/[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 770 - (show annotations)
Mon Nov 28 20:39:30 2011 UTC (7 years, 9 months ago) by zherczeg
File MIME type: text/plain
File size: 88774 byte(s)
Make character ranges 16 bit friendly
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 /* We do not support both EBCDIC and UTF-8/16 at the same time. The "configure"
61 script prevents both being selected, but not everybody uses "configure". */
62
63 #if defined EBCDIC && (defined SUPPORT_UTF8 || defined SUPPORT_UTF16)
64 #error The use of both EBCDIC and SUPPORT_UTF8/16 is not supported.
65 #endif
66
67 /* If SUPPORT_UCP is defined, SUPPORT_UTF8 must also be defined. The
68 "configure" script ensures this, but not everybody uses "configure". */
69
70 #if defined SUPPORT_UCP && !defined SUPPORT_UTF8
71 #define SUPPORT_UTF8 1
72 #endif
73
74 /* If SUPPORT_UCP is defined, SUPPORT_UTF16 must also be defined. The
75 "configure" script ensures this, but not everybody uses "configure". */
76
77 #if defined SUPPORT_UCP && defined COMPILE_PCRE16 && !defined SUPPORT_UTF16
78 #define SUPPORT_UTF16 1
79 #endif
80
81 /* This macro is defined if either UTF-8 or UTF-16 support or both are
82 enabled. */
83
84 #if defined SUPPORT_UTF8 || defined SUPPORT_UTF16
85 /* Unicode Transformation Format is enabled. */
86 #define SUPPORT_UTF 1
87 #endif
88
89 /* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef
90 inline, and there are *still* stupid compilers about that don't like indented
91 pre-processor statements, or at least there were when I first wrote this. After
92 all, it had only been about 10 years then...
93
94 It turns out that the Mac Debugging.h header also defines the macro DPRINTF, so
95 be absolutely sure we get our version. */
96
97 #undef DPRINTF
98 #ifdef PCRE_DEBUG
99 #define DPRINTF(p) printf p
100 #else
101 #define DPRINTF(p) /* Nothing */
102 #endif
103
104
105 /* Standard C headers plus the external interface definition. The only time
106 setjmp and stdarg are used is when NO_RECURSE is set. */
107
108 #include <ctype.h>
109 #include <limits.h>
110 #include <stddef.h>
111 #include <stdio.h>
112 #include <stdlib.h>
113 #include <string.h>
114
115 /* When compiling a DLL for Windows, the exported symbols have to be declared
116 using some MS magic. I found some useful information on this web page:
117 http://msdn2.microsoft.com/en-us/library/y4h7bcy6(VS.80).aspx. According to the
118 information there, using __declspec(dllexport) without "extern" we have a
119 definition; with "extern" we have a declaration. The settings here override the
120 setting in pcre.h (which is included below); it defines only PCRE_EXP_DECL,
121 which is all that is needed for applications (they just import the symbols). We
122 use:
123
124 PCRE_EXP_DECL for declarations
125 PCRE_EXP_DEFN for definitions of exported functions
126 PCRE_EXP_DATA_DEFN for definitions of exported variables
127
128 The reason for the two DEFN macros is that in non-Windows environments, one
129 does not want to have "extern" before variable definitions because it leads to
130 compiler warnings. So we distinguish between functions and variables. In
131 Windows, the two should always be the same.
132
133 The reason for wrapping this in #ifndef PCRE_EXP_DECL is so that pcretest,
134 which is an application, but needs to import this file in order to "peek" at
135 internals, can #include pcre.h first to get an application's-eye view.
136
137 In principle, people compiling for non-Windows, non-Unix-like (i.e. uncommon,
138 special-purpose environments) might want to stick other stuff in front of
139 exported symbols. That's why, in the non-Windows case, we set PCRE_EXP_DEFN and
140 PCRE_EXP_DATA_DEFN only if they are not already set. */
141
142 #ifndef PCRE_EXP_DECL
143 # ifdef _WIN32
144 # ifndef PCRE_STATIC
145 # define PCRE_EXP_DECL extern __declspec(dllexport)
146 # define PCRE_EXP_DEFN __declspec(dllexport)
147 # define PCRE_EXP_DATA_DEFN __declspec(dllexport)
148 # else
149 # define PCRE_EXP_DECL extern
150 # define PCRE_EXP_DEFN
151 # define PCRE_EXP_DATA_DEFN
152 # endif
153 # else
154 # ifdef __cplusplus
155 # define PCRE_EXP_DECL extern "C"
156 # else
157 # define PCRE_EXP_DECL extern
158 # endif
159 # ifndef PCRE_EXP_DEFN
160 # define PCRE_EXP_DEFN PCRE_EXP_DECL
161 # endif
162 # ifndef PCRE_EXP_DATA_DEFN
163 # define PCRE_EXP_DATA_DEFN
164 # endif
165 # endif
166 #endif
167
168 /* When compiling with the MSVC compiler, it is sometimes necessary to include
169 a "calling convention" before exported function names. (This is secondhand
170 information; I know nothing about MSVC myself). For example, something like
171
172 void __cdecl function(....)
