHAR_RIGHT_CURLY_BRACKET '\175'
1004 #define CHAR_TILDE '\176'
1005
1006 #define STR_HT "\011"
1007 #define STR_VT "\013"
1008 #define STR_FF "\014"
1009 #define STR_CR "\015"
1010 #define STR_NL "\012"
1011 #define STR_BS "\010"
1012 #define STR_BEL "\007"
1013 #define STR_ESC "\033"
1014 #define STR_DEL "\177"
1015
1016 #define STR_SPACE "\040"
1017 #define STR_EXCLAMATION_MARK "\041"
1018 #define STR_QUOTATION_MARK "\042"
1019 #define STR_NUMBER_SIGN "\043"
1020 #define STR_DOLLAR_SIGN "\044"
1021 #define STR_PERCENT_SIGN "\045"
1022 #define STR_AMPERSAND "\046"
1023 #define STR_APOSTROPHE "\047"
1024 #define STR_LEFT_PARENTHESIS "\050"
1025 #define STR_RIGHT_PARENTHESIS "\051"
1026 #define STR_ASTERISK "\052"
1027 #define STR_PLUS "\053"
1028 #define STR_COMMA "\054"
1029 #define STR_MINUS "\055"
1030 #define STR_DOT "\056"
1031 #define STR_SLASH "\057"
1032 #define STR_0 "\060"
1033 #define STR_1 "\061"
1034 #define STR_2 "\062"
1035 #define STR_3 "\063"
1036 #define STR_4 "\064"
1037 #define STR_5 "\065"
1038 #define STR_6 "\066"
1039 #define STR_7 "\067"
1040 #define STR_8 "\070"
1041 #define STR_9 "\071"
1042 #define STR_COLON "\072"
1043 #define STR_SEMICOLON "\073"
1044 #define STR_LESS_THAN_SIGN "\074"
1045 #define STR_EQUALS_SIGN "\075"
1046 #define STR_GREATER_THAN_SIGN "\076"
1047 #define STR_QUESTION_MARK "\077"
1048 #define STR_COMMERCIAL_AT "\100"
1049 #define STR_A "\101"
1050 #define STR_B "\102"
1051 #define STR_C "\103"
1052 #define STR_D "\104"
1053 #define STR_E "\105"
1054 #define STR_F "\106"
1055 #define STR_G "\107"
1056 #define STR_H "\110"
1057 #define STR_I "\111"
1058 #define STR_J "\112"
1059 #define STR_K "\113"
1060 #define STR_L "\114"
1061 #define STR_M "\115"
1062 #define STR_N "\116"
1063 #define STR_O "\117"
1064 #define STR_P "\120"
1065 #define STR_Q "\121"
1066 #define STR_R "\122"
1067 #define STR_S "\123"
1068 #define STR_T "\124"
1069 #define STR_U "\125"
1070 #define STR_V "\126"
1071 #define STR_W "\127"
1072 #define STR_X "\130"
1073 #define STR_Y "\131"
1074 #define STR_Z "\132"
1075 #define STR_LEFT_SQUARE_BRACKET "\133"
1076 #define STR_BACKSLASH "\134"
1077 #define STR_RIGHT_SQUARE_BRACKET "\135"
1078 #define STR_CIRCUMFLEX_ACCENT "\136"
1079 #define STR_UNDERSCORE "\137"
1080 #define STR_GRAVE_ACCENT "\140"
1081 #define STR_a "\141"
1082 #define STR_b "\142"
1083 #define STR_c "\143"
1084 #define STR_d "\144"
1085 #define STR_e "\145"
1086 #define STR_f "\146"
1087 #define STR_g "\147"
1088 #define STR_h "\150"
1089 #define STR_i "\151"
1090 #define STR_j "\152"
1091 #define STR_k "\153"
1092 #define STR_l "\154"
1093 #define STR_m "\155"
1094 #define STR_n "\156"
1095 #define STR_o "\157"
1096 #define STR_p "\160"
1097 #define STR_q "\161"
1098 #define STR_r "\162"
1099 #define STR_s "\163"
1100 #define STR_t "\164"
1101 #define STR_u "\165"
1102 #define STR_v "\166"
1103 #define STR_w "\167"
1104 #define STR_x "\170"
1105 #define STR_y "\171"
1106 #define STR_z "\172"
1107 #define STR_LEFT_CURLY_BRACKET "\173"
1108 #define STR_VERTICAL_LINE "\174"
1109 #define STR_RIGHT_CURLY_BRACKET "\175"
1110 #define STR_TILDE "\176"
1111
1112 #define STRING_ACCEPT0 STR_A STR_C STR_C STR_E STR_P STR_T "\0"
1113 #define STRING_COMMIT0 STR_C STR_O STR_M STR_M STR_I STR_T "\0"
1114 #define STRING_F0 STR_F "\0"
1115 #define STRING_FAIL0 STR_F STR_A STR_I STR_L "\0"
1116 #define STRING_PRUNE0 STR_P STR_R STR_U STR_N STR_E "\0"
1117 #define STRING_SKIP0 STR_S STR_K STR_I STR_P "\0"
1118 #define STRING_THEN STR_T STR_H STR_E STR_N
1119
