/[pcre]/code/branches/pcre16/pcre_internal.h
ViewVC logotype

Contents of /code/branches/pcre16/pcre_internal.h

Parent Directory Parent Directory | Revision Log Revision Log


Revision 795 - (show annotations)
Sat Dec 10 02:20:06 2011 UTC (7 years, 8 months ago) by zherczeg
File MIME type: text/plain
File size: 93611 byte(s)
extending the 16 bit API, mode check, and fixes
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) 1997-2011 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_" or "_pcre16_" depending on
44 the PRIV macro. */
45
46 #ifndef PCRE_INTERNAL_H
47 #define PCRE_INTERNAL_H
48
49 /* Define PCRE_DEBUG to get debugging output on stdout. */
50
51 #if 0
52 #define PCRE_DEBUG
53 #endif
54
55 /* PCRE is compiled as an 8 bit library if it is not requested otherwise. */
56 #ifndef COMPILE_PCRE16
57 #define COMPILE_PCRE8
58 #endif
59
60 /* If SUPPORT_UCP is defined, SUPPORT_UTF must also be defined. The
61 "configure" script ensures this, but not everybody uses "configure". */
62
63 #if defined SUPPORT_UCP && !(defined SUPPORT_UTF)
64 #define SUPPORT_UTF 1
65 #endif
66
67 /* We define SUPPORT_UTF if SUPPORT_UTF8 is enabled for compatibility
68 reasons with existing code. */
69
70 #if defined SUPPORT_UTF8 && !(defined SUPPORT_UTF)
71 #define SUPPORT_UTF 1
72 #endif
73
74 /* Fixme: SUPPORT_UTF8 should be eventually disappear from the code.
75 Until then we define it if SUPPORT_UTF is defined. */
76
77 #if defined SUPPORT_UTF && !(defined SUPPORT_UTF8)
78 #define SUPPORT_UTF8 1
79 #endif
80
81 /* We do not support both EBCDIC and UTF-8/16 at the same time. The "configure"
82 script prevents both being selected, but not everybody uses "configure". */
83
84 #if defined EBCDIC && defined SUPPORT_UTF
85 #error The use of both EBCDIC and SUPPORT_UTF8/16 is not supported.
86 #endif
87
88 /* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef
89 inline, and there are *still* stupid compilers about that don't like indented
90 pre-processor statements, or at least there were when I first wrote this. After
91 all, it had only been about 10 years then...
92
93 It turns out that the Mac Debugging.h header also defines the macro DPRINTF, so
94 be absolutely sure we get our version. */
95
96 #undef DPRINTF
97 #ifdef PCRE_DEBUG
98 #define DPRINTF(p) printf p
99 #else
100 #define DPRINTF(p) /* Nothing */
101 #endif
102
103
104 /* Standard C headers plus the external interface definition. The only time
105 setjmp and stdarg are used is when NO_RECURSE is set. */
106
107 #include <ctype.h>
108 #include <limits.h>
109 #include <stddef.h>
110 #include <stdio.h>
111 #include <stdlib.h>
112 #include <string.h>
113
114 /* When compiling a DLL for Windows, the exported symbols have to be declared
115 using some MS magic. I found some useful information on this web page:
116 http://msdn2.microsoft.com/en-us/library/y4h7bcy6(VS.80).aspx. According to the
117 information there, using __declspec(dllexport) without "extern" we have a
118 definition; with "extern" we have a declaration. The settings here override the
119 setting in pcre.h (which is included below); it defines only PCRE_EXP_DECL,
120 which is all that is needed for applications (they just import the symbols). We
121 use:
122
123 PCRE_EXP_DECL for declarations
124 PCRE_EXP_DEFN for definitions of exported functions
125 PCRE_EXP_DATA_DEFN for definitions of exported variables
126
127 The reason for the two DEFN macros is that in non-Windows environments, one
128 does not want to have "extern" before variable definitions because it leads to
129 compiler warnings. So we distinguish between functions and variables. In
130 Windows, the two should always be the same.
131
132 The reason for wrapping this in #ifndef PCRE_EXP_DECL is so that pcretest,
133 which is an application, but needs to import this file in order to "peek" at
134 internals, can #include pcre.h first to get an application's-eye view.
135
136 In principle, people compiling for non-Windows, non-Unix-like (i.e. uncommon,
137 special-purpose environments) might want to stick other stuff in front of
138 exported symbols. That's why, in the non-Windows case, we set PCRE_EXP_DEFN and
139 PCRE_EXP_DATA_DEFN only if they are not already set. */
140
141 #ifndef PCRE_EXP_DECL
142 # ifdef _WIN32
143 # ifndef PCRE_STATIC
144 # define PCRE_EXP_DECL extern __declspec(dllexport)
145 # define PCRE_EXP_DEFN __declspec(dllexport)
146 # define PCRE_EXP_DATA_DEFN __declspec(dllexport)
147 # else
148 # define PCRE_EXP_DECL extern
149 # define PCRE_EXP_DEFN
150 # define PCRE_EXP_DATA_DEFN
151 # endif
152 # else
153 # ifdef __cplusplus
154 # define PCRE_EXP_DECL extern "C"
155 # else
156 # define PCRE_EXP_DECL extern
157 # endif
158 # ifndef PCRE_EXP_DEFN
159 # define PCRE_EXP_DEFN PCRE_EXP_DECL
160 # endif
161 # ifndef PCRE_EXP_DATA_DEFN
162 # define PCRE_EXP_DATA_DEFN
163 # endif
164 # endif
165 #endif
166
167 /* When compiling with the MSVC compiler, it is sometimes necessary to include
168 a "calling convention" before exported function names. (This is secondhand
169 information; I know nothing about MSVC myself). For example, something like
170
171 void __cdecl function(....)
172
173 might be needed. In order so make this easy, all the exported functions have
174 PCRE_CALL_CONVENTION just before their names. It is rarely needed; if not
175 set, we ensure here that it has no effect. */
176
177 #ifndef PCRE_CALL_CONVENTION
178 #define PCRE_CALL_CONVENTION
179 #endif
180
181 /* We need to have types that specify unsigned 8, 16 and 32-bit integers. We
182 cannot determine these outside the compilation (e.g. by running a program as
183 part of "configure") because PCRE is often cross-compiled for use on other
184 systems. Instead we make use of the maximum sizes that are available at
185 preprocessor time in standard C environments. */
186
187 typedef unsigned char pcre_uint8;
188
189 #if USHRT_MAX == 65535
190 typedef unsigned short pcre_uint16;
191 typedef short pcre_int16;
192 #elif UINT_MAX == 65535
193 typedef unsigned int pcre_uint16;
194 typedef int pcre_int16;
195 #else
196 #error Cannot determine a type for 16-bit unsigned integers
197 #endif
198
199 #if UINT_MAX == 4294967295
200 typedef unsigned int pcre_uint32;
201 typedef int pcre_int32;
202 #elif ULONG_MAX == 4294967295
203 typedef unsigned long int pcre_uint32;
204 typedef long int pcre_int32;
205 #else
206 #error Cannot determine a type for 32-bit unsigned integers
207 #endif
208
209 /* When checking for integer overflow in pcre_compile(), we need to handle
210 large integers. If a 64-bit integer type is available, we can use that.
211 Otherwise we have to cast to double, which of course requires floating point
212 arithmetic. Handle this by defining a macro for the appropriate type. If
213 stdint.h is available, include it; it may define INT64_MAX. Systems that do not
214 have stdint.h (e.g. Solaris) may have inttypes.h. The macro int64_t may be set
215 by "configure". */
216
217 #if HAVE_STDINT_H
218 #include <stdint.h>
219 #elif HAVE_INTTYPES_H
220 #include <inttypes.h>
221 #endif
222
223 #if defined INT64_MAX || defined int64_t
224 #define INT64_OR_DOUBLE int64_t
225 #else
226 #define INT64_OR_DOUBLE double
227 #endif
228
229 /* All character handling must be done as unsigned characters. Otherwise there
230 are problems with top-bit-set characters and functions such as isspace().
231 However, we leave the interface to the outside world as char * or short *,
232 because that should make things easier for callers. This character type is
233 called pcre_uchar.
234
235 The IN_UCHARS macro multiply its argument with the byte size of the current
236 pcre_uchar type. Useful for memcpy and such operations, whose require the
237 byte size of their input/output buffers.
238
239 The MAX_255 macro checks whether its pcre_uchar input is less than 256.
240
241 The TABLE_GET macro is designed for accessing elements of tables whose contain
242 exactly 256 items. When the character is able to contain more than 256
243 items, some check is needed before accessing these tables.
244 */
245
246 #ifdef COMPILE_PCRE8
247
248 typedef unsigned char pcre_uchar;
249 #define IN_UCHARS(x) (x)
250 #define MAX_255(c) 1
251 #define TABLE_GET(c, table, default) ((table)[c])
252
253 #else
254
255 #ifdef COMPILE_PCRE16
256 #if USHRT_MAX != 65535
257 /* This is a warning message. Change PCRE_SCHAR16 to a 16 bit data type in
258 pcre.h(.in) and disable (comment out) this message. */
259 #error Warning: PCRE_SCHAR16 is not a 16 bit data type.
260 #endif
261
262 typedef pcre_uint16 pcre_uchar;
263 #define IN_UCHARS(x) ((x) << 1)
264 #define MAX_255(c) ((c) <= 255u)
265 #define TABLE_GET(c, table, default) (MAX_255(c)? ((table)[c]):(default))
266
267 #else
268 #error Unsupported compiling mode
269 #endif /* COMPILE_PCRE16 */
270
271 #endif /* COMPILE_PCRE8 */
272
273 /* This is an unsigned int value that no character can ever have. UTF-8
274 characters only go up to 0x7fffffff (though Unicode doesn't go beyond
275 0x0010ffff). */
276
277 #define NOTACHAR 0xffffffff
278
279 /* PCRE is able to support several different kinds of newline (CR, LF, CRLF,
280 "any" and "anycrlf" at present). The following macros are used to package up
281 testing for newlines. NLBLOCK, PSSTART, and PSEND are defined in the various
282 modules to indicate in which datablock the parameters exist, and what the
283 start/end of string field names are. */
284
285 #define NLTYPE_FIXED 0 /* Newline is a fixed length string */
286 #define NLTYPE_ANY 1 /* Newline is any Unicode line ending */
287 #define NLTYPE_ANYCRLF 2 /* Newline is CR, LF, or CRLF */
288
289 /* This macro checks for a newline at the given position */
290
291 #define IS_NEWLINE(p) \
292 ((NLBLOCK->nltype != NLTYPE_FIXED)? \
293 ((p) < NLBLOCK->PSEND && \
294 PRIV(is_newline)((p), NLBLOCK->nltype, NLBLOCK->PSEND, \
295 &(NLBLOCK->nllen), utf)) \
296 : \
297 ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \
298 (p)[0] == NLBLOCK->nl[0] && \
299 (NLBLOCK->nllen == 1 || (p)[1] == NLBLOCK->nl[1]) \
300 ) \
301 )
302
303 /* This macro checks for a newline immediately preceding the given position */
304
305 #define WAS_NEWLINE(p) \
306 ((NLBLOCK->nltype != NLTYPE_FIXED)? \
307 ((p) > NLBLOCK->PSSTART && \
308 PRIV(was_newline)((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \
309 &(NLBLOCK->nllen), utf)) \
310 : \
311 ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \
312 (p)[-NLBLOCK->nllen] == NLBLOCK->nl[0] && \
313 (NLBLOCK->nllen == 1 || (p)[-NLBLOCK->nllen+1] == NLBLOCK->nl[1]) \
314 ) \
315 )
316
317 /* When PCRE is compiled as a C++ library, the subject pointer can be replaced
318 with a custom type. This makes it possible, for example, to allow pcre_exec()
319 to process subject strings that are discontinuous by using a smart pointer
320 class. It must always be possible to inspect all of the subject string in
321 pcre_exec() because of the way it backtracks. Two macros are required in the
322 normal case, for sign-unspecified and unsigned char pointers. The former is
323 used for the external interface and appears in pcre.h, which is why its name
324 must begin with PCRE_. */
325
326 #ifdef CUSTOM_SUBJECT_PTR
327 #define PCRE_PUCHAR CUSTOM_SUBJECT_PTR
328 #else
329 #define PCRE_PUCHAR const pcre_uchar *
330 #endif
331
332 /* Include the public PCRE header and the definitions of UCP character property
333 values. */
334
335 #include "pcre.h"
336 #include "ucp.h"
337
338 /* When compiling for use with the Virtual Pascal compiler, these functions
339 need to have their names changed. PCRE must be compiled with the -DVPCOMPAT
340 option on the command line. */
341
342 #ifdef VPCOMPAT
343 #define strlen(s) _strlen(s)
344 #define strncmp(s1,s2,m) _strncmp(s1,s2,m)
345 #define memcmp(s,c,n) _memcmp(s,c,n)
346 #define memcpy(d,s,n) _memcpy(d,s,n)
347 #define memmove(d,s,n) _memmove(d,s,n)
348 #define memset(s,c,n) _memset(s,c,n)
349 #else /* VPCOMPAT */
350
351 /* To cope with SunOS4 and other systems that lack memmove() but have bcopy(),
352 define a macro for memmove() if HAVE_MEMMOVE is false, provided that HAVE_BCOPY
353 is set. Otherwise, include an emulating function for those systems that have
354 neither (there some non-Unix environments where this is the case). */
355
356 #ifndef HAVE_MEMMOVE
357 #undef memmove /* some systems may have a macro */
358 #ifdef HAVE_BCOPY
359 #define memmove(a, b, c) bcopy(b, a, c)
360 #else /* HAVE_BCOPY */
361 static void *
362 pcre_memmove(void *d, const void *s, size_t n)
363 {
364 size_t i;
365 unsigned char *dest = (unsigned char *)d;
366 const unsigned char *src = (const unsigned char *)s;
367 if (dest > src)
368 {
369 dest += n;
370 src += n;
371 for (i = 0; i < n; ++i) *(--dest) = *(--src);
372 return (void *)dest;
373 }
374 else
375 {
376 for (i = 0; i < n; ++i) *dest++ = *src++;
377 return (void *)(dest - n);
378 }
379 }
380 #define memmove(a, b, c) pcre_memmove(a, b, c)
381 #endif /* not HAVE_BCOPY */
382 #endif /* not HAVE_MEMMOVE */
383 #endif /* not VPCOMPAT */
384
385
386 /* PCRE keeps offsets in its compiled code as 2-byte quantities (always stored
387 in big-endian order) by default. These are used, for example, to link from the
388 start of a subpattern to its alternatives and its end. The use of 2 bytes per
389 offset limits the size of the compiled regex to around 64K, which is big enough
390 for almost everybody. However, I received a request for an even bigger limit.