173
174 might be needed. In order so make this easy, all the exported functions have
175 PCRE_CALL_CONVENTION just before their names. It is rarely needed; if not
176 set, we ensure here that it has no effect. */
177
178 #ifndef PCRE_CALL_CONVENTION
179 #define PCRE_CALL_CONVENTION
180 #endif
181
182 /* We need to have types that specify unsigned 8, 16 and 32-bit integers. We
183 cannot determine these outside the compilation (e.g. by running a program as
184 part of "configure") because PCRE is often cross-compiled for use on other
185 systems. Instead we make use of the maximum sizes that are available at
186 preprocessor time in standard C environments. */
187
188 typedef unsigned char pcre_uint8;
189
190 #if USHRT_MAX == 65535
191 typedef unsigned short pcre_uint16;
192 typedef short pcre_int16;
193 #elif UINT_MAX == 65535
194 typedef unsigned int pcre_uint16;
195 typedef int pcre_int16;
196 #else
197 #error Cannot determine a type for 16-bit unsigned integers
198 #endif
199
200 #if UINT_MAX == 4294967295
201 typedef unsigned int pcre_uint32;
202 typedef int pcre_int32;
203 #elif ULONG_MAX == 4294967295
204 typedef unsigned long int pcre_uint32;
205 typedef long int pcre_int32;
206 #else
207 #error Cannot determine a type for 32-bit unsigned integers
208 #endif
209
210 /* When checking for integer overflow in pcre_compile(), we need to handle
211 large integers. If a 64-bit integer type is available, we can use that.
212 Otherwise we have to cast to double, which of course requires floating point
213 arithmetic. Handle this by defining a macro for the appropriate type. If
214 stdint.h is available, include it; it may define INT64_MAX. Systems that do not
215 have stdint.h (e.g. Solaris) may have inttypes.h. The macro int64_t may be set
216 by "configure". */
217
218 #if HAVE_STDINT_H
219 #include <stdint.h>
220 #elif HAVE_INTTYPES_H
221 #include <inttypes.h>
222 #endif
223
224 #if defined INT64_MAX || defined int64_t
225 #define INT64_OR_DOUBLE int64_t
226 #else
227 #define INT64_OR_DOUBLE double
228 #endif
229
230 /* All character handling must be done as unsigned characters. Otherwise there
231 are problems with top-bit-set characters and functions such as isspace().
232 However, we leave the interface to the outside world as char * or short *,
233 because that should make things easier for callers. We define a short type
234 for the current character representation (either 8 or 16 bit) to save lots
235 of typing. I tried "uchar", but it causes problems on Digital Unix, where
236 it is defined in sys/types, so use "uschar" instead. */
237
238 #ifdef COMPILE_PCRE8
239
240 typedef unsigned char pcre_uchar;
241 #define IN_UCHARS(x) (x)
242
243 #else
244
245 #ifdef COMPILE_PCRE16
246 #if USHRT_MAX != 65535
247 /* This is a warning message. Change PCRE_SCHAR16 to a 16 bit data type in
248 pcre.h(.in) and disable (comment out) this message. */
249 #error Warning: PCRE_SCHAR16 is not a 16 bit data type.