1120 #define STRING_alpha0 STR_a STR_l STR_p STR_h STR_a "\0"
1121 #define STRING_lower0 STR_l STR_o STR_w STR_e STR_r "\0"
1122 #define STRING_upper0 STR_u STR_p STR_p STR_e STR_r "\0"
1123 #define STRING_alnum0 STR_a STR_l STR_n STR_u STR_m "\0"
1124 #define STRING_ascii0 STR_a STR_s STR_c STR_i STR_i "\0"
1125 #define STRING_blank0 STR_b STR_l STR_a STR_n STR_k "\0"
1126 #define STRING_cntrl0 STR_c STR_n STR_t STR_r STR_l "\0"
1127 #define STRING_digit0 STR_d STR_i STR_g STR_i STR_t "\0"
1128 #define STRING_graph0 STR_g STR_r STR_a STR_p STR_h "\0"
1129 #define STRING_print0 STR_p STR_r STR_i STR_n STR_t "\0"
1130 #define STRING_punct0 STR_p STR_u STR_n STR_c STR_t "\0"
1131 #define STRING_space0 STR_s STR_p STR_a STR_c STR_e "\0"
1132 #define STRING_word0 STR_w STR_o STR_r STR_d "\0"
1133 #define STRING_xdigit STR_x STR_d STR_i STR_g STR_i STR_t
1134
1135 #define STRING_DEFINE STR_D STR_E STR_F STR_I STR_N STR_E
1136
1137 #define STRING_CR_RIGHTPAR STR_C STR_R STR_RIGHT_PARENTHESIS
1138 #define STRING_LF_RIGHTPAR STR_L STR_F STR_RIGHT_PARENTHESIS
1139 #define STRING_CRLF_RIGHTPAR STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1140 #define STRING_ANY_RIGHTPAR STR_A STR_N STR_Y STR_RIGHT_PARENTHESIS
1141 #define STRING_ANYCRLF_RIGHTPAR STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1142 #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
1143 #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
1144 #define STRING_UTF8_RIGHTPAR STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS
1145
1146 #endif /* SUPPORT_UTF8 */
1147
1148 /* Escape items that are just an encoding of a particular data value. */
1149
1150 #ifndef ESC_e
1151 #define ESC_e CHAR_ESC
1152 #endif
1153
1154 #ifndef ESC_f
1155 #define ESC_f CHAR_FF
1156 #endif
1157
1158 #ifndef ESC_n
1159 #define ESC_n CHAR_NL
1160 #endif
1161
1162 #ifndef ESC_r
1163 #define ESC_r CHAR_CR
1164 #endif
1165
1166 /* We can't officially use ESC_t because it is a POSIX reserved identifier
1167 (presumably because of all the others like size_t). */
1168
1169 #ifndef ESC_tee
1170 #define ESC_tee CHAR_HT
1171 #endif
1172
1173 /* Codes for different types of Unicode property */
1174
1175 #define PT_ANY 0 /* Any property - matches all chars */
1176 #define PT_LAMP 1 /* L& - the union of Lu, Ll, Lt */
1177 #define PT_GC 2 /* General characteristic (e.g. L) */
1178 #define PT_PC 3 /* Particular characteristic (e.g. Lu) */
1179 #define PT_SC 4 /* Script (e.g. Han) */
1180
1181 /* Flag bits and data types for the extended class (OP_XCLASS) for classes that
1182 contain UTF-8 characters with values greater than 255. */
1183
1184 #define XCL_NOT 0x01 /* Flag: this is a negative class */
1185 #define XCL_MAP 0x02 /* Flag: a 32-byte map is present */
1186
1187 #define XCL_END 0 /* Marks end of individual items */
1188 #define XCL_SINGLE 1 /* Single item (one multibyte char) follows */
1189 #define XCL_RANGE 2 /* A range (two multibyte chars) follows */
1190 #define XCL_PROP 3 /* Unicode property (2-byte property code follows) */
1191 #define XCL_NOTPROP 4 /* Unicode inverted property (ditto) */
1192
1193 /* These are escaped items that aren't just an encoding of a particular data
1194 value such as \n. They must have non-zero values, as check_escape() returns
1195 their negation. Also, they must appear in the same order as in the opcode
1196 definitions below, up to ESC_z. There's a dummy for OP_ANY because it
1197 corresponds to "." rather than an escape sequence, and another for OP_ALLANY
1198 (which is used for [^] in JavaScript compatibility mode).