391 For this reason, and also to make the code easier to maintain, the storing and
392 loading of offsets from the byte string is now handled by the macros that are
393 defined here.
394
395 The macros are controlled by the value of LINK_SIZE. This defaults to 2 in
396 the config.h file, but can be overridden by using -D on the command line. This
397 is automated on Unix systems via the "configure" command. */
398
399 #ifdef COMPILE_PCRE8
400
401 #if LINK_SIZE == 2
402
403 #define PUT(a,n,d) \
404 (a[n] = (d) >> 8), \
405 (a[(n)+1] = (d) & 255)
406
407 #define GET(a,n) \
408 (((a)[n] << 8) | (a)[(n)+1])
409
410 #define MAX_PATTERN_SIZE (1 << 16)
411
412
413 #elif LINK_SIZE == 3
414
415 #define PUT(a,n,d) \
416 (a[n] = (d) >> 16), \
417 (a[(n)+1] = (d) >> 8), \
418 (a[(n)+2] = (d) & 255)
419
420 #define GET(a,n) \
421 (((a)[n] << 16) | ((a)[(n)+1] << 8) | (a)[(n)+2])
422
423 #define MAX_PATTERN_SIZE (1 << 24)
424
425
426 #elif LINK_SIZE == 4
427
428 #define PUT(a,n,d) \
429 (a[n] = (d) >> 24), \
430 (a[(n)+1] = (d) >> 16), \
431 (a[(n)+2] = (d) >> 8), \
432 (a[(n)+3] = (d) & 255)
433
434 #define GET(a,n) \
435 (((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3])
436
437 /* Keep it positive */
438 #define MAX_PATTERN_SIZE (1 << 30)
439
440 #else
441 #error LINK_SIZE must be either 2, 3, or 4
442 #endif
443
444 #else /* COMPILE_PCRE8 */
445
446 #ifdef COMPILE_PCRE16
447
448 #if LINK_SIZE == 2
449
450 #undef LINK_SIZE
451 #define LINK_SIZE 1
452
453 #define PUT(a,n,d) \
454 (a[n] = (d))
455
456 #define GET(a,n) \
457 (a[n])
458
459 #define MAX_PATTERN_SIZE (1 << 16)
460
461 #elif LINK_SIZE == 3 || LINK_SIZE == 4
462
463 #undef LINK_SIZE
464 #define LINK_SIZE 2
465
466 #define PUT(a,n,d) \
467 (a[n] = (d) >> 16), \
468 (a[(n)+1] = (d) & 65536)
469
470 #define GET(a,n) \
471 (((a)[n] << 16) | (a)[(n)+1])
472
473 /* Keep it positive */
474 #define MAX_PATTERN_SIZE (1 << 30)
475
476 #else
477 #error LINK_SIZE must be either 2, 3, or 4
478 #endif
479
480 #else
481 #error Unsupported compiling mode
482 #endif /* COMPILE_PCRE16 */
483
484 #endif /* COMPILE_PCRE8 */
485
486 /* Convenience macro defined in terms of the others */
487
488 #define PUTINC(a,n,d) PUT(a,n,d), a += LINK_SIZE
489
490
491 /* PCRE uses some other 2-byte quantities that do not change when the size of
492 offsets changes. There are used for repeat counts and for other things such as
493 capturing parenthesis numbers in back references. */
494
495 #ifdef COMPILE_PCRE8
496
497 #define IMM2_SIZE 2
498
499 #define PUT2(a,n,d) \
500 a[n] = (d) >> 8; \
501 a[(n)+1] = (d) & 255
502
503 #define GET2(a,n) \
504 (((a)[n] << 8) | (a)[(n)+1])
505
506 #else /* COMPILE_PCRE8 */
507
508 #ifdef COMPILE_PCRE16
509
510 #define IMM2_SIZE 1
511
512 #define PUT2(a,n,d) \
513 a[n] = d
514
515 #define GET2(a,n) \
516 a[n]
517
518 #else
519 #error Unsupported compiling mode
520 #endif /* COMPILE_PCRE16 */
521
522 #endif /* COMPILE_PCRE8 */
523
524 #define PUT2INC(a,n,d) PUT2(a,n,d), a += IMM2_SIZE
525
526 /* When UTF encoding is being used, a character is no longer just a single
527 character. The macros for character handling generate simple sequences when
528 used in character-mode, and more complicated ones for UTF characters.
529 GETCHARLENTEST and other macros are not used when UTF is not supported,
530 so they are not defined. To make sure they can never even appear when
531 UTF support is omitted, we don't even define them. */
532
533 #ifndef SUPPORT_UTF
534
535 /* #define HAS_EXTRALEN(c) */
536 /* #define GET_EXTRALEN(c) */
537 /* #define NOT_FIRSTCHAR(c) */
538 #define GETCHAR(c, eptr) c = *eptr;
539 #define GETCHARTEST(c, eptr) c = *eptr;
540 #define GETCHARINC(c, eptr) c = *eptr++;
541 #define GETCHARINCTEST(c, eptr) c = *eptr++;
542 #define GETCHARLEN(c, eptr, len) c = *eptr;
543 /* #define GETCHARLENTEST(c, eptr, len) */
544 /* #define BACKCHAR(eptr) */
545 /* #define FORWARDCHAR(eptr) */
546 /* #define ACROSSCHAR(condition, eptr, action) */
547
548 #else /* SUPPORT_UTF */
549
550 #ifdef COMPILE_PCRE8
551
552 /* These macros were originally written in the form of loops that used data
553 from the tables whose names start with PRIV(utf8_table). They were rewritten by
554 a user so as not to use loops, because in some environments this gives a
555 significant performance advantage, and it seems never to do any harm. */
556
557 /* Tests whether the code point needs extra characters to decode. */
558
559 #define HAS_EXTRALEN(c) ((c) >= 0xc0)
560
561 /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
562 Otherwise it has an undefined behaviour. */
563
564 #define GET_EXTRALEN(c) (PRIV(utf8_table4)[(c) & 0x3f])
565
566 /* Returns TRUE, if the given character is not the first character
567 of a UTF sequence. */
568
569 #define NOT_FIRSTCHAR(c) (((c) & 0xc0) == 0x80)
570
571 /* Base macro to pick up the remaining bytes of a UTF-8 character, not
572 advancing the pointer. */
573
574 #define GETUTF8(c, eptr) \
575 { \
576 if ((c & 0x20) == 0) \
577 c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
578 else if ((c & 0x10) == 0) \
579 c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
580 else if ((c & 0x08) == 0) \
581 c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
582 ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
583 else if ((c & 0x04) == 0) \
584 c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
585 ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
586 (eptr[4] & 0x3f); \
587 else \
588 c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
589 ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
590 ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
591 }
592
593 /* Get the next UTF-8 character, not advancing the pointer. This is called when
594 we know we are in UTF-8 mode. */
595
596 #define GETCHAR(c, eptr) \
597 c = *eptr; \
598 if (c >= 0xc0) GETUTF8(c, eptr);
599
600 /* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the
601 pointer. */
602
603 #define GETCHARTEST(c, eptr) \
604 c = *eptr; \
605 if (utf && c >= 0xc0) GETUTF8(c, eptr);
606
607 /* Base macro to pick up the remaining bytes of a UTF-8 character, advancing
608 the pointer. */
609
610 #define GETUTF8INC(c, eptr) \
611 { \
612 if ((c & 0x20) == 0) \
613 c = ((c & 0x1f) << 6) | (*eptr++ & 0x3f); \
614 else if ((c & 0x10) == 0) \
615 { \
616 c = ((c & 0x0f) << 12) | ((*eptr & 0x3f) << 6) | (eptr[1] & 0x3f); \
617 eptr += 2; \
618 } \
619 else if ((c & 0x08) == 0) \
620 { \
621 c = ((c & 0x07) << 18) | ((*eptr & 0x3f) << 12) | \
622 ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
623 eptr += 3; \
624 } \
625 else if ((c & 0x04) == 0) \
626 { \
627 c = ((c & 0x03) << 24) | ((*eptr & 0x3f) << 18) | \
628 ((eptr[1] & 0x3f) << 12) | ((eptr[2] & 0x3f) << 6) | \
629 (eptr[3] & 0x3f); \
630 eptr += 4; \
631 } \
632 else \
633 { \
634 c = ((c & 0x01) << 30) | ((*eptr & 0x3f) << 24) | \
635 ((eptr[1] & 0x3f) << 18) | ((eptr[2] & 0x3f) << 12) | \
636 ((eptr[3] & 0x3f) << 6) | (eptr[4] & 0x3f); \
637 eptr += 5; \
638 } \
639 }
640
641 /* Get the next UTF-8 character, advancing the pointer. This is called when we
642 know we are in UTF-8 mode. */
643
644 #define GETCHARINC(c, eptr) \
645 c = *eptr++; \
646 if (c >= 0xc0) GETUTF8INC(c, eptr);
647
648 /* Get the next character, testing for UTF-8 mode, and advancing the pointer.