250 #endif
251 typedef pcre_uint16 pcre_uchar;
252 #define IN_UCHARS(x) ((x) << 1)
253
254 #else
255 #error Unsupported compiling mode
256 #endif /* COMPILE_PCRE16 */
257
258 #endif /* COMPILE_PCRE8 */
259
260 /* This is an unsigned int value that no character can ever have. UTF-8
261 characters only go up to 0x7fffffff (though Unicode doesn't go beyond
262 0x0010ffff). */
263
264 #define NOTACHAR 0xffffffff
265
266 /* PCRE is able to support several different kinds of newline (CR, LF, CRLF,
267 "any" and "anycrlf" at present). The following macros are used to package up
268 testing for newlines. NLBLOCK, PSSTART, and PSEND are defined in the various
269 modules to indicate in which datablock the parameters exist, and what the
270 start/end of string field names are. */
271
272 #define NLTYPE_FIXED 0 /* Newline is a fixed length string */
273 #define NLTYPE_ANY 1 /* Newline is any Unicode line ending */
274 #define NLTYPE_ANYCRLF 2 /* Newline is CR, LF, or CRLF */
275
276 /* This macro checks for a newline at the given position */
277
278 #define IS_NEWLINE(p) \
279 ((NLBLOCK->nltype != NLTYPE_FIXED)? \
280 ((p) < NLBLOCK->PSEND && \
281 PRIV(is_newline)((p), NLBLOCK->nltype, NLBLOCK->PSEND, &(NLBLOCK->nllen),\
282 utf8)) \
283 : \
284 ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \
285 (p)[0] == NLBLOCK->nl[0] && \
286 (NLBLOCK->nllen == 1 || (p)[1] == NLBLOCK->nl[1]) \
287 ) \
288 )
289
290 /* This macro checks for a newline immediately preceding the given position */
291
292 #define WAS_NEWLINE(p) \
293 ((NLBLOCK->nltype != NLTYPE_FIXED)? \
294 ((p) > NLBLOCK->PSSTART && \
295 PRIV(was_newline)((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \
296 &(NLBLOCK->nllen), utf8)) \
297 : \
298 ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \
299 (p)[-NLBLOCK->nllen] == NLBLOCK->nl[0] && \
300 (NLBLOCK->nllen == 1 || (p)[-NLBLOCK->nllen+1] == NLBLOCK->nl[1]) \
301 ) \
302 )
303
304 /* When PCRE is compiled as a C++ library, the subject pointer can be replaced
305 with a custom type. This makes it possible, for example, to allow pcre_exec()
306 to process subject strings that are discontinuous by using a smart pointer
307 class. It must always be possible to inspect all of the subject string in
308 pcre_exec() because of the way it backtracks. Two macros are required in the
309 normal case, for sign-unspecified and unsigned char pointers. The former is
310 used for the external interface and appears in pcre.h, which is why its name
311 must begin with PCRE_. */
312
313 #ifdef CUSTOM_SUBJECT_PTR
314 #define PCRE_PUCHAR CUSTOM_SUBJECT_PTR
315 #else
316 #define PCRE_PUCHAR const pcre_uchar *
317 #endif
318
319 /* Include the public PCRE header and the definitions of UCP character property
320 values. */
321
322 #include "pcre.h"
323 #include "ucp.h"
324
325 /* When compiling for use with the Virtual Pascal compiler, these functions
326 need to have their names changed. PCRE must be compiled with the -DVPCOMPAT
327 option on the command line. */
328
329 #ifdef VPCOMPAT
330 #define strlen(s) _strlen(s)
331 #define strncmp(s1,s2,m) _strncmp(s1,s2,m)
332 #define memcmp(s,c,n) _memcmp(s,c,n)
333 #define memcpy(d,s,n) _memcpy(d,s,n)
334 #define memmove(d,s,n) _memmove(d,s,n)
335 #define memset(s,c,n) _memset(s,c,n)
336 #else /* VPCOMPAT */
337
338 /* To cope with SunOS4 and other systems that lack memmove() but have bcopy(),
339 define a macro for memmove() if HAVE_MEMMOVE is false, provided that HAVE_BCOPY
340 is set. Otherwise, include an emulating function for those systems that have
341 neither (there some non-Unix environments where this is the case). */
342
343 #ifndef HAVE_MEMMOVE
344 #undef memmove /* some systems may have a macro */
345 #ifdef HAVE_BCOPY
346 #define memmove(a, b, c) bcopy(b, a, c)
347 #else /* HAVE_BCOPY */
348 static void *
349 pcre_memmove(void *d, const void *s, size_t n)
350 {
351 size_t i;
352 unsigned char *dest = (unsigned char *)d;
353 const unsigned char *src = (const unsigned char *)s;
354 if (dest > src)
355 {
356 dest += n;
357 src += n;
358 for (i = 0; i < n; ++i) *(--dest) = *(--src);
359 return (void *)dest;
360 }
361 else
362 {
363 for (i = 0; i < n; ++i) *dest++ = *src++;
364 return (void *)(dest - n);
365 }
366 }
367 #define memmove(a, b, c) pcre_memmove(a, b, c)
368 #endif /* not HAVE_BCOPY */
369 #endif /* not HAVE_MEMMOVE */
370 #endif /* not VPCOMPAT */
371
372
373 /* PCRE keeps offsets in its compiled code as 2-byte quantities (always stored
374 in big-endian order) by default. These are used, for example, to link from the
375 start of a subpattern to its alternatives and its end. The use of 2 bytes per
376 offset limits the size of the compiled regex to around 64K, which is big enough
377 for almost everybody. However, I received a request for an even bigger limit.