1199
1200 The final escape must be ESC_REF as subsequent values are used for
1201 backreferences (\1, \2, \3, etc). There are two tests in the code for an escape
1202 greater than ESC_b and less than ESC_Z to detect the types that may be
1203 repeated. These are the types that consume characters. If any new escapes are
1204 put in between that don't consume a character, that code will have to change.
1205 */
1206
1207 enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s,
1208 ESC_W, ESC_w, ESC_dum1, ESC_dum2, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H,
1209 ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z, ESC_E, ESC_Q, ESC_g, ESC_k,
1210 ESC_REF };
1211
1212
1213 /* Opcode table: Starting from 1 (i.e. after OP_END), the values up to
1214 OP_EOD must correspond in order to the list of escapes immediately above.
1215
1216 *** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions
1217 that follow must also be updated to match. There is also a table called
1218 "coptable" in pcre_dfa_exec.c that must be updated. */
1219
1220 enum {
1221 OP_END, /* 0 End of pattern */
1222
1223 /* Values corresponding to backslashed metacharacters */
1224
1225 OP_SOD, /* 1 Start of data: \A */
1226 OP_SOM, /* 2 Start of match (subject + offset): \G */
1227 OP_SET_SOM, /* 3 Set start of match (\K) */
1228 OP_NOT_WORD_BOUNDARY, /* 4 \B */
1229 OP_WORD_BOUNDARY, /* 5 \b */
1230 OP_NOT_DIGIT, /* 6 \D */
1231 OP_DIGIT, /* 7 \d */
1232 OP_NOT_WHITESPACE, /* 8 \S */
1233 OP_WHITESPACE, /* 9 \s */
1234 OP_NOT_WORDCHAR, /* 10 \W */
1235 OP_WORDCHAR, /* 11 \w */
1236 OP_ANY, /* 12 Match any character (subject to DOTALL) */
1237 OP_ALLANY, /* 13 Match any character (not subject to DOTALL) */
1238 OP_ANYBYTE, /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */
1239 OP_NOTPROP, /* 15 \P (not Unicode property) */
1240 OP_PROP, /* 16 \p (Unicode property) */
1241 OP_ANYNL, /* 17 \R (any newline sequence) */
1242 OP_NOT_HSPACE, /* 18 \H (not horizontal whitespace) */
1243 OP_HSPACE, /* 19 \h (horizontal whitespace) */
1244 OP_NOT_VSPACE, /* 20 \V (not vertical whitespace) */
1245 OP_VSPACE, /* 21 \v (vertical whitespace) */
1246 OP_EXTUNI, /* 22 \X (extended Unicode sequence */
1247 OP_EODN, /* 23 End of data or \n at end of data: \Z. */
1248 OP_EOD, /* 24 End of data: \z */
1249
1250 OP_OPT, /* 25 Set runtime options */
1251 OP_CIRC, /* 26 Start of line - varies with multiline switch */
1252 OP_DOLL, /* 27 End of line - varies with multiline switch */
1253 OP_CHAR, /* 28 Match one character, casefully */
1254 OP_CHARNC, /* 29 Match one character, caselessly */
1255 OP_NOT, /* 30 Match one character, not the following one */
1256
1257 OP_STAR, /* 31 The maximizing and minimizing versions of */
1258 OP_MINSTAR, /* 32 these six opcodes must come in pairs, with */
1259 OP_PLUS, /* 33 the minimizing one second. */
1260 OP_MINPLUS, /* 34 This first set applies to single characters.*/
1261 OP_QUERY, /* 35 */
1262 OP_MINQUERY, /* 36 */
1263
1264 OP_UPTO, /* 37 From 0 to n matches */
1265 OP_MINUPTO, /* 38 */
1266 OP_EXACT, /* 39 Exactly n matches */
1267
1268 OP_POSSTAR, /* 40 Possessified star */
1269 OP_POSPLUS, /* 41 Possessified plus */
1270 OP_POSQUERY, /* 42 Posesssified query */
1271 OP_POSUPTO, /* 43 Possessified upto */
1272
1273 OP_NOTSTAR, /* 44 The maximizing and minimizing versions of */
1274 OP_NOTMINSTAR, /* 45 these six opcodes must come in pairs, with */
1275 OP_NOTPLUS, /* 46 the minimizing one second. They must be in */
1276 OP_NOTMINPLUS, /* 47 exactly the same order as those above. */
1277 OP_NOTQUERY, /* 48 This set applies to "not" single characters. */
1278 OP_NOTMINQUERY, /* 49 */
1279
1280 OP_NOTUPTO, /* 50 From 0 to n matches */
1281 OP_NOTMINUPTO, /* 51 */
1282 OP_NOTEXACT, /* 52 Exactly n matches */
1283
1284 OP_NOTPOSSTAR, /* 53 Possessified versions */
1285 OP_NOTPOSPLUS, /* 54 */
1286 OP_NOTPOSQUERY, /* 55 */
1287 OP_NOTPOSUPTO, /* 56 */
1288