649 This is called when we don't know if we are in UTF-8 mode. */
650
651 #define GETCHARINCTEST(c, eptr) \
652 c = *eptr++; \
653 if (utf && c >= 0xc0) GETUTF8INC(c, eptr);
654
655 /* Base macro to pick up the remaining bytes of a UTF-8 character, not
656 advancing the pointer, incrementing the length. */
657
658 #define GETUTF8LEN(c, eptr, len) \
659 { \
660 if ((c & 0x20) == 0) \
661 { \
662 c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
663 len++; \
664 } \
665 else if ((c & 0x10) == 0) \
666 { \
667 c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
668 len += 2; \
669 } \
670 else if ((c & 0x08) == 0) \
671 {\
672 c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
673 ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
674 len += 3; \
675 } \
676 else if ((c & 0x04) == 0) \
677 { \
678 c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
679 ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
680 (eptr[4] & 0x3f); \
681 len += 4; \
682 } \
683 else \
684 {\
685 c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
686 ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
687 ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
688 len += 5; \
689 } \
690 }
691
692 /* Get the next UTF-8 character, not advancing the pointer, incrementing length
693 if there are extra bytes. This is called when we know we are in UTF-8 mode. */
694
695 #define GETCHARLEN(c, eptr, len) \
696 c = *eptr; \
697 if (c >= 0xc0) GETUTF8LEN(c, eptr, len);
698
699 /* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the
700 pointer, incrementing length if there are extra bytes. This is called when we
701 do not know if we are in UTF-8 mode. */
702
703 #define GETCHARLENTEST(c, eptr, len) \
704 c = *eptr; \
705 if (utf && c >= 0xc0) GETUTF8LEN(c, eptr, len);
706
707 /* If the pointer is not at the start of a character, move it back until
708 it is. This is called only in UTF-8 mode - we don't put a test within the macro
709 because almost all calls are already within a block of UTF-8 only code. */
710
711 #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--
712
713 /* Same as above, just in the other direction. */
714 #define FORWARDCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr++
715
716 /* Same as above, but it allows a fully customizable form. */
717 #define ACROSSCHAR(condition, eptr, action) \
718 while((condition) && ((eptr) & 0xc0) == 0x80) action
719
720 #else /* COMPILE_PCRE8 */
721
722 #ifdef COMPILE_PCRE16
723
724 /* Tests whether the code point needs extra characters to decode. */
725
726 #define HAS_EXTRALEN(c) (((c) & 0xfc00) == 0xd800)
727
728 /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
729 Otherwise it has an undefined behaviour. */
730
731 #define GET_EXTRALEN(c) 1
732
733 /* Returns TRUE, if the given character is not the first character
734 of a UTF sequence. */
735
736 #define NOT_FIRSTCHAR(c) (((c) & 0xfc00) == 0xdc00)
737
738 /* Base macro to pick up the low surrogate of a UTF-16 character, not
739 advancing the pointer. */
740
741 #define GETUTF16(c, eptr) \
742 { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; }
743
744 /* Get the next UTF-16 character, not advancing the pointer. This is called when
745 we know we are in UTF-16 mode. */
746
747 #define GETCHAR(c, eptr) \
748 c = *eptr; \
749 if ((c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
750
751 /* Get the next UTF-16 character, testing for UTF-16 mode, and not advancing the
752 pointer. */
753
754 #define GETCHARTEST(c, eptr) \
755 c = *eptr; \
756 if (utf && (c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
757
758 /* Base macro to pick up the low surrogate of a UTF-16 character, advancing
759 the pointer. */
760
761 #define GETUTF16INC(c, eptr) \
762 { c = (((c & 0x3ff) << 10) | (*eptr++ & 0x3ff)) + 0x10000; }
763
764 /* Get the next UTF-16 character, advancing the pointer. This is called when we
765 know we are in UTF-16 mode. */
766
767 #define GETCHARINC(c, eptr) \
768 c = *eptr++; \
769 if ((c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
770
771 /* Get the next character, testing for UTF-16 mode, and advancing the pointer.
772 This is called when we don't know if we are in UTF-16 mode. */
773
774 #define GETCHARINCTEST(c, eptr) \
775 c = *eptr++; \
776 if (utf && (c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
777
778 /* Base macro to pick up the low surrogate of a UTF-16 character, not
779 advancing the pointer, incrementing the length. */
780
781 #define GETUTF16LEN(c, eptr, len) \
782 { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; len++; }
783
784 /* Get the next UTF-16 character, not advancing the pointer, incrementing
785 length if there is a low surrogate. This is called when we know we are in
786 UTF-16 mode. */
787
788 #define GETCHARLEN(c, eptr, len) \
789 c = *eptr; \
790 if ((c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
791
792 /* Get the next UTF-816character, testing for UTF-16 mode, not advancing the
793 pointer, incrementing length if there is a low surrogate. This is called when
794 we do not know if we are in UTF-16 mode. */
795
796 #define GETCHARLENTEST(c, eptr, len) \
797 c = *eptr; \
798 if (utf && (c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
799
800 /* If the pointer is not at the start of a character, move it back until
801 it is. This is called only in UTF-16 mode - we don't put a test within the
802 macro because almost all calls are already within a block of UTF-16 only
803 code. */
804
805 #define BACKCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr--
806
807 /* Same as above, just in the other direction. */
808 #define FORWARDCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr++
809
810 /* Same as above, but it allows a fully customizable form. */
811 #define ACROSSCHAR(condition, eptr, action) \
812 if ((condition) && ((eptr) & 0xfc00) == 0xdc00) action
813
814 #endif
815
816 #endif /* COMPILE_PCRE8 */
817
818 #endif /* SUPPORT_UTF */
819
820
821 /* In case there is no definition of offsetof() provided - though any proper
822 Standard C system should have one. */
823
824 #ifndef offsetof
825 #define offsetof(p_type,field) ((size_t)&(((p_type *)0)->field))
826 #endif
827
828
829 /* Private flags containing information about the compiled regex. They used to
830 live at the top end of the options word, but that got almost full, so now they
831 are in a 16-bit flags word. From release 8.00, PCRE_NOPARTIAL is unused, as
832 the restrictions on partial matching have been lifted. It remains for backwards
833 compatibility. */
834
835 #ifdef COMPILE_PCRE8
836 #define PCRE_MODE 0x0001 /* compiled in 8 bit mode */
837 #endif
838 #ifdef COMPILE_PCRE16
839 #define PCRE_MODE 0x0002 /* compiled in 16 bit mode */
840 #endif
841 #define PCRE_FIRSTSET 0x0010 /* first_char is set */
842 #define PCRE_FCH_CASELESS 0x0020 /* caseless first char */
843 #define PCRE_REQCHSET 0x0040 /* req_byte is set */
844 #define PCRE_RCH_CASELESS 0x0080 /* caseless requested char */
845 #define PCRE_STARTLINE 0x0100 /* start after \n for multiline */
846 #define PCRE_NOPARTIAL 0x0200 /* can't use partial with this regex */
847 #define PCRE_JCHANGED 0x0400 /* j option used in regex */
848 #define PCRE_HASCRORLF 0x0800 /* explicit \r or \n in pattern */
849 #define PCRE_HASTHEN 0x1000 /* pattern contains (*THEN) */
850
851 /* Flags for the "extra" block produced by pcre_study(). */
852
853 #define PCRE_STUDY_MAPPED 0x0001 /* a map of starting chars exists */
854 #define PCRE_STUDY_MINLEN 0x0002 /* a minimum length field exists */
855
856 /* Masks for identifying the public options that are permitted at compile
857 time, run time, or study time, respectively. */
858
859 #define PCRE_NEWLINE_BITS (PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_ANY| \
860 PCRE_NEWLINE_ANYCRLF)
861
862 #define PUBLIC_COMPILE_OPTIONS \
863 (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \
864 PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \
865 PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \
866 PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
867 PCRE_JAVASCRIPT_COMPAT|PCRE_UCP|PCRE_NO_START_OPTIMIZE)
868
869 #define PUBLIC_EXEC_OPTIONS \
870 (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
871 PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_NEWLINE_BITS| \
872 PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE|PCRE_NO_START_OPTIMIZE)
873
874 #define PUBLIC_DFA_EXEC_OPTIONS \
875 (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
876 PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_DFA_SHORTEST| \
877 PCRE_DFA_RESTART|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
878 PCRE_NO_START_OPTIMIZE)
879
880 #define PUBLIC_STUDY_OPTIONS \
881 PCRE_STUDY_JIT_COMPILE
882
883 /* Magic number to provide a small check against being handed junk. Also used
884 to detect whether a pattern was compiled on a host of different endianness. */
885
886 #define MAGIC_NUMBER 0x50435245UL /* 'PCRE' */
887
888 /* Negative values for the firstchar and reqchar variables */
889
890 #define REQ_UNSET (-2)
891 #define REQ_NONE (-1)
892
893 /* The maximum remaining length of subject we are prepared to search for a
894 req_byte match. */
895
896 #define REQ_BYTE_MAX 1000
897
898 /* Miscellaneous definitions. The #ifndef is to pacify compiler warnings in
899 environments where these macros are defined elsewhere. Unfortunately, there
900 is no way to do the same for the typedef. */
901
902 typedef int BOOL;
903
904 #ifndef FALSE
905 #define FALSE 0
906 #define TRUE 1
907 #endif
908
909 /* If PCRE is to support UTF-8 on EBCDIC platforms, we cannot use normal
910 character constants like '*' because the compiler would emit their EBCDIC code,
911 which is different from their ASCII/UTF-8 code. Instead we define macros for
912 the characters so that they always use the ASCII/UTF-8 code when UTF-8 support
913 is enabled. When UTF-8 support is not enabled, the definitions use character
914 literals. Both character and string versions of each character are needed, and
915 there are some longer strings as well.
916
917 This means that, on EBCDIC platforms, the PCRE library can handle either
918 EBCDIC, or UTF-8, but not both. To support both in the same compiled library
919 would need different lookups depending on whether PCRE_UTF8 was set or not.
920 This would make it impossible to use characters in switch/case statements,
921 which would reduce performance. For a theoretical use (which nobody has asked
922 for) in a minority area (EBCDIC platforms), this is not sensible. Any
923 application that did need both could compile two versions of the library, using
924 macros to give the functions distinct names. */
925
926 #ifndef SUPPORT_UTF
927
928 /* UTF-8 support is not enabled; use the platform-dependent character literals
929 so that PCRE works on both ASCII and EBCDIC platforms, in non-UTF-mode only. */
930
931 #define CHAR_HT '\t'
932 #define CHAR_VT '\v'
933 #define CHAR_FF '\f'
934 #define CHAR_CR '\r'
935 #define CHAR_NL '\n'
936 #define CHAR_BS '\b'
937 #define CHAR_BEL '\a'
938 #ifdef EBCDIC
939 #define CHAR_ESC '\047'
940 #define CHAR_DEL '\007'
941 #else
942 #define CHAR_ESC '\033'
943 #define CHAR_DEL '\177'
944 #endif
945
946 #define CHAR_SPACE ' '
947 #define CHAR_EXCLAMATION_MARK '!'
948 #define CHAR_QUOTATION_MARK '"'
949 #define CHAR_NUMBER_SIGN '#'
950 #define CHAR_DOLLAR_SIGN '$'
951 #define CHAR_PERCENT_SIGN '%'
952 #define CHAR_AMPERSAND '&'
953 #define CHAR_APOSTROPHE '\''
954 #define CHAR_LEFT_PARENTHESIS '('
955 #define CHAR_RIGHT_PARENTHESIS ')'
956 #define CHAR_ASTERISK '*'
957 #define CHAR_PLUS '+'
958 #define CHAR_COMMA ','
959 #define CHAR_MINUS '-'
960 #define CHAR_DOT '.'
961 #define CHAR_SLASH '/'
962 #define CHAR_0 '0'
963 #define CHAR_1 '1'
964 #define CHAR_2 '2'
965 #define CHAR_3 '3'
966 #define CHAR_4 '4'
967 #define CHAR_5 '5'
968 #define CHAR_6 '6'
969 #define CHAR_7 '7'
970 #define CHAR_8 '8'
971 #define CHAR_9 '9'
972 #define CHAR_COLON ':'
973 #define CHAR_SEMICOLON ';'
974 #define CHAR_LESS_THAN_SIGN '<'
975 #define CHAR_EQUALS_SIGN '='
976 #define CHAR_GREATER_THAN_SIGN '>'
977 #define CHAR_QUESTION_MARK '?'
978 #define CHAR_COMMERCIAL_AT '@'
979 #define CHAR_A 'A'
980 #define CHAR_B 'B'
981 #define CHAR_C 'C'
982 #define CHAR_D 'D'
983 #define CHAR_E 'E'
984 #define CHAR_F 'F'
985 #define CHAR_G 'G'
986 #define CHAR_H 'H'
987 #define CHAR_I 'I'
988 #define CHAR_J 'J'
989 #define CHAR_K 'K'
990 #define CHAR_L 'L'
991 #define CHAR_M 'M'
992 #define CHAR_N 'N'
993 #define CHAR_O 'O'
994 #define CHAR_P 'P'
995 #define CHAR_Q 'Q'
996 #define CHAR_R 'R'
997 #define CHAR_S 'S'
998 #define CHAR_T 'T'
999 #define CHAR_U 'U'
1000 #define CHAR_V 'V'
1001 #define CHAR_W 'W'
1002 #define CHAR_X 'X'
1003 #define CHAR_Y 'Y'
1004 #define CHAR_Z 'Z'
1005 #define CHAR_LEFT_SQUARE_BRACKET '['
1006 #define CHAR_BACKSLASH '\\'
1007 #define CHAR_RIGHT_SQUARE_BRACKET ']'
1008 #define CHAR_CIRCUMFLEX_ACCENT '^'
1009 #define CHAR_UNDERSCORE '_'
1010 #define CHAR_GRAVE_ACCENT '`'
1011 #define CHAR_a 'a'
1012 #define CHAR_b 'b'
1013 #define CHAR_c 'c'
1014 #define CHAR_d 'd'
1015 #define CHAR_e 'e'
1016 #define CHAR_f 'f'
1017 #define CHAR_g 'g'
1018 #define CHAR_h 'h'
1019 #define CHAR_i 'i'
1020 #define CHAR_j 'j'
1021 #define CHAR_k 'k'
1022 #define CHAR_l 'l'
1023 #define CHAR_m 'm'
1024 #define CHAR_n 'n'
1025 #define CHAR_o 'o'
1026 #define CHAR_p 'p'
1027 #define CHAR_q 'q'
1028 #define CHAR_r 'r'
1029 #define CHAR_s 's'
1030 #define CHAR_t 't'
1031 #define CHAR_u 'u'
1032 #define CHAR_v 'v'
1033 #define CHAR_w 'w'
1034 #define CHAR_x 'x'
1035 #define CHAR_y 'y'
1036 #define CHAR_z 'z'
1037 #define CHAR_LEFT_CURLY_BRACKET '{'
1038 #define CHAR_VERTICAL_LINE '|'
1039 #define CHAR_RIGHT_CURLY_BRACKET '}'
1040 #define CHAR_TILDE '~'
1041
1042 #define STR_HT "\t"
1043 #define STR_VT "\v"
1044 #define STR_FF "\f"
1045 #define STR_CR "\r"
1046 #define STR_NL "\n"
1047 #define STR_BS "\b"
1048 #define STR_BEL "\a"
1049 #ifdef EBCDIC
1050 #define STR_ESC "\047"
1051 #define STR_DEL "\007"
1052 #else
1053 #define STR_ESC "\033"
1054 #define STR_DEL "\177"
1055 #endif
1056
1057 #define STR_SPACE " "
1058 #define STR_EXCLAMATION_MARK "!"