378 For this reason, and also to make the code easier to maintain, the storing and
379 loading of offsets from the byte string is now handled by the macros that are
380 defined here.
381
382 The macros are controlled by the value of LINK_SIZE. This defaults to 2 in
383 the config.h file, but can be overridden by using -D on the command line. This
384 is automated on Unix systems via the "configure" command. */
385
386 #ifdef COMPILE_PCRE8
387
388 #if LINK_SIZE == 2
389
390 #define PUT(a,n,d) \
391 (a[n] = (d) >> 8), \
392 (a[(n)+1] = (d) & 255)
393
394 #define GET(a,n) \
395 (((a)[n] << 8) | (a)[(n)+1])
396
397 #define MAX_PATTERN_SIZE (1 << 16)
398
399
400 #elif LINK_SIZE == 3
401
402 #define PUT(a,n,d) \
403 (a[n] = (d) >> 16), \
404 (a[(n)+1] = (d) >> 8), \
405 (a[(n)+2] = (d) & 255)
406
407 #define GET(a,n) \
408 (((a)[n] << 16) | ((a)[(n)+1] << 8) | (a)[(n)+2])
409
410 #define MAX_PATTERN_SIZE (1 << 24)
411
412
413 #elif LINK_SIZE == 4
414
415 #define PUT(a,n,d) \
416 (a[n] = (d) >> 24), \
417 (a[(n)+1] = (d) >> 16), \
418 (a[(n)+2] = (d) >> 8), \
419 (a[(n)+3] = (d) & 255)
420
421 #define GET(a,n) \
422 (((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3])
423
424 /* Keep it positive */
425 #define MAX_PATTERN_SIZE (1 << 30)
426
427 #else
428 #error LINK_SIZE must be either 2, 3, or 4
429 #endif
430
431 #else /* COMPILE_PCRE8 */
432
433 #ifdef COMPILE_PCRE16
434
435 #if LINK_SIZE == 2
436
437 #undef LINK_SIZE
438 #define LINK_SIZE 1
439
440 #define PUT(a,n,d) \
441 (a[n] = (d))
442
443 #define GET(a,n) \
444 (a[n])
445
446 #define MAX_PATTERN_SIZE (1 << 16)
447
448 #elif LINK_SIZE == 3 || LINK_SIZE == 4
449
450 #undef LINK_SIZE
451 #define LINK_SIZE 2
452
453 #define PUT(a,n,d) \
454 (a[n] = (d) >> 16), \
455 (a[(n)+1] = (d) & 65536)
456
457 #define GET(a,n) \
458 (((a)[n] << 16) | (a)[(n)+1])
459
460 /* Keep it positive */
461 #define MAX_PATTERN_SIZE (1 << 30)
462
463 #else
464 #error LINK_SIZE must be either 2, 3, or 4
465 #endif
466
467 #else
468 #error Unsupported compiling mode
469 #endif /* COMPILE_PCRE16 */
470
471 #endif /* COMPILE_PCRE8 */
472
473 /* Convenience macro defined in terms of the others */
474
475 #define PUTINC(a,n,d) PUT(a,n,d), a += LINK_SIZE
476
477
478 /* PCRE uses some other 2-byte quantities that do not change when the size of
479 offsets changes. There are used for repeat counts and for other things such as
480 capturing parenthesis numbers in back references. */
481
482 #ifdef COMPILE_PCRE8
483
484 #define IMM2_SIZE 2
485
486 #define PUT2(a,n,d) \
487 a[n] = (d) >> 8; \
488 a[(n)+1] = (d) & 255
489
490 #define GET2(a,n) \
491 (((a)[n] << 8) | (a)[(n)+1])
492
493 #else /* COMPILE_PCRE8 */
494
495 #ifdef COMPILE_PCRE16
496
497 #define IMM2_SIZE 1
498
499 #define PUT2(a,n,d) \
500 a[n] = d
501
502 #define GET2(a,n) \
503 a[n]
504
505 #else
506 #error Unsupported compiling mode
507 #endif /* COMPILE_PCRE16 */
508
509 #endif /* COMPILE_PCRE8 */
510
511 #define PUT2INC(a,n,d) PUT2(a,n,d), a += IMM2_SIZE
512
513 /* When UTF-8 encoding is being used, a character is no longer just a single
514 byte. The macros for character handling generate simple sequences when used in