1289 OP_TYPESTAR, /* 57 The maximizing and minimizing versions of */
1290 OP_TYPEMINSTAR, /* 58 these six opcodes must come in pairs, with */
1291 OP_TYPEPLUS, /* 59 the minimizing one second. These codes must */
1292 OP_TYPEMINPLUS, /* 60 be in exactly the same order as those above. */
1293 OP_TYPEQUERY, /* 61 This set applies to character types such as \d */
1294 OP_TYPEMINQUERY, /* 62 */
1295
1296 OP_TYPEUPTO, /* 63 From 0 to n matches */
1297 OP_TYPEMINUPTO, /* 64 */
1298 OP_TYPEEXACT, /* 65 Exactly n matches */
1299
1300 OP_TYPEPOSSTAR, /* 66 Possessified versions */
1301 OP_TYPEPOSPLUS, /* 67 */
1302 OP_TYPEPOSQUERY, /* 68 */
1303 OP_TYPEPOSUPTO, /* 69 */
1304
1305 OP_CRSTAR, /* 70 The maximizing and minimizing versions of */
1306 OP_CRMINSTAR, /* 71 all these opcodes must come in pairs, with */
1307 OP_CRPLUS, /* 72 the minimizing one second. These codes must */
1308 OP_CRMINPLUS, /* 73 be in exactly the same order as those above. */
1309 OP_CRQUERY, /* 74 These are for character classes and back refs */
1310 OP_CRMINQUERY, /* 75 */
1311 OP_CRRANGE, /* 76 These are different to the three sets above. */
1312 OP_CRMINRANGE, /* 77 */
1313
1314 OP_CLASS, /* 78 Match a character class, chars < 256 only */
1315 OP_NCLASS, /* 79 Same, but the bitmap was created from a negative
1316 class - the difference is relevant only when a UTF-8
1317 character > 255 is encountered. */
1318
1319 OP_XCLASS, /* 80 Extended class for handling UTF-8 chars within the
1320 class. This does both positive and negative. */
1321
1322 OP_REF, /* 81 Match a back reference */
1323 OP_RECURSE, /* 82 Match a numbered subpattern (possibly recursive) */
1324 OP_CALLOUT, /* 83 Call out to external function if provided */
1325
1326 OP_ALT, /* 84 Start of alternation */
1327 OP_KET, /* 85 End of group that doesn't have an unbounded repeat */
1328 OP_KETRMAX, /* 86 These two must remain together and in this */
1329 OP_KETRMIN, /* 87 order. They are for groups the repeat for ever. */
1330
1331 /* The assertions must come before BRA, CBRA, ONCE, and COND.*/
1332
1333 OP_ASSERT, /* 88 Positive lookahead */
1334 OP_ASSERT_NOT, /* 89 Negative lookahead */
1335 OP_ASSERTBACK, /* 90 Positive lookbehind */
1336 OP_ASSERTBACK_NOT, /* 91 Negative lookbehind */
1337 OP_REVERSE, /* 92 Move pointer back - used in lookbehind assertions */
1338
1339 /* ONCE, BRA, CBRA, and COND must come after the assertions, with ONCE first,
1340 as there's a test for >= ONCE for a subpattern that isn't an assertion. */
1341
1342 OP_ONCE, /* 93 Atomic group */
1343 OP_BRA, /* 94 Start of non-capturing bracket */
1344 OP_CBRA, /* 95 Start of capturing bracket */
1345 OP_COND, /* 96 Conditional group */
1346
1347 /* These three must follow the previous three, in the same order. There's a
1348 check for >= SBRA to distinguish the two sets. */
1349
1350 OP_SBRA, /* 97 Start of non-capturing bracket, check empty */
1351 OP_SCBRA, /* 98 Start of capturing bracket, check empty */
1352 OP_SCOND, /* 99 Conditional group, check empty */
1353
1354 OP_CREF, /* 100 Used to hold a capture number as condition */
1355 OP_RREF, /* 101 Used to hold a recursion number as condition */
1356 OP_DEF, /* 102 The DEFINE condition */
1357
1358 OP_BRAZERO, /* 103 These two must remain together and in this */
1359 OP_BRAMINZERO, /* 104 order. */
1360
1361 /* These are backtracking control verbs */
1362
1363 OP_PRUNE, /* 105 */
1364 OP_SKIP, /* 106 */
1365 OP_THEN, /* 107 */
1366 OP_COMMIT, /* 108 */
1367
1368 /* These are forced failure and success verbs */
1369
1370 OP_FAIL, /* 109 */
1371 OP_ACCEPT, /* 110 */
1372
1373 /* This is used to skip a subpattern with a {0} quantifier */
1374
1375 OP_SKIPZERO /* 111 */
1376 };
1377
1378
1379 /* This macro defines textual names for all the opcodes. These are used only
1380 for debugging. The macro is referenced only in pcre_printint.c. */
1381
1382 #define OP_NAME_LIST \
1383 "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d", \