1059 #define STR_QUOTATION_MARK "\""
1060 #define STR_NUMBER_SIGN "#"
1061 #define STR_DOLLAR_SIGN "$"
1062 #define STR_PERCENT_SIGN "%"
1063 #define STR_AMPERSAND "&"
1064 #define STR_APOSTROPHE "'"
1065 #define STR_LEFT_PARENTHESIS "("
1066 #define STR_RIGHT_PARENTHESIS ")"
1067 #define STR_ASTERISK "*"
1068 #define STR_PLUS "+"
1069 #define STR_COMMA ","
1070 #define STR_MINUS "-"
1071 #define STR_DOT "."
1072 #define STR_SLASH "/"
1073 #define STR_0 "0"
1074 #define STR_1 "1"
1075 #define STR_2 "2"
1076 #define STR_3 "3"
1077 #define STR_4 "4"
1078 #define STR_5 "5"
1079 #define STR_6 "6"
1080 #define STR_7 "7"
1081 #define STR_8 "8"
1082 #define STR_9 "9"
1083 #define STR_COLON ":"
1084 #define STR_SEMICOLON ";"
1085 #define STR_LESS_THAN_SIGN "<"
1086 #define STR_EQUALS_SIGN "="
1087 #define STR_GREATER_THAN_SIGN ">"
1088 #define STR_QUESTION_MARK "?"
1089 #define STR_COMMERCIAL_AT "@"
1090 #define STR_A "A"
1091 #define STR_B "B"
1092 #define STR_C "C"
1093 #define STR_D "D"
1094 #define STR_E "E"
1095 #define STR_F "F"
1096 #define STR_G "G"
1097 #define STR_H "H"
1098 #define STR_I "I"
1099 #define STR_J "J"
1100 #define STR_K "K"
1101 #define STR_L "L"
1102 #define STR_M "M"
1103 #define STR_N "N"
1104 #define STR_O "O"
1105 #define STR_P "P"
1106 #define STR_Q "Q"
1107 #define STR_R "R"
1108 #define STR_S "S"
1109 #define STR_T "T"
1110 #define STR_U "U"
1111 #define STR_V "V"
1112 #define STR_W "W"
1113 #define STR_X "X"
1114 #define STR_Y "Y"
1115 #define STR_Z "Z"
1116 #define STR_LEFT_SQUARE_BRACKET "["
1117 #define STR_BACKSLASH "\\"
1118 #define STR_RIGHT_SQUARE_BRACKET "]"
1119 #define STR_CIRCUMFLEX_ACCENT "^"
1120 #define STR_UNDERSCORE "_"
1121 #define STR_GRAVE_ACCENT "`"
1122 #define STR_a "a"
1123 #define STR_b "b"
1124 #define STR_c "c"
1125 #define STR_d "d"
1126 #define STR_e "e"
1127 #define STR_f "f"
1128 #define STR_g "g"
1129 #define STR_h "h"
1130 #define STR_i "i"
1131 #define STR_j "j"
1132 #define STR_k "k"
1133 #define STR_l "l"
1134 #define STR_m "m"
1135 #define STR_n "n"
1136 #define STR_o "o"
1137 #define STR_p "p"
1138 #define STR_q "q"
1139 #define STR_r "r"
1140 #define STR_s "s"
1141 #define STR_t "t"
1142 #define STR_u "u"
1143 #define STR_v "v"
1144 #define STR_w "w"
1145 #define STR_x "x"
1146 #define STR_y "y"
1147 #define STR_z "z"
1148 #define STR_LEFT_CURLY_BRACKET "{"
1149 #define STR_VERTICAL_LINE "|"
1150 #define STR_RIGHT_CURLY_BRACKET "}"
1151 #define STR_TILDE "~"
1152
1153 #define STRING_ACCEPT0 "ACCEPT\0"
1154 #define STRING_COMMIT0 "COMMIT\0"
1155 #define STRING_F0 "F\0"
1156 #define STRING_FAIL0 "FAIL\0"
1157 #define STRING_MARK0 "MARK\0"
1158 #define STRING_PRUNE0 "PRUNE\0"
1159 #define STRING_SKIP0 "SKIP\0"
1160 #define STRING_THEN "THEN"
1161
1162 #define STRING_alpha0 "alpha\0"
1163 #define STRING_lower0 "lower\0"
1164 #define STRING_upper0 "upper\0"
1165 #define STRING_alnum0 "alnum\0"
1166 #define STRING_ascii0 "ascii\0"
1167 #define STRING_blank0 "blank\0"
1168 #define STRING_cntrl0 "cntrl\0"
1169 #define STRING_digit0 "digit\0"
1170 #define STRING_graph0 "graph\0"
1171 #define STRING_print0 "print\0"
1172 #define STRING_punct0 "punct\0"
1173 #define STRING_space0 "space\0"
1174 #define STRING_word0 "word\0"
1175 #define STRING_xdigit "xdigit"
1176
1177 #define STRING_DEFINE "DEFINE"
1178
1179 #define STRING_CR_RIGHTPAR "CR)"
1180 #define STRING_LF_RIGHTPAR "LF)"
1181 #define STRING_CRLF_RIGHTPAR "CRLF)"
1182 #define STRING_ANY_RIGHTPAR "ANY)"
1183 #define STRING_ANYCRLF_RIGHTPAR "ANYCRLF)"
1184 #define STRING_BSR_ANYCRLF_RIGHTPAR "BSR_ANYCRLF)"
1185 #define STRING_BSR_UNICODE_RIGHTPAR "BSR_UNICODE)"
1186 #ifdef COMPILE_PCRE8
1187 #define STRING_UTF_RIGHTPAR "UTF8)"
1188 #endif
1189 #ifdef COMPILE_PCRE16
1190 #define STRING_UTF_RIGHTPAR "UTF16)"
1191 #endif
1192 #define STRING_UCP_RIGHTPAR "UCP)"
1193 #define STRING_NO_START_OPT_RIGHTPAR "NO_START_OPT)"
1194
1195 #else /* SUPPORT_UTF */
1196
1197 /* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This
1198 works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode
1199 only. */
1200
1201 #define CHAR_HT '\011'
1202 #define CHAR_VT '\013'
1203 #define CHAR_FF '\014'
1204 #define CHAR_CR '\015'
1205 #define CHAR_NL '\012'
1206 #define CHAR_BS '\010'
1207 #define CHAR_BEL '\007'
1208 #define CHAR_ESC '\033'
1209 #define CHAR_DEL '\177'
1210
1211 #define CHAR_SPACE '\040'
1212 #define CHAR_EXCLAMATION_MARK '\041'
1213 #define CHAR_QUOTATION_MARK '\042'
1214 #define CHAR_NUMBER_SIGN '\043'
1215 #define CHAR_DOLLAR_SIGN '\044'
1216 #define CHAR_PERCENT_SIGN '\045'
1217 #define CHAR_AMPERSAND '\046'
1218 #define CHAR_APOSTROPHE '\047'
1219 #define CHAR_LEFT_PARENTHESIS '\050'
1220 #define CHAR_RIGHT_PARENTHESIS '\051'
1221 #define CHAR_ASTERISK '\052'
1222 #define CHAR_PLUS '\053'
1223 #define CHAR_COMMA '\054'
1224 #define CHAR_MINUS '\055'
1225 #define CHAR_DOT '\056'
1226 #define CHAR_SLASH '\057'
1227 #define CHAR_0 '\060'
1228 #define CHAR_1 '\061'
1229 #define CHAR_2 '\062'
1230 #define CHAR_3 '\063'
1231 #define CHAR_4 '\064'
1232 #define CHAR_5 '\065'
1233 #define CHAR_6 '\066'
1234 #define CHAR_7 '\067'
1235 #define CHAR_8 '\070'
1236 #define CHAR_9 '\071'
1237 #define CHAR_COLON '\072'
1238 #define CHAR_SEMICOLON '\073'
1239 #define CHAR_LESS_THAN_SIGN '\074'
1240 #define CHAR_EQUALS_SIGN '\075'
1241 #define CHAR_GREATER_THAN_SIGN '\076'
1242 #define CHAR_QUESTION_MARK '\077'
1243 #define CHAR_COMMERCIAL_AT '\100'
1244 #define CHAR_A '\101'
1245 #define CHAR_B '\102'
1246 #define CHAR_C '\103'
1247 #define CHAR_D '\104'
1248 #define CHAR_E '\105'
1249 #define CHAR_F '\106'
1250 #define CHAR_G '\107'
1251 #define CHAR_H '\110'
1252 #define CHAR_I '\111'
1253 #define CHAR_J '\112'
1254 #define CHAR_K '\113'
1255 #define CHAR_L '\114'
1256 #define CHAR_M '\115'
1257 #define CHAR_N '\116'
1258 #define CHAR_O '\117'
1259 #define CHAR_P '\120'
1260 #define CHAR_Q '\121'
1261 #define CHAR_R '\122'
1262 #define CHAR_S '\123'
1263 #define CHAR_T '\124'
1264 #define CHAR_U '\125'
1265 #define CHAR_V '\126'
1266 #define CHAR_W '\127'
1267 #define CHAR_X '\130'
1268 #define CHAR_Y '\131'
1269 #define CHAR_Z '\132'
1270 #define CHAR_LEFT_SQUARE_BRACKET '\133'
1271 #define CHAR_BACKSLASH '\134'
1272 #define CHAR_RIGHT_SQUARE_BRACKET '\135'
1273 #define CHAR_CIRCUMFLEX_ACCENT '\136'
1274 #define CHAR_UNDERSCORE '\137'
1275 #define CHAR_GRAVE_ACCENT '\140'
1276 #define CHAR_a '\141'
1277 #define CHAR_b '\142'
1278 #define CHAR_c '\143'
1279 #define CHAR_d '\144'
1280 #define CHAR_e '\145'
1281 #define CHAR_f '\146'
1282 #define CHAR_g '\147'
1283 #define CHAR_h '\150'
1284 #define CHAR_i '\151'
1285 #define CHAR_j '\152'
1286 #define CHAR_k '\153'
1287 #define CHAR_l '\154'
1288 #define CHAR_m '\155'
1289 #define CHAR_n '\156'
1290 #define CHAR_o '\157'
1291 #define CHAR_p '\160'
1292 #define CHAR_q '\161'
1293 #define CHAR_r '\162'
1294 #define CHAR_s '\163'
1295 #define CHAR_t '\164'
1296 #define CHAR_u '\165'
1297 #define CHAR_v '\166'
1298 #define CHAR_w '\167'
1299 #define CHAR_x '\170'
1300 #define CHAR_y '\171'
1301 #define CHAR_z '\172'
1302 #define CHAR_LEFT_CURLY_BRACKET '\173'
1303 #define CHAR_VERTICAL_LINE '\174'
1304 #define CHAR_RIGHT_CURLY_BRACKET '\175'
1305 #define CHAR_TILDE '\176'
1306
1307 #define STR_HT "\011"
1308 #define STR_VT "\013"
1309 #define STR_FF "\014"
1310 #define STR_CR "\015"
1311 #define STR_NL "\012"
1312 #define STR_BS "\010"
1313 #define STR_BEL "\007"
1314 #define STR_ESC "\033"
1315 #define STR_DEL "\177"
1316
1317 #define STR_SPACE "\040"
1318 #define STR_EXCLAMATION_MARK "\041"
1319 #define STR_QUOTATION_MARK "\042"
1320 #define STR_NUMBER_SIGN "\043"
1321 #define STR_DOLLAR_SIGN "\044"
1322 #define STR_PERCENT_SIGN "\045"
1323 #define STR_AMPERSAND "\046"
1324 #define STR_APOSTROPHE "\047"
1325 #define STR_LEFT_PARENTHESIS "\050"
1326 #define STR_RIGHT_PARENTHESIS "\051"