1384 "\\S", "\\s", "\\W", "\\w", "Any", "AllAny", "Anybyte", \
1385 "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v", \
1386 "extuni", "\\Z", "\\z", \
1387 "Opt", "^", "$", "char", "charnc", "not", \
1388 "*", "*?", "+", "+?", "?", "??", "{", "{", "{", \
1389 "*+","++", "?+", "{", \
1390 "*", "*?", "+", "+?", "?", "??", "{", "{", "{", \
1391 "*+","++", "?+", "{", \
1392 "*", "*?", "+", "+?", "?", "??", "{", "{", "{", \
1393 "*+","++", "?+", "{", \
1394 "*", "*?", "+", "+?", "?", "??", "{", "{", \
1395 "class", "nclass", "xclass", "Ref", "Recurse", "Callout", \
1396 "Alt", "Ket", "KetRmax", "KetRmin", "Assert", "Assert not", \
1397 "AssertB", "AssertB not", "Reverse", \
1398 "Once", "Bra", "CBra", "Cond", "SBra", "SCBra", "SCond", \
1399 "Cond ref", "Cond rec", "Cond def", "Brazero", "Braminzero", \
1400 "*PRUNE", "*SKIP", "*THEN", "*COMMIT", "*FAIL", "*ACCEPT", \
1401 "Skip zero"
1402
1403
1404 /* This macro defines the length of fixed length operations in the compiled
1405 regex. The lengths are used when searching for specific things, and also in the
1406 debugging printing of a compiled regex. We use a macro so that it can be
1407 defined close to the definitions of the opcodes themselves.
1408
1409 As things have been extended, some of these are no longer fixed lenths, but are
1410 minima instead. For example, the length of a single-character repeat may vary
1411 in UTF-8 mode. The code that uses this table must know about such things. */
1412
1413 #define OP_LENGTHS \
1414 1, /* End */ \
1415 1, 1, 1, 1, 1, /* \A, \G, \K, \B, \b */ \
1416 1, 1, 1, 1, 1, 1, /* \D, \d, \S, \s, \W, \w */ \
1417 1, 1, 1, /* Any, AllAny, Anybyte */ \
1418 3, 3, 1, /* NOTPROP, PROP, EXTUNI */ \
1419 1, 1, 1, 1, 1, /* \R, \H, \h, \V, \v */ \
1420 1, 1, 2, 1, 1, /* \Z, \z, Opt, ^, $ */ \
1421 2, /* Char - the minimum length */ \
1422 2, /* Charnc - the minimum length */ \
1423 2, /* not */ \
1424 /* Positive single-char repeats ** These are */ \
1425 2, 2, 2, 2, 2, 2, /* *, *?, +, +?, ?, ?? ** minima in */ \
1426 4, 4, 4, /* upto, minupto, exact ** UTF-8 mode */ \
1427 2, 2, 2, 4, /* *+, ++, ?+, upto+ */ \
1428 /* Negative single-char repeats - only for chars < 256 */ \
1429 2, 2, 2, 2, 2, 2, /* NOT *, *?, +, +?, ?, ?? */ \
1430 4, 4, 4, /* NOT upto, minupto, exact */ \
1431 2, 2, 2, 4, /* Possessive *, +, ?, upto */ \
1432 /* Positive type repeats */ \
1433 2, 2, 2, 2, 2, 2, /* Type *, *?, +, +?, ?, ?? */ \
1434 4, 4, 4, /* Type upto, minupto, exact */ \
1435 2, 2, 2, 4, /* Possessive *+, ++, ?+, upto+ */ \
1436 /* Character class & ref repeats */ \
1437 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ \
1438 5, 5, /* CRRANGE, CRMINRANGE */ \
1439 33, /* CLASS */ \
1440 33, /* NCLASS */ \
1441 0, /* XCLASS - variable length */ \
1442 3, /* REF */ \
1443 1+LINK_SIZE, /* RECURSE */ \
1444 2+2*LINK_SIZE, /* CALLOUT */ \
1445 1+LINK_SIZE, /* Alt */ \
1446 1+LINK_SIZE, /* Ket */ \
1447 1+LINK_SIZE, /* KetRmax */ \
1448 1+LINK_SIZE, /* KetRmin */ \
1449 1+LINK_SIZE, /* Assert */ \
1450 1+LINK_SIZE, /* Assert not */ \
1451 1+LINK_SIZE, /* Assert behind */ \
1452 1+LINK_SIZE, /* Assert behind not */ \
1453 1+LINK_SIZE, /* Reverse */ \
1454 1+LINK_SIZE, /* ONCE */ \
1455 1+LINK_SIZE, /* BRA */ \
1456 3+LINK_SIZE, /* CBRA */ \
1457 1+LINK_SIZE, /* COND */ \
1458 1+LINK_SIZE, /* SBRA */ \
1459 3+LINK_SIZE, /* SCBRA */ \
1460 1+LINK_SIZE, /* SCOND */ \
1461 3, /* CREF */ \
1462 3, /* RREF */ \
1463 1, /* DEF */ \
1464 1, 1, /* BRAZERO, BRAMINZERO */ \
1465 1, 1, 1, 1, /* PRUNE, SKIP, THEN, COMMIT, */ \
1466 1, 1, 1 /* FAIL, ACCEPT, SKIPZERO */
1467
1468
1469 /* A magic value for OP_RREF to indicate the "any recursion" condition. */
1470
1471 #define RREF_ANY 0xffff
1472
1473 /* Error code numbers. They are given names so that they can more easily be
1474 tracked. */
1475
1476 enum { ERR0, ERR1, ERR2, ERR3, ERR4, ERR5, ERR6, ERR7, ERR8, ERR9,
1477 ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,
1478 ERR20, ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR28, ERR29,
1479 ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,
1480 ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,
1481 ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,
1482 ERR60, ERR61, ERR62, ERR63, ERR64 };
1483
1484 /* The real format of the start of the pcre block; the index of names and the
1485 code vector run on as long as necessary after the end. We store an explicit
1486 offset to the name table so that if a regex is compiled on one host, saved, and
1487 then run on another where the size of pointers is different, all might still
1488 be well. For the case of compiled-on-4 and run-on-8, we include an extra
1489 pointer that is always NULL. For future-proofing, a few dummy fields were
1490 originally included - even though you can never get this planning right - but
1491 there is only one left now.
1492
1493 NOTE NOTE NOTE:
1494 Because people can now save and re-use compiled patterns, any additions to this
1495 structure should be made at the end, and something earlier (e.g. a new
1496 flag in the options or one of the dummy fields) should indicate that the new
1497 fields are present. Currently PCRE always sets the dummy fields to zero.
1498 NOTE NOTE NOTE:
1499 */
1500
1501 typedef struct real_pcre {
1502 pcre_uint32 magic_number;
1503 pcre_uint32 size; /* Total that was malloced */
1504 pcre_uint32 options; /* Public options */
1505 pcre_uint16 flags; /* Private flags */
1506 pcre_uint16 dummy1; /* For future use */
1507 pcre_uint16 top_bracket;
1508 pcre_uint16 top_backref;
1509 pcre_uint16 first_byte;
1510 pcre_uint16 req_byte;
1511 pcre_uint16 name_table_offset; /* Offset to name table that follows */
1512 pcre_uint16 name_entry_size; /* Size of any name items */
1513 pcre_uint16 name_count; /* Number of name items */
1514 pcre_uint16 ref_count; /* Reference count */
1515
1516 const unsigned char *tables; /* Pointer to tables or NULL for std */
1517 const unsigned char *nullpad; /* NULL padding */
1518 } real_pcre;
1519
1520 /* The format of the block used to store data from pcre_study(). The same
1521 remark (see NOTE above) about extending this structure applies. */
1522
1523 typedef struct pcre_study_data {
1524 pcre_uint32 size; /* Total that was malloced */
1525 pcre_uint32 options;
1526 uschar start_bits[32];
1527 } pcre_study_data;
1528
1529 /* Structure for passing "static" information around between the functions
1530 doing the compiling, so that they are thread-safe. */
1531
1532 typedef struct compile_data {
1533 const uschar *lcc; /* Points to lower casing table */
1534 const uschar *fcc; /* Points to case-flipping table */
1535 const uschar *cbits; /* Points to character type table */
1536 const uschar *ctypes; /* Points to table of type maps */
1537 const uschar *start_workspace;/* The start of working space */
1538 const uschar *start_code; /* The start of the compiled code */
1539 const uschar *start_pattern; /* The start of the pattern */
1540 const uschar *end_pattern; /* The end of the pattern */
1541 uschar *hwm; /* High watermark of workspace */
1542 uschar *name_table; /* The name/number table */
1543 int names_found; /* Number of entries so far */
1544 int name_entry_size; /* Size of each entry */
1545 int bracount; /* Count of capturing parens as we compile */
1546 int final_bracount; /* Saved value after first pass */
1547 int top_backref; /* Maximum back reference */
1548 unsigned int backref_map; /* Bitmap of low back refs */
1549 int external_options; /* External (initial) options */
1550 int external_flags; /* External flag bits to be set */
1551 int req_varyopt; /* "After variable item" flag for reqbyte */
1552 BOOL had_accept; /* (*ACCEPT) encountered */
1553 int nltype; /* Newline type */
1554 int nllen; /* Newline string length */
1555 uschar nl[4]; /* Newline string when fixed length */
1556 } compile_data;
1557
1558 /* Structure for maintaining a chain of pointers to the currently incomplete