1327 #define STR_ASTERISK "\052"
1328 #define STR_PLUS "\053"
1329 #define STR_COMMA "\054"
1330 #define STR_MINUS "\055"
1331 #define STR_DOT "\056"
1332 #define STR_SLASH "\057"
1333 #define STR_0 "\060"
1334 #define STR_1 "\061"
1335 #define STR_2 "\062"
1336 #define STR_3 "\063"
1337 #define STR_4 "\064"
1338 #define STR_5 "\065"
1339 #define STR_6 "\066"
1340 #define STR_7 "\067"
1341 #define STR_8 "\070"
1342 #define STR_9 "\071"
1343 #define STR_COLON "\072"
1344 #define STR_SEMICOLON "\073"
1345 #define STR_LESS_THAN_SIGN "\074"
1346 #define STR_EQUALS_SIGN "\075"
1347 #define STR_GREATER_THAN_SIGN "\076"
1348 #define STR_QUESTION_MARK "\077"
1349 #define STR_COMMERCIAL_AT "\100"
1350 #define STR_A "\101"
1351 #define STR_B "\102"
1352 #define STR_C "\103"
1353 #define STR_D "\104"
1354 #define STR_E "\105"
1355 #define STR_F "\106"
1356 #define STR_G "\107"
1357 #define STR_H "\110"
1358 #define STR_I "\111"
1359 #define STR_J "\112"
1360 #define STR_K "\113"
1361 #define STR_L "\114"
1362 #define STR_M "\115"
1363 #define STR_N "\116"
1364 #define STR_O "\117"
1365 #define STR_P "\120"
1366 #define STR_Q "\121"
1367 #define STR_R "\122"
1368 #define STR_S "\123"
1369 #define STR_T "\124"
1370 #define STR_U "\125"
1371 #define STR_V "\126"
1372 #define STR_W "\127"
1373 #define STR_X "\130"
1374 #define STR_Y "\131"
1375 #define STR_Z "\132"
1376 #define STR_LEFT_SQUARE_BRACKET "\133"
1377 #define STR_BACKSLASH "\134"
1378 #define STR_RIGHT_SQUARE_BRACKET "\135"
1379 #define STR_CIRCUMFLEX_ACCENT "\136"
1380 #define STR_UNDERSCORE "\137"
1381 #define STR_GRAVE_ACCENT "\140"
1382 #define STR_a "\141"
1383 #define STR_b "\142"
1384 #define STR_c "\143"
1385 #define STR_d "\144"
1386 #define STR_e "\145"
1387 #define STR_f "\146"
1388 #define STR_g "\147"
1389 #define STR_h "\150"
1390 #define STR_i "\151"
1391 #define STR_j "\152"
1392 #define STR_k "\153"
1393 #define STR_l "\154"
1394 #define STR_m "\155"
1395 #define STR_n "\156"
1396 #define STR_o "\157"
1397 #define STR_p "\160"
1398 #define STR_q "\161"
1399 #define STR_r "\162"
1400 #define STR_s "\163"
1401 #define STR_t "\164"
1402 #define STR_u "\165"
1403 #define STR_v "\166"
1404 #define STR_w "\167"
1405 #define STR_x "\170"
1406 #define STR_y "\171"
1407 #define STR_z "\172"
1408 #define STR_LEFT_CURLY_BRACKET "\173"
1409 #define STR_VERTICAL_LINE "\174"
1410 #define STR_RIGHT_CURLY_BRACKET "\175"
1411 #define STR_TILDE "\176"
1412
1413 #define STRING_ACCEPT0 STR_A STR_C STR_C STR_E STR_P STR_T "\0"
1414 #define STRING_COMMIT0 STR_C STR_O STR_M STR_M STR_I STR_T "\0"
1415 #define STRING_F0 STR_F "\0"
1416 #define STRING_FAIL0 STR_F STR_A STR_I STR_L "\0"
1417 #define STRING_MARK0 STR_M STR_A STR_R STR_K "\0"
1418 #define STRING_PRUNE0 STR_P STR_R STR_U STR_N STR_E "\0"
1419 #define STRING_SKIP0 STR_S STR_K STR_I STR_P "\0"
1420 #define STRING_THEN STR_T STR_H STR_E STR_N
1421
1422 #define STRING_alpha0 STR_a STR_l STR_p STR_h STR_a "\0"
1423 #define STRING_lower0 STR_l STR_o STR_w STR_e STR_r "\0"
1424 #define STRING_upper0 STR_u STR_p STR_p STR_e STR_r "\0"
1425 #define STRING_alnum0 STR_a STR_l STR_n STR_u STR_m "\0"
1426 #define STRING_ascii0 STR_a STR_s STR_c STR_i STR_i "\0"
1427 #define STRING_blank0 STR_b STR_l STR_a STR_n STR_k "\0"
1428 #define STRING_cntrl0 STR_c STR_n STR_t STR_r STR_l "\0"
1429 #define STRING_digit0 STR_d STR_i STR_g STR_i STR_t "\0"
1430 #define STRING_graph0 STR_g STR_r STR_a STR_p STR_h "\0"
1431 #define STRING_print0 STR_p STR_r STR_i STR_n STR_t "\0"
1432 #define STRING_punct0 STR_p STR_u STR_n STR_c STR_t "\0"
1433 #define STRING_space0 STR_s STR_p STR_a STR_c STR_e "\0"
1434 #define STRING_word0 STR_w STR_o STR_r STR_d "\0"
1435 #define STRING_xdigit STR_x STR_d STR_i STR_g STR_i STR_t
1436
1437 #define STRING_DEFINE STR_D STR_E STR_F STR_I STR_N STR_E
1438
1439 #define STRING_CR_RIGHTPAR STR_C STR_R STR_RIGHT_PARENTHESIS
1440 #define STRING_LF_RIGHTPAR STR_L STR_F STR_RIGHT_PARENTHESIS
1441 #define STRING_CRLF_RIGHTPAR STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1442 #define STRING_ANY_RIGHTPAR STR_A STR_N STR_Y STR_RIGHT_PARENTHESIS
1443 #define STRING_ANYCRLF_RIGHTPAR STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1444 #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
1445 #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
1446 #ifdef COMPILE_PCRE8
1447 #define STRING_UTF_RIGHTPAR STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS
1448 #endif
1449 #ifdef COMPILE_PCRE16
1450 #define STRING_UTF_RIGHTPAR STR_U STR_T STR_F STR_1 STR_6 STR_RIGHT_PARENTHESIS
1451 #endif
1452 #define STRING_UCP_RIGHTPAR STR_U STR_C STR_P STR_RIGHT_PARENTHESIS
1453 #define STRING_NO_START_OPT_RIGHTPAR STR_N STR_O STR_UNDERSCORE STR_S STR_T STR_A STR_R STR_T STR_UNDERSCORE STR_O STR_P STR_T STR_RIGHT_PARENTHESIS
1454
1455 #endif /* SUPPORT_UTF */
1456
1457 /* Escape items that are just an encoding of a particular data value. */
1458
1459 #ifndef ESC_e
1460 #define ESC_e CHAR_ESC
1461 #endif
1462
1463 #ifndef ESC_f
1464 #define ESC_f CHAR_FF
1465 #endif
1466
1467 #ifndef ESC_n
1468 #define ESC_n CHAR_NL
1469 #endif
1470
1471 #ifndef ESC_r
1472 #define ESC_r CHAR_CR
1473 #endif
1474
1475 /* We can't officially use ESC_t because it is a POSIX reserved identifier
1476 (presumably because of all the others like size_t). */
1477
1478 #ifndef ESC_tee
1479 #define ESC_tee CHAR_HT
1480 #endif
1481
1482 /* Codes for different types of Unicode property */
1483
1484 #define PT_ANY 0 /* Any property - matches all chars */
1485 #define PT_LAMP 1 /* L& - the union of Lu, Ll, Lt */
1486 #define PT_GC 2 /* Specified general characteristic (e.g. L) */
1487 #define PT_PC 3 /* Specified particular characteristic (e.g. Lu) */
1488 #define PT_SC 4 /* Script (e.g. Han) */
1489 #define PT_ALNUM 5 /* Alphanumeric - the union of L and N */
1490 #define PT_SPACE 6 /* Perl space - Z plus 9,10,12,13 */
1491 #define PT_PXSPACE 7 /* POSIX space - Z plus 9,10,11,12,13 */
1492 #define PT_WORD 8 /* Word - L plus N plus underscore */
1493
1494 /* Flag bits and data types for the extended class (OP_XCLASS) for classes that
1495 contain characters with values greater than 255. */
1496
1497 #define XCL_NOT 0x01 /* Flag: this is a negative class */
1498 #define XCL_MAP 0x02 /* Flag: a 32-byte map is present */
1499
1500 #define XCL_END 0 /* Marks end of individual items */
1501 #define XCL_SINGLE 1 /* Single item (one multibyte char) follows */
1502 #define XCL_RANGE 2 /* A range (two multibyte chars) follows */
1503 #define XCL_PROP 3 /* Unicode property (2-byte property code follows) */
1504 #define XCL_NOTPROP 4 /* Unicode inverted property (ditto) */
1505
1506 /* These are escaped items that aren't just an encoding of a particular data
1507 value such as \n. They must have non-zero values, as check_escape() returns
1508 their negation. Also, they must appear in the same order as in the opcode
1509 definitions below, up to ESC_z. There's a dummy for OP_ALLANY because it
1510 corresponds to "." in DOTALL mode rather than an escape sequence. It is also
1511 used for [^] in JavaScript compatibility mode, and for \C in non-utf mode. In
1512 non-DOTALL mode, "." behaves like \N.
1513
1514 The special values ESC_DU, ESC_du, etc. are used instead of ESC_D, ESC_d, etc.
1515 when PCRE_UCP is set, when replacement of \d etc by \p sequences is required.
1516 They must be contiguous, and remain in order so that the replacements can be
1517 looked up from a table.
1518
1519 The final escape must be ESC_REF as subsequent values are used for
1520 backreferences (\1, \2, \3, etc). There are two tests in the code for an escape
1521 greater than ESC_b and less than ESC_Z to detect the types that may be
1522 repeated. These are the types that consume characters. If any new escapes are
1523 put in between that don't consume a character, that code will have to change.
1524 */
1525
1526 enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s,
1527 ESC_W, ESC_w, ESC_N, ESC_dum, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H,
1528 ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z,
1529 ESC_E, ESC_Q, ESC_g, ESC_k,
1530 ESC_DU, ESC_du, ESC_SU, ESC_su, ESC_WU, ESC_wu,
1531 ESC_REF };
1532
1533 /* Opcode table: Starting from 1 (i.e. after OP_END), the values up to
1534 OP_EOD must correspond in order to the list of escapes immediately above.
1535
1536 *** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions
1537 that follow must also be updated to match. There are also tables called
1538 "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
1539
1540 enum {
1541 OP_END, /* 0 End of pattern */
1542
1543 /* Values corresponding to backslashed metacharacters */
1544
1545 OP_SOD, /* 1 Start of data: \A */
1546 OP_SOM, /* 2 Start of match (subject + offset): \G */
1547 OP_SET_SOM, /* 3 Set start of match (\K) */
1548 OP_NOT_WORD_BOUNDARY, /* 4 \B */
1549 OP_WORD_BOUNDARY, /* 5 \b */
1550 OP_NOT_DIGIT, /* 6 \D */
1551 OP_DIGIT, /* 7 \d */
1552 OP_NOT_WHITESPACE, /* 8 \S */
1553 OP_WHITESPACE, /* 9 \s */
1554 OP_NOT_WORDCHAR, /* 10 \W */
1555 OP_WORDCHAR, /* 11 \w */
1556
1557 OP_ANY, /* 12 Match any character except newline */
1558 OP_ALLANY, /* 13 Match any character */
1559 OP_ANYBYTE, /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */
1560 OP_NOTPROP, /* 15 \P (not Unicode property) */
1561 OP_PROP, /* 16 \p (Unicode property) */
1562 OP_ANYNL, /* 17 \R (any newline sequence) */
1563 OP_NOT_HSPACE, /* 18 \H (not horizontal whitespace) */
1564 OP_HSPACE, /* 19 \h (horizontal whitespace) */
1565 OP_NOT_VSPACE, /* 20 \V (not vertical whitespace) */
1566 OP_VSPACE, /* 21 \v (vertical whitespace) */
1567 OP_EXTUNI, /* 22 \X (extended Unicode sequence */
1568 OP_EODN, /* 23 End of data or \n at end of data: \Z. */
1569 OP_EOD, /* 24 End of data: \z */
1570
1571 OP_CIRC, /* 25 Start of line - not multiline */
1572 OP_CIRCM, /* 26 Start of line - multiline */
1573 OP_DOLL, /* 27 End of line - not multiline */
1574 OP_DOLLM, /* 28 End of line - multiline */
1575 OP_CHAR, /* 29 Match one character, casefully */
1576 OP_CHARI, /* 30 Match one character, caselessly */
1577 OP_NOT, /* 31 Match one character, not the given one, casefully */
1578 OP_NOTI, /* 32 Match one character, not the given one, caselessly */
1579
1580 /* The following sets of 13 opcodes must always be kept in step because
1581 the offset from the first one is used to generate the others. */
1582
1583 /**** Single characters, caseful, must precede the caseless ones ****/
1584
1585 OP_STAR, /* 33 The maximizing and minimizing versions of */
1586 OP_MINSTAR, /* 34 these six opcodes must come in pairs, with */
1587 OP_PLUS, /* 35 the minimizing one second. */
1588 OP_MINPLUS, /* 36 */
1589 OP_QUERY, /* 37 */
1590 OP_MINQUERY, /* 38 */
1591
1592 OP_UPTO, /* 39 From 0 to n matches of one character, caseful*/
1593 OP_MINUPTO, /* 40 */
1594 OP_EXACT, /* 41 Exactly n matches */
1595
1596 OP_POSSTAR, /* 42 Possessified star, caseful */
1597 OP_POSPLUS, /* 43 Possessified plus, caseful */
1598 OP_POSQUERY, /* 44 Posesssified query, caseful */
1599 OP_POSUPTO, /* 45 Possessified upto, caseful */
1600
1601 /**** Single characters, caseless, must follow the caseful ones */
1602
1603 OP_STARI, /* 46 */
1604 OP_MINSTARI, /* 47 */
1605 OP_PLUSI, /* 48 */
1606 OP_MINPLUSI, /* 49 */
1607 OP_QUERYI, /* 50 */
1608 OP_MINQUERYI, /* 51 */
1609
1610 OP_UPTOI, /* 52 From 0 to n matches of one character, caseless */
1611 OP_MINUPTOI, /* 53 */
1612 OP_EXACTI, /* 54 */
1613
1614 OP_POSSTARI, /* 55 Possessified star, caseless */
1615 OP_POSPLUSI, /* 56 Possessified plus, caseless */
1616 OP_POSQUERYI, /* 57 Posesssified query, caseless */
1617 OP_POSUPTOI, /* 58 Possessified upto, caseless */
1618
1619 /**** The negated ones must follow the non-negated ones, and match them ****/
1620 /**** Negated single character, caseful; must precede the caseless ones ****/
1621
1622 OP_NOTSTAR, /* 59 The maximizing and minimizing versions of */
1623 OP_NOTMINSTAR, /* 60 these six opcodes must come in pairs, with */
1624 OP_NOTPLUS, /* 61 the minimizing one second. They must be in */
1625 OP_NOTMINPLUS, /* 62 exactly the same order as those above. */
1626 OP_NOTQUERY, /* 63 */
1627 OP_NOTMINQUERY, /* 64 */
1628
1629 OP_NOTUPTO, /* 65 From 0 to n matches, caseful */
1630 OP_NOTMINUPTO, /* 66 */
1631 OP_NOTEXACT, /* 67 Exactly n matches */
1632
1633 OP_NOTPOSSTAR, /* 68 Possessified versions, caseful */
1634 OP_NOTPOSPLUS, /* 69 */
1635 OP_NOTPOSQUERY, /* 70 */
1636 OP_NOTPOSUPTO, /* 71 */
1637
1638 /**** Negated single character, caseless; must follow the caseful ones ****/
1639
1640 OP_NOTSTARI, /* 72 */
1641 OP_NOTMINSTARI, /* 73 */
1642 OP_NOTPLUSI, /* 74 */
1643 OP_NOTMINPLUSI, /* 75 */
1644 OP_NOTQUERYI, /* 76 */
1645 OP_NOTMINQUERYI, /* 77 */
1646
1647 OP_NOTUPTOI, /* 78 From 0 to n matches, caseless */
1648 OP_NOTMINUPTOI, /* 79 */
1649 OP_NOTEXACTI, /* 80 Exactly n matches */
1650
1651 OP_NOTPOSSTARI, /* 81 Possessified versions, caseless */
1652 OP_NOTPOSPLUSI, /* 82 */
1653 OP_NOTPOSQUERYI, /* 83 */
1654 OP_NOTPOSUPTOI, /* 84 */
1655
1656 /**** Character types ****/
1657
1658 OP_TYPESTAR, /* 85 The maximizing and minimizing versions of */
1659 OP_TYPEMINSTAR, /* 86 these six opcodes must come in pairs, with */
1660 OP_TYPEPLUS, /* 87 the minimizing one second. These codes must */
1661 OP_TYPEMINPLUS, /* 88 be in exactly the same order as those above. */
1662 OP_TYPEQUERY, /* 89 */
1663 OP_TYPEMINQUERY, /* 90 */
1664
1665 OP_TYPEUPTO, /* 91 From 0 to n matches */
1666 OP_TYPEMINUPTO, /* 92 */
1667 OP_TYPEEXACT, /* 93 Exactly n matches */
1668
1669 OP_TYPEPOSSTAR, /* 94 Possessified versions */
1670 OP_TYPEPOSPLUS, /* 95 */
1671 OP_TYPEPOSQUERY, /* 96 */
1672 OP_TYPEPOSUPTO, /* 97 */
1673
1674 /* These are used for character classes and back references; only the
1675 first six are the same as the sets above. */
1676
1677 OP_CRSTAR, /* 98 The maximizing and minimizing versions of */
1678 OP_CRMINSTAR, /* 99 all these opcodes must come in pairs, with */
1679 OP_CRPLUS, /* 100 the minimizing one second. These codes must */
1680 OP_CRMINPLUS, /* 101 be in exactly the same order as those above. */
1681 OP_CRQUERY, /* 102 */
1682 OP_CRMINQUERY, /* 103 */
1683
1684 OP_CRRANGE, /* 104 These are different to the three sets above. */
1685 OP_CRMINRANGE, /* 105 */
1686
1687 /* End of quantifier opcodes */
1688
1689 OP_CLASS, /* 106 Match a character class, chars < 256 only */
1690 OP_NCLASS, /* 107 Same, but the bitmap was created from a negative
1691 class - the difference is relevant only when a
1692 character > 255 is encountered. */
1693 OP_XCLASS, /* 108 Extended class for handling > 255 chars within the
1694 class. This does both positive and negative. */
1695 OP_REF, /* 109 Match a back reference, casefully */
1696 OP_REFI, /* 110 Match a back reference, caselessly */
1697 OP_RECURSE, /* 111 Match a numbered subpattern (possibly recursive) */
1698 OP_CALLOUT, /* 112 Call out to external function if provided */
1699
1700 OP_ALT, /* 113 Start of alternation */
1701 OP_KET, /* 114 End of group that doesn't have an unbounded repeat */
1702 OP_KETRMAX, /* 115 These two must remain together and in this */
1703 OP_KETRMIN, /* 116 order. They are for groups the repeat for ever. */
1704 OP_KETRPOS, /* 117 Possessive unlimited repeat. */
1705
1706 /* The assertions must come before BRA, CBRA, ONCE, and COND, and the four
1707 asserts must remain in order. */
1708
1709 OP_REVERSE, /* 118 Move pointer back - used in lookbehind assertions */
1710 OP_ASSERT, /* 119 Positive lookahead */
1711 OP_ASSERT_NOT, /* 120 Negative lookahead */
1712 OP_ASSERTBACK, /* 121 Positive lookbehind */
1713 OP_ASSERTBACK_NOT, /* 122 Negative lookbehind */
1714
1715 /* ONCE, ONCE_NC, BRA, BRAPOS, CBRA, CBRAPOS, and COND must come immediately
1716 after the assertions, with ONCE first, as there's a test for >= ONCE for a
1717 subpattern that isn't an assertion. The POS versions must immediately follow
1718 the non-POS versions in each case. */
1719
1720 OP_ONCE, /* 123 Atomic group, contains captures */
1721 OP_ONCE_NC, /* 124 Atomic group containing no captures */
1722 OP_BRA, /* 125 Start of non-capturing bracket */
1723 OP_BRAPOS, /* 126 Ditto, with unlimited, possessive repeat */
1724 OP_CBRA, /* 127 Start of capturing bracket */
1725 OP_CBRAPOS, /* 128 Ditto, with unlimited, possessive repeat */
1726 OP_COND, /* 129 Conditional group */
1727
1728 /* These five must follow the previous five, in the same order. There's a
1729 check for >= SBRA to distinguish the two sets. */
1730
1731 OP_SBRA, /* 130 Start of non-capturing bracket, check empty */
1732 OP_SBRAPOS, /* 131 Ditto, with unlimited, possessive repeat */
1733 OP_SCBRA, /* 132 Start of capturing bracket, check empty */
1734 OP_SCBRAPOS, /* 133 Ditto, with unlimited, possessive repeat */
1735 OP_SCOND, /* 134 Conditional group, check empty */
1736
1737 /* The next two pairs must (respectively) be kept together. */
1738
1739 OP_CREF, /* 135 Used to hold a capture number as condition */
1740 OP_NCREF, /* 136 Same, but generated by a name reference*/
1741 OP_RREF, /* 137 Used to hold a recursion number as condition */
1742 OP_NRREF, /* 138 Same, but generated by a name reference*/
1743 OP_DEF, /* 139 The DEFINE condition */
1744
1745 OP_BRAZERO, /* 140 These two must remain together and in this */
1746 OP_BRAMINZERO, /* 141 order. */
1747 OP_BRAPOSZERO, /* 142 */
1748
1749 /* These are backtracking control verbs */
1750
1751 OP_MARK, /* 143 always has an argument */
1752 OP_PRUNE, /* 144 */
1753 OP_PRUNE_ARG, /* 145 same, but with argument */
1754 OP_SKIP, /* 146 */
1755 OP_SKIP_ARG, /* 147 same, but with argument */
1756 OP_THEN, /* 148 */
1757 OP_THEN_ARG, /* 149 same, but with argument */
1758 OP_COMMIT, /* 150 */
1759
1760 /* These are forced failure and success verbs */
1761
1762 OP_FAIL, /* 151 */
1763 OP_ACCEPT, /* 152 */
1764 OP_ASSERT_ACCEPT, /* 153 Used inside assertions */
1765 OP_CLOSE, /* 154 Used before OP_ACCEPT to close open captures */
1766
1767 /* This is used to skip a subpattern with a {0} quantifier */
1768
1769 OP_SKIPZERO, /* 155 */
1770
1771 /* This is not an opcode, but is used to check that tables indexed by opcode
1772 are the correct length, in order to catch updating errors - there have been
1773 some in the past. */
1774
1775 OP_TABLE_LENGTH
1776 };
1777
1778 /* *** NOTE NOTE NOTE *** Whenever the list above is updated, the two macro
1779 definitions that follow must also be updated to match. There are also tables
1780 called "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
1781
1782
1783 /* This macro defines textual names for all the opcodes. These are used only
1784 for debugging, and some of them are only partial names. The macro is referenced
1785 only in pcre_printint.c, which fills out the full names in many cases (and in
1786 some cases doesn't actually use these names at all). */
1787
1788 #define OP_NAME_LIST \
1789 "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d", \
1790 "\\S", "\\s", "\\W", "\\w", "Any", "AllAny", "Anybyte", \
1791 "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v", \
1792 "extuni", "\\Z", "\\z", \
1793 "^", "^", "$", "$", "char", "chari", "not", "noti", \
1794 "*", "*?", "+", "+?", "?", "??", \
1795 "{", "{", "{", \
1796 "*+","++", "?+", "{", \
1797 "*", "*?", "+", "+?", "?", "??", \
1798 "{", "{", "{", \
1799 "*+","++", "?+", "{", \
1800 "*", "*?", "+", "+?", "?", "??", \
1801 "{", "{", "{", \
1802 "*+","++", "?+", "{", \
1803 "*", "*?", "+", "+?", "?", "??", \
1804 "{", "{", "{", \
1805 "*+","++", "?+", "{", \
1806 "*", "*?", "+", "+?", "?", "??", "{", "{", "{", \
1807 "*+","++", "?+", "{", \
1808 "*", "*?", "+", "+?", "?", "??", "{", "{", \
1809 "class", "nclass", "xclass", "Ref", "Refi", \
1810 "Recurse", "Callout", \
1811 "Alt", "Ket", "KetRmax", "KetRmin", "KetRpos", \
1812 "Reverse", "Assert", "Assert not", "AssertB", "AssertB not", \
1813 "Once", "Once_NC", \
1814 "Bra", "BraPos", "CBra", "CBraPos", \
1815 "Cond", \
1816 "SBra", "SBraPos", "SCBra", "SCBraPos", \
1817 "SCond", \
1818 "Cond ref", "Cond nref", "Cond rec", "Cond nrec", "Cond def", \
1819 "Brazero", "Braminzero", "Braposzero", \
1820 "*MARK", "*PRUNE", "*PRUNE", "*SKIP", "*SKIP", \
1821 "*THEN", "*THEN", "*COMMIT", "*FAIL", \
1822 "*ACCEPT", "*ASSERT_ACCEPT", \
1823 "Close", "Skip zero"
1824
1825
1826 /* This macro defines the length of fixed length operations in the compiled
1827 regex. The lengths are used when searching for specific things, and also in the
1828 debugging printing of a compiled regex. We use a macro so that it can be
1829 defined close to the definitions of the opcodes themselves.