1559 branches, for testing for left recursion. */
1560
1561 typedef struct branch_chain {
1562 struct branch_chain *outer;
1563 uschar *current;
1564 } branch_chain;
1565
1566 /* Structure for items in a linked list that represents an explicit recursive
1567 call within the pattern. */
1568
1569 typedef struct recursion_info {
1570 struct recursion_info *prevrec; /* Previous recursion record (or NULL) */
1571 int group_num; /* Number of group that was called */
1572 const uschar *after_call; /* "Return value": points after the call in the expr */
1573 USPTR save_start; /* Old value of mstart */
1574 int *offset_save; /* Pointer to start of saved offsets */
1575 int saved_max; /* Number of saved offsets */
1576 } recursion_info;
1577
1578 /* Structure for building a chain of data for holding the values of the subject
1579 pointer at the start of each subpattern, so as to detect when an empty string
1580 has been matched by a subpattern - to break infinite loops. */
1581
1582 typedef struct eptrblock {
1583 struct eptrblock *epb_prev;
1584 USPTR epb_saved_eptr;
1585 } eptrblock;
1586
1587
1588 /* Structure for passing "static" information around between the functions
1589 doing traditional NFA matching, so that they are thread-safe. */
1590
1591 typedef struct match_data {
1592 unsigned long int match_call_count; /* As it says */
1593 unsigned long int match_limit; /* As it says */
1594 unsigned long int match_limit_recursion; /* As it says */
1595 int *offset_vector; /* Offset vector */
1596 int offset_end; /* One past the end */
1597 int offset_max; /* The maximum usable for return data */
1598 int nltype; /* Newline type */
1599 int nllen; /* Newline string length */
1600 uschar nl[4]; /* Newline string when fixed */
1601 const uschar *lcc; /* Points to lower casing table */
1602 const uschar *ctypes; /* Points to table of type maps */
1603 BOOL offset_overflow; /* Set if too many extractions */
1604 BOOL notbol; /* NOTBOL flag */
1605 BOOL noteol; /* NOTEOL flag */
1606 BOOL utf8; /* UTF8 flag */
1607 BOOL jscript_compat; /* JAVASCRIPT_COMPAT flag */
1608 BOOL endonly; /* Dollar not before final \n */
1609 BOOL notempty; /* Empty string match not wanted */
1610 BOOL partial; /* PARTIAL flag */
1611 BOOL hitend; /* Hit the end of the subject at some point */
1612 BOOL bsr_anycrlf; /* \R is just any CRLF, not full Unicode */
1613 const uschar *start_code; /* For use when recursing */
1614 USPTR start_subject; /* Start of the subject string */
1615 USPTR end_subject; /* End of the subject string */
1616 USPTR start_match_ptr; /* Start of matched string */
1617 USPTR end_match_ptr; /* Subject position at end match */
1618 int end_offset_top; /* Highwater mark at end of match */
1619 int capture_last; /* Most recent capture number */
1620 int start_offset; /* The start offset value */
1621 eptrblock *eptrchain; /* Chain of eptrblocks for tail recursions */
1622 int eptrn; /* Next free eptrblock */
1623 recursion_info *recursive; /* Linked list of recursion data */
1624 void *callout_data; /* To pass back to callouts */
1625 } match_data;
1626
1627 /* A similar structure is used for the same purpose by the DFA matching
1628 functions. */
1629
1630 typedef struct dfa_match_data {
1631 const uschar *start_code; /* Start of the compiled pattern */
1632 const uschar *start_subject; /* Start of the subject string */
1633 const uschar *end_subject; /* End of subject string */
1634 const uschar *tables; /* Character tables */
1635 int moptions; /* Match options */
1636 int poptions; /* Pattern options */
1637 int nltype; /* Newline type */
1638 int nllen; /* Newline string length */
1639 uschar nl[4]; /* Newline string when fixed */
1640 void *callout_data; /* To pass back to callouts */
1641 } dfa_match_data;
1642
1643 /* Bit definitions for entries in the pcre_ctypes table. */
1644
1645 #define ctype_space 0x01
1646 #define ctype_letter 0x02
1647 #define ctype_digit 0x04
1648 #define ctype_xdigit 0x08