1830
1831 As things have been extended, some of these are no longer fixed lenths, but are
1832 minima instead. For example, the length of a single-character repeat may vary
1833 in UTF-8 mode. The code that uses this table must know about such things. */
1834
1835 #define OP_LENGTHS \
1836 1, /* End */ \
1837 1, 1, 1, 1, 1, /* \A, \G, \K, \B, \b */ \
1838 1, 1, 1, 1, 1, 1, /* \D, \d, \S, \s, \W, \w */ \
1839 1, 1, 1, /* Any, AllAny, Anybyte */ \
1840 3, 3, /* \P, \p */ \
1841 1, 1, 1, 1, 1, /* \R, \H, \h, \V, \v */ \
1842 1, /* \X */ \
1843 1, 1, 1, 1, 1, 1, /* \Z, \z, ^, ^M, $, $M */ \
1844 2, /* Char - the minimum length */ \
1845 2, /* Chari - the minimum length */ \
1846 2, /* not */ \
1847 2, /* noti */ \
1848 /* Positive single-char repeats ** These are */ \
1849 2, 2, 2, 2, 2, 2, /* *, *?, +, +?, ?, ?? ** minima in */ \
1850 2+IMM2_SIZE, 2+IMM2_SIZE, /* upto, minupto ** mode */ \
1851 2+IMM2_SIZE, /* exact */ \
1852 2, 2, 2, 2+IMM2_SIZE, /* *+, ++, ?+, upto+ */ \
1853 2, 2, 2, 2, 2, 2, /* *I, *?I, +I, +?I, ?I, ??I ** UTF-8 */ \
1854 2+IMM2_SIZE, 2+IMM2_SIZE, /* upto I, minupto I */ \
1855 2+IMM2_SIZE, /* exact I */ \
1856 2, 2, 2, 2+IMM2_SIZE, /* *+I, ++I, ?+I, upto+I */ \
1857 /* Negative single-char repeats - only for chars < 256 */ \
1858 2, 2, 2, 2, 2, 2, /* NOT *, *?, +, +?, ?, ?? */ \
1859 2+IMM2_SIZE, 2+IMM2_SIZE, /* NOT upto, minupto */ \
1860 2+IMM2_SIZE, /* NOT exact */ \
1861 2, 2, 2, 2+IMM2_SIZE, /* Possessive NOT *, +, ?, upto */ \
1862 2, 2, 2, 2, 2, 2, /* NOT *I, *?I, +I, +?I, ?I, ??I */ \
1863 2+IMM2_SIZE, 2+IMM2_SIZE, /* NOT upto I, minupto I */ \
1864 2+IMM2_SIZE, /* NOT exact I */ \
1865 2, 2, 2, 2+IMM2_SIZE, /* Possessive NOT *I, +I, ?I, upto I */ \
1866 /* Positive type repeats */ \
1867 2, 2, 2, 2, 2, 2, /* Type *, *?, +, +?, ?, ?? */ \
1868 2+IMM2_SIZE, 2+IMM2_SIZE, /* Type upto, minupto */ \
1869 2+IMM2_SIZE, /* Type exact */ \
1870 2, 2, 2, 2+IMM2_SIZE, /* Possessive *+, ++, ?+, upto+ */ \
1871 /* Character class & ref repeats */ \
1872 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ \
1873 1+2*IMM2_SIZE, 1+2*IMM2_SIZE, /* CRRANGE, CRMINRANGE */ \
1874 1+(32/sizeof(pcre_uchar)), /* CLASS */ \
1875 1+(32/sizeof(pcre_uchar)), /* NCLASS */ \
1876 0, /* XCLASS - variable length */ \
1877 1+IMM2_SIZE, /* REF */ \
1878 1+IMM2_SIZE, /* REFI */ \
1879 1+LINK_SIZE, /* RECURSE */ \
1880 2+2*LINK_SIZE, /* CALLOUT */ \
1881 1+LINK_SIZE, /* Alt */ \
1882 1+LINK_SIZE, /* Ket */ \
1883 1+LINK_SIZE, /* KetRmax */ \
1884 1+LINK_SIZE, /* KetRmin */ \
1885 1+LINK_SIZE, /* KetRpos */ \
1886 1+LINK_SIZE, /* Reverse */ \
1887 1+LINK_SIZE, /* Assert */ \
1888 1+LINK_SIZE, /* Assert not */ \
1889 1+LINK_SIZE, /* Assert behind */ \
1890 1+LINK_SIZE, /* Assert behind not */ \
1891 1+LINK_SIZE, /* ONCE */ \
1892 1+LINK_SIZE, /* ONCE_NC */ \
1893 1+LINK_SIZE, /* BRA */ \
1894 1+LINK_SIZE, /* BRAPOS */ \
1895 1+LINK_SIZE+IMM2_SIZE, /* CBRA */ \
1896 1+LINK_SIZE+IMM2_SIZE, /* CBRAPOS */ \
1897 1+LINK_SIZE, /* COND */ \
1898 1+LINK_SIZE, /* SBRA */ \
1899 1+LINK_SIZE, /* SBRAPOS */ \
1900 1+LINK_SIZE+IMM2_SIZE, /* SCBRA */ \
1901 1+LINK_SIZE+IMM2_SIZE, /* SCBRAPOS */ \
1902 1+LINK_SIZE, /* SCOND */ \
1903 1+IMM2_SIZE, 1+IMM2_SIZE, /* CREF, NCREF */ \
1904 1+IMM2_SIZE, 1+IMM2_SIZE, /* RREF, NRREF */ \
1905 1, /* DEF */ \
1906 1, 1, 1, /* BRAZERO, BRAMINZERO, BRAPOSZERO */ \
1907 3, 1, 3, /* MARK, PRUNE, PRUNE_ARG */ \
1908 1, 3, /* SKIP, SKIP_ARG */ \
1909 1, 3, /* THEN, THEN_ARG */ \
1910 1, 1, 1, 1, /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT */ \
1911 1+IMM2_SIZE, 1 /* CLOSE, SKIPZERO */
1912
1913 /* A magic value for OP_RREF and OP_NRREF to indicate the "any recursion"
1914 condition. */
1915
1916 #define RREF_ANY 0xffff
1917
1918 /* Compile time error code numbers. They are given names so that they can more
1919 easily be tracked. When a new number is added, the table called eint in
1920 pcreposix.c must be updated. */
1921
1922 enum { ERR0, ERR1, ERR2, ERR3, ERR4, ERR5, ERR6, ERR7, ERR8, ERR9,
1923 ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,
1924 ERR20, ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR28, ERR29,
1925 ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,
1926 ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,
1927 ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,
1928 ERR60, ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERR68, ERR69,
1929 ERR70, ERR71, ERRCOUNT };
1930
1931 /* The real format of the start of the pcre block; the index of names and the
1932 code vector run on as long as necessary after the end. We store an explicit
1933 offset to the name table so that if a regex is compiled on one host, saved, and
1934 then run on another where the size of pointers is different, all might still
1935 be well. For the case of compiled-on-4 and run-on-8, we include an extra
1936 pointer that is always NULL. For future-proofing, a few dummy fields were
1937 originally included - even though you can never get this planning right - but
1938 there is only one left now.
1939
1940 NOTE NOTE NOTE:
1941 Because people can now save and re-use compiled patterns, any additions to this
1942 structure should be made at the end, and something earlier (e.g. a new
1943 flag in the options or one of the dummy fields) should indicate that the new
1944 fields are present. Currently PCRE always sets the dummy fields to zero.
1945 NOTE NOTE NOTE
1946 */
1947
1948 typedef struct real_pcre {
1949 pcre_uint32 magic_number;
1950 pcre_uint32 size; /* Total that was malloced */
1951 pcre_uint32 options; /* Public options */
1952 pcre_uint16 flags; /* Private flags */
1953 pcre_uint16 dummy1; /* For future use */
1954 pcre_uint16 top_bracket;
1955 pcre_uint16 top_backref;
1956 pcre_uint16 first_char; /* Starting character */
1957 pcre_uint16 req_char; /* This character must be seen */
1958 pcre_uint16 name_table_offset; /* Offset to name table that follows */
1959 pcre_uint16 name_entry_size; /* Size of any name items */
1960 pcre_uint16 name_count; /* Number of name items */
1961 pcre_uint16 ref_count; /* Reference count */
1962
1963 const pcre_uint8 *tables; /* Pointer to tables or NULL for std */
1964 const pcre_uint8 *nullpad; /* NULL padding */
1965 } real_pcre;
1966
1967 /* The format of the block used to store data from pcre_study(). The same
1968 remark (see NOTE above) about extending this structure applies. */
1969
1970 typedef struct pcre_study_data {
1971 pcre_uint32 size; /* Total that was malloced */
1972 pcre_uint32 flags; /* Private flags */
1973 pcre_uint8 start_bits[32]; /* Starting char bits */
1974 pcre_uint32 minlength; /* Minimum subject length */
1975 } pcre_study_data;
1976
1977 /* Structure for building a chain of open capturing subpatterns during
1978 compiling, so that instructions to close them can be compiled when (*ACCEPT) is
1979 encountered. This is also used to identify subpatterns that contain recursive
1980 back references to themselves, so that they can be made atomic. */
1981
1982 typedef struct open_capitem {
1983 struct open_capitem *next; /* Chain link */
1984 pcre_uint16 number; /* Capture number */
1985 pcre_uint16 flag; /* Set TRUE if recursive back ref */
1986 } open_capitem;
1987
1988 /* Structure for passing "static" information around between the functions
1989 doing the compiling, so that they are thread-safe. */
1990
1991 typedef struct compile_data {
1992 const pcre_uint8 *lcc; /* Points to lower casing table */
1993 const pcre_uint8 *fcc; /* Points to case-flipping table */
1994 const pcre_uint8 *cbits; /* Points to character type table */
1995 const pcre_uint8 *ctypes; /* Points to table of type maps */
1996 const pcre_uchar *start_workspace;/* The start of working space */
1997 const pcre_uchar *start_code; /* The start of the compiled code */
1998 const pcre_uchar *start_pattern; /* The start of the pattern */
1999 const pcre_uchar *end_pattern; /* The end of the pattern */
2000 open_capitem *open_caps; /* Chain of open capture items */
2001 pcre_uchar *hwm; /* High watermark of workspace */
2002 pcre_uchar *name_table; /* The name/number table */
2003 int names_found; /* Number of entries so far */
2004 int name_entry_size; /* Size of each entry */
2005 int bracount; /* Count of capturing parens as we compile */
2006 int final_bracount; /* Saved value after first pass */
2007 int top_backref; /* Maximum back reference */
2008 unsigned int backref_map; /* Bitmap of low back refs */
2009 int assert_depth; /* Depth of nested assertions */
2010 int external_options; /* External (initial) options */
2011 int external_flags; /* External flag bits to be set */
2012 int req_varyopt; /* "After variable item" flag for reqbyte */
2013 BOOL had_accept; /* (*ACCEPT) encountered */
2014 BOOL check_lookbehind; /* Lookbehinds need later checking */
2015 int nltype; /* Newline type */
2016 int nllen; /* Newline string length */
2017 pcre_uchar nl[4]; /* Newline string when fixed length */
2018 } compile_data;
2019
2020 /* Structure for maintaining a chain of pointers to the currently incomplete
2021 branches, for testing for left recursion while compiling. */
2022
2023 typedef struct branch_chain {
2024 struct branch_chain *outer;
2025 pcre_uchar *current_branch;
2026 } branch_chain;
2027
2028 /* Structure for items in a linked list that represents an explicit recursive
2029 call within the pattern; used by pcre_exec(). */
2030
2031 typedef struct recursion_info {
2032 struct recursion_info *prevrec; /* Previous recursion record (or NULL) */
2033 int group_num; /* Number of group that was called */
2034 int *offset_save; /* Pointer to start of saved offsets */
2035 int saved_max; /* Number of saved offsets */
2036 PCRE_PUCHAR subject_position; /* Position at start of recursion */
2037 } recursion_info;
2038
2039 /* A similar structure for pcre_dfa_exec(). */
2040
2041 typedef struct dfa_recursion_info {
2042 struct dfa_recursion_info *prevrec;
2043 int group_num;
2044 PCRE_PUCHAR subject_position;
2045 } dfa_recursion_info;
2046
2047 /* Structure for building a chain of data for holding the values of the subject
2048 pointer at the start of each subpattern, so as to detect when an empty string
2049 has been matched by a subpattern - to break infinite loops; used by
2050 pcre_exec(). */
2051
2052 typedef struct eptrblock {
2053 struct eptrblock *epb_prev;
2054 PCRE_PUCHAR epb_saved_eptr;
2055 } eptrblock;
2056
2057
2058 /* Structure for passing "static" information around between the functions
2059 doing traditional NFA matching, so that they are thread-safe. */
2060
2061 typedef struct match_data {
2062 unsigned long int match_call_count; /* As it says */
2063 unsigned long int match_limit; /* As it says */
2064 unsigned long int match_limit_recursion; /* As it says */
2065 int *offset_vector; /* Offset vector */
2066 int offset_end; /* One past the end */
2067 int offset_max; /* The maximum usable for return data */
2068 int nltype; /* Newline type */
2069 int nllen; /* Newline string length */
2070 int name_count; /* Number of names in name table */
2071 int name_entry_size; /* Size of entry in names table */
2072 pcre_uchar *name_table; /* Table of names */
2073 pcre_uchar nl[4]; /* Newline string when fixed */
2074 const pcre_uint8 *lcc; /* Points to lower casing table */
2075 const pcre_uint8 *fcc; /* Points to case-flipping table */
2076 const pcre_uint8 *ctypes; /* Points to table of type maps */
2077 BOOL offset_overflow; /* Set if too many extractions */
2078 BOOL notbol; /* NOTBOL flag */
2079 BOOL noteol; /* NOTEOL flag */
2080 BOOL utf; /* UTF-8 / UTF-16 flag */
2081 BOOL jscript_compat; /* JAVASCRIPT_COMPAT flag */
2082 BOOL use_ucp; /* PCRE_UCP flag */
2083 BOOL endonly; /* Dollar not before final \n */
2084 BOOL notempty; /* Empty string match not wanted */
2085 BOOL notempty_atstart; /* Empty string match at start not wanted */
2086 BOOL hitend; /* Hit the end of the subject at some point */
2087 BOOL bsr_anycrlf; /* \R is just any CRLF, not full Unicode */
2088 BOOL hasthen; /* Pattern contains (*THEN) */
2089 const pcre_uchar *start_code; /* For use when recursing */
2090 PCRE_PUCHAR start_subject; /* Start of the subject string */
2091 PCRE_PUCHAR end_subject; /* End of the subject string */
2092 PCRE_PUCHAR start_match_ptr; /* Start of matched string */
2093 PCRE_PUCHAR end_match_ptr; /* Subject position at end match */
2094 PCRE_PUCHAR start_used_ptr; /* Earliest consulted character */
2095 int partial; /* PARTIAL options */
2096 int end_offset_top; /* Highwater mark at end of match */
2097 int capture_last; /* Most recent capture number */
2098 int start_offset; /* The start offset value */
2099 int match_function_type; /* Set for certain special calls of MATCH() */
2100 eptrblock *eptrchain; /* Chain of eptrblocks for tail recursions */
2101 int eptrn; /* Next free eptrblock */
2102 recursion_info *recursive; /* Linked list of recursion data */
2103 void *callout_data; /* To pass back to callouts */
2104 const pcre_uchar *mark; /* Mark pointer to pass back */
2105 const pcre_uchar *once_target; /* Where to back up to for atomic groups */
2106 } match_data;
2107
2108 /* A similar structure is used for the same purpose by the DFA matching
2109 functions. */
2110
2111 typedef struct dfa_match_data {
2112 const pcre_uchar *start_code; /* Start of the compiled pattern */
2113 const pcre_uchar *start_subject ; /* Start of the subject string */
2114 const pcre_uchar *end_subject; /* End of subject string */
2115 const pcre_uchar *start_used_ptr; /* Earliest consulted character */
2116 const pcre_uint8 *tables; /* Character tables */
2117 int start_offset; /* The start offset value */
2118 int moptions; /* Match options */
2119 int poptions; /* Pattern options */
2120 int nltype; /* Newline type */
2121 int nllen; /* Newline string length */
2122 pcre_uchar nl[4]; /* Newline string when fixed */
2123 void *callout_data; /* To pass back to callouts */
2124 dfa_recursion_info *recursive; /* Linked list of recursion data */
2125 } dfa_match_data;
2126
2127 /* Bit definitions for entries in the pcre_ctypes table. */
2128
2129 #define ctype_space 0x01
2130 #define ctype_letter 0x02
2131 #define ctype_digit 0x04
2132 #define ctype_xdigit 0x08
2133 #define ctype_word 0x10 /* alphanumeric or '_' */
2134 #define ctype_meta 0x80 /* regexp meta char or zero (end pattern) */
2135
2136 /* Offsets for the bitmap tables in pcre_cbits. Each table contains a set
2137 of bits for a class map. Some classes are built by combining these tables. */
2138
2139 #define cbit_space 0 /* [:space:] or \s */
2140 #define cbit_xdigit 32 /* [:xdigit:] */
2141 #define cbit_digit 64 /* [:digit:] or \d */
2142 #define cbit_upper 96 /* [:upper:] */
2143 #define cbit_lower 128 /* [:lower:] */
2144 #define cbit_word 160 /* [:word:] or \w */
2145 #define cbit_graph 192 /* [:graph:] */
2146 #define cbit_print 224 /* [:print:] */
2147 #define cbit_punct 256 /* [:punct:] */
2148 #define cbit_cntrl 288 /* [:cntrl:] */
2149 #define cbit_length 320 /* Length of the cbits table */
2150
2151 /* Offsets of the various tables from the base tables pointer, and
2152 total length. */
2153
2154 #define lcc_offset 0
2155 #define fcc_offset 256
2156 #define cbits_offset 512
2157 #define ctypes_offset (cbits_offset + cbit_length)
2158 #define tables_length (ctypes_offset + 256)
2159
2160 /* Internal function prefix */
2161
2162 #ifdef COMPILE_PCRE8
2163 #define PRIV(name) _pcre_##name
2164 #else
2165 #ifdef COMPILE_PCRE16
2166 #define PRIV(name) _pcre16_##name
2167 #else
2168 #error Unsupported compiling mode
2169 #endif /* COMPILE_PCRE16 */
2170 #endif /* COMPILE_PCRE8 */
2171
2172 /* Layout of the UCP type table that translates property names into types and
2173 codes. Each entry used to point directly to a name, but to reduce the number of
2174 relocations in shared libraries, it now has an offset into a single string
2175 instead. */
2176
2177 typedef struct {
2178 pcre_uint16 name_offset;
2179 pcre_uint16 type;
2180 pcre_uint16 value;
2181 } ucp_type_table;
2182
2183
2184 /* Internal shared data tables. These are tables that are used by more than one
2185 of the exported public functions. They have to be "external" in the C sense,
2186 but are not part of the PCRE public API. The data for these tables is in the
2187 pcre_tables.c module. */
2188
2189 #ifdef COMPILE_PCRE8
2190
2191 extern const int PRIV(utf8_table1)[];
2192 extern const int PRIV(utf8_table1_size);
2193 extern const int PRIV(utf8_table2)[];
2194 extern const int PRIV(utf8_table3)[];
2195 extern const pcre_uint8 PRIV(utf8_table4)[];
2196
2197 #endif /* COMPILE_PCRE8 */
2198
2199 extern const char PRIV(utt_names)[];
2200 extern const ucp_type_table PRIV(utt)[];
2201 extern const int PRIV(utt_size);
2202
2203 extern const pcre_uint8 PRIV(default_tables)[];
2204
2205 extern const pcre_uint8 PRIV(OP_lengths)[];
2206
2207
2208 /* Internal shared functions. These are functions that are used by more than
2209 one of the exported public functions. They have to be "external" in the C
2210 sense, but are not part of the PCRE public API. */
2211
2212 /* String comparison functions. */
2213 #ifdef COMPILE_PCRE8
2214
2215 #define STRCMP_UC_UC(str1, str2) \
2216 strcmp((char *)(str1), (char *)(str2))
2217 #define STRCMP_UC_C8(str1, str2) \
2218 strcmp((char *)(str1), (str2))
2219 #define STRNCMP_UC_UC(str1, str2, num) \
2220 strncmp((char *)(str1), (char *)(str2), (num))
2221 #define STRNCMP_UC_C8(str1, str2, num) \
2222 strncmp((char *)(str1), (str2), (num))
2223 #define STRLEN_UC(str) strlen((const char *)str)
2224
2225 #else
2226
2227 extern int PRIV(strcmp_uc_uc)(const pcre_uchar *,
2228 const pcre_uchar *);
2229 extern int PRIV(strcmp_uc_c8)(const pcre_uchar *,
2230 const char *);
2231 extern int PRIV(strncmp_uc_uc)(const pcre_uchar *,
2232 const pcre_uchar *, unsigned int num);
2233 extern int PRIV(strncmp_uc_c8)(const pcre_uchar *,
2234 const char *, unsigned int num);
2235 extern unsigned int PRIV(strlen_uc)(const pcre_uchar *str);
2236
2237 #define STRCMP_UC_UC(str1, str2) \
2238 PRIV(strcmp_uc_uc)((str1), (str2))
2239 #define STRCMP_UC_C8(str1, str2) \
2240 PRIV(strcmp_uc_c8)((str1), (str2))
2241 #define STRNCMP_UC_UC(str1, str2, num) \
2242 PRIV(strncmp_uc_uc)((str1), (str2), (num))
2243 #define STRNCMP_UC_C8(str1, str2, num) \
2244 PRIV(strncmp_uc_c8)((str1), (str2), (num))
2245 #define STRLEN_UC(str) PRIV(strlen_uc)(str)
2246
2247 #endif /* COMPILE_PCRE8 */
2248
2249 extern const pcre_uchar *PRIV(find_bracket)(const pcre_uchar *, BOOL, int);
2250 extern BOOL PRIV(is_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2251 int *, BOOL);
2252 extern int PRIV(ord2utf)(pcre_uint32, pcre_uchar *);
2253 extern real_pcre *PRIV(try_flipped)(const real_pcre *, real_pcre *,
2254 const pcre_study_data *, pcre_study_data *);
2255 extern int PRIV(valid_utf)(PCRE_PUCHAR, int, int *);
2256 extern BOOL PRIV(was_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2257 int *, BOOL);
2258 extern BOOL PRIV(xclass)(int, const pcre_uchar *, BOOL);
2259
2260 #ifdef SUPPORT_JIT
2261 extern void PRIV(jit_compile)(const real_pcre *, pcre_extra *);
2262 extern int PRIV(jit_exec)(const real_pcre *, void *,
2263 const pcre_uchar *, int, int, int, int, int *, int);
2264 extern void PRIV(jit_free)(void *);
2265 #endif
2266
2267 /* Unicode character database (UCD) */
2268
2269 typedef struct {
2270 pcre_uint8 script;
2271 pcre_uint8 chartype;
2272 pcre_int32 other_case;
2273 } ucd_record;
2274
2275 extern const ucd_record PRIV(ucd_records)[];
2276 extern const pcre_uint8 PRIV(ucd_stage1)[];
2277 extern const pcre_uint16 PRIV(ucd_stage2)[];
2278 extern const int PRIV(ucp_gentype)[];
2279 #ifdef SUPPORT_JIT
2280 extern const int PRIV(ucp_typerange)[];
2281 #endif
2282
2283 #ifdef SUPPORT_UCP
2284 /* UCD access macros */
2285
2286 #define UCD_BLOCK_SIZE 128
2287 #define GET_UCD(ch) (PRIV(ucd_records) + \
2288 PRIV(ucd_stage2)[PRIV(ucd_stage1)[(ch) / UCD_BLOCK_SIZE] * \
2289 UCD_BLOCK_SIZE + (ch) % UCD_BLOCK_SIZE])
2290
2291 #define UCD_CHARTYPE(ch) GET_UCD(ch)->chartype
2292 #define UCD_SCRIPT(ch) GET_UCD(ch)->script
2293 #define UCD_CATEGORY(ch) PRIV(ucp_gentype)[UCD_CHARTYPE(ch)]
2294 #define UCD_OTHERCASE(ch) (ch + GET_UCD(ch)->other_case)
2295
2296 #endif /* SUPPORT_UCP */
2297
2298 #endif
2299
2300 /* End of pcre_internal.h */

Properties

Name Value
svn:eol-style native
svn:keywords "Author Date Id Revision Url"

  ViewVC Help
Powered by ViewVC 1.1.5