1649 #define ctype_word 0x10 /* alphanumeric or '_' */
1650 #define ctype_meta 0x80 /* regexp meta char or zero (end pattern) */
1651
1652 /* Offsets for the bitmap tables in pcre_cbits. Each table contains a set
1653 of bits for a class map. Some classes are built by combining these tables. */
1654
1655 #define cbit_space 0 /* [:space:] or \s */
1656 #define cbit_xdigit 32 /* [:xdigit:] */
1657 #define cbit_digit 64 /* [:digit:] or \d */
1658 #define cbit_upper 96 /* [:upper:] */
1659 #define cbit_lower 128 /* [:lower:] */
1660 #define cbit_word 160 /* [:word:] or \w */
1661 #define cbit_graph 192 /* [:graph:] */
1662 #define cbit_print 224 /* [:print:] */
1663 #define cbit_punct 256 /* [:punct:] */
1664 #define cbit_cntrl 288 /* [:cntrl:] */
1665 #define cbit_length 320 /* Length of the cbits table */
1666
1667 /* Offsets of the various tables from the base tables pointer, and
1668 total length. */
1669
1670 #define lcc_offset 0
1671 #define fcc_offset 256
1672 #define cbits_offset 512
1673 #define ctypes_offset (cbits_offset + cbit_length)
1674 #define tables_length (ctypes_offset + 256)
1675
1676 /* Layout of the UCP type table that translates property names into types and
1677 codes. Each entry used to point directly to a name, but to reduce the number of
1678 relocations in shared libraries, it now has an offset into a single string
1679 instead. */
1680
1681 typedef struct {
1682 pcre_uint16 name_offset;
1683 pcre_uint16 type;
1684 pcre_uint16 value;
1685 } ucp_type_table;
1686
1687
1688 /* Internal shared data tables. These are tables that are used by more than one
1689 of the exported public functions. They have to be "external" in the C sense,
1690 but are not part of the PCRE public API. The data for these tables is in the
1691 pcre_tables.c module. */
1692
1693 extern const int _pcre_utf8_table1[];
1694 extern const int _pcre_utf8_table2[];
1695 extern const int _pcre_utf8_table3[];
1696 extern const uschar _pcre_utf8_table4[];
1697
1698 extern const int _pcre_utf8_table1_size;
1699
1700 extern const char _pcre_utt_names[];
1701 extern const ucp_type_table _pcre_utt[];
1702 extern const int _pcre_utt_size;
1703
1704 extern const uschar _pcre_default_tables[];
1705
1706 extern const uschar _pcre_OP_lengths[];
1707
1708
1709 /* Internal shared functions. These are functions that are used by more than
1710 one of the exported public functions. They have to be "external" in the C
1711 sense, but are not part of the PCRE public API. */
1712
1713 extern BOOL _pcre_is_newline(const uschar *, int, const uschar *,
1714 int *, BOOL);
1715 extern int _pcre_ord2utf8(int, uschar *);
1716 extern real_pcre *_pcre_try_flipped(const real_pcre *, real_pcre *,
1717 const pcre_study_data *, pcre_study_data *);
1718 extern int _pcre_valid_utf8(const uschar *, int);
1719 extern BOOL _pcre_was_newline(const uschar *, int, const uschar *,
1720 int *, BOOL);
1721 extern BOOL _pcre_xclass(int, const uschar *);
1722
1723
1724 /* Unicode character database (UCD) */
1725
1726 typedef struct {
1727 uschar script;
1728 uschar chartype;
1729 pcre_int32 other_case;
1730 } ucd_record;
1731
1732 extern const ucd_record _pcre_ucd_records[];
1733 extern const uschar _pcre_ucd_stage1[];
1734 extern const pcre_uint16 _pcre_ucd_stage2[];
1735 extern const int _pcre_ucp_gentype[];
1736
1737
1738 /* UCD access macros */
1739
1740 #define UCD_BLOCK_SIZE 128
1741 #define GET_UCD(ch) (_pcre_ucd_records + \
1742 _pcre_ucd_stage2[_pcre_ucd_stage1[(ch) / UCD_BLOCK_SIZE] * \
1743 UCD_BLOCK_SIZE + ch % UCD_BLOCK_SIZE])
1744
1745 #define UCD_CHARTYPE(ch) GET_UCD(ch)->chartype
1746 #define UCD_SCRIPT(ch) GET_UCD(ch)->script
1747 #define UCD_CATEGORY(ch) _pcre_ucp_gentype[UCD_CHARTYPE(ch)]
1748 #define UCD_OTHERCASE(ch) (ch + GET_UCD(ch)->other_case)
1749
1750 #endif
1751
1752 /* End of pcre_internal.h */
1753 |
This page was automatically generated by the
LXR engine.
Visit the LXR main site for more information. |