515 byte-mode, and more complicated ones for UTF-8 characters. GETCHARLENTEST is
516 not used when UTF-8 is not supported, so it is not defined, and BACKCHAR should
517 never be called in byte mode. To make sure they can never even appear when
518 UTF-8 support is omitted, we don't even define them. */
519
520 #ifndef SUPPORT_UTF8
521 #define GETCHAR(c, eptr) c = *eptr;
522 #define GETCHARTEST(c, eptr) c = *eptr;
523 #define GETCHARINC(c, eptr) c = *eptr++;
524 #define GETCHARINCTEST(c, eptr) c = *eptr++;
525 #define GETCHARLEN(c, eptr, len) c = *eptr;
526 /* #define GETCHARLENTEST(c, eptr, len) */
527 /* #define BACKCHAR(eptr) */
528
529 #else /* SUPPORT_UTF8 */
530
531 /* These macros were originally written in the form of loops that used data
532 from the tables whose names start with PRIV(utf8_table). They were rewritten by
533 a user so as not to use loops, because in some environments this gives a
534 significant performance advantage, and it seems never to do any harm. */
535
536 /* Base macro to pick up the remaining bytes of a UTF-8 character, not
537 advancing the pointer. */
538
539 #define GETUTF8(c, eptr) \
540 { \
541 if ((c & 0x20) == 0) \
542 c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
543 else if ((c & 0x10) == 0) \
544 c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
545 else if ((c & 0x08) == 0) \
546 c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
547 ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
548 else if ((c & 0x04) == 0) \
549 c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
550 ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
551 (eptr[4] & 0x3f); \
552 else \
553 c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
554 ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
555 ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
556 }
557
558 /* Get the next UTF-8 character, not advancing the pointer. This is called when
559 we know we are in UTF-8 mode. */
560
561 #define GETCHAR(c, eptr) \
562 c = *eptr; \
563 if (c >= 0xc0) GETUTF8(c, eptr);
564
565 /* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the
566 pointer. */
567
568 #define GETCHARTEST(c, eptr) \
569 c = *eptr; \
570 if (utf8 && c >= 0xc0) GETUTF8(c, eptr);
571
572 /* Base macro to pick up the remaining bytes of a UTF-8 character, advancing
573 the pointer. */
574
575 #define GETUTF8INC(c, eptr) \
576 { \
577 if ((c & 0x20) == 0) \
578 c = ((c & 0x1f) << 6) | (*eptr++ & 0x3f); \
579 else if ((c & 0x10) == 0) \
580 { \
581 c = ((c & 0x0f) << 12) | ((*eptr & 0x3f) << 6) | (eptr[1] & 0x3f); \
582 eptr += 2; \
583 } \
584 else if ((c & 0x08) == 0) \
585 { \
586 c = ((c & 0x07) << 18) | ((*eptr & 0x3f) << 12) | \
587 ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
588 eptr += 3; \
589 } \
590 else if ((c & 0x04) == 0) \
591 { \
592 c = ((c & 0x03) << 24) | ((*eptr & 0x3f) << 18) | \
593 ((eptr[1] & 0x3f) << 12) | ((eptr[2] & 0x3f) << 6) | \
594 (eptr[3] & 0x3f); \
595 eptr += 4; \
596 } \
597 else \
598 { \
599 c = ((c & 0x01) << 30) | ((*eptr & 0x3f) << 24) | \
600 ((eptr[1] & 0x3f) << 18) | ((eptr[2] & 0x3f) << 12) | \
601 ((eptr[3] & 0x3f) << 6) | (eptr[4] & 0x3f); \
602 eptr += 5; \
603 } \
604 }
605
606 /* Get the next UTF-8 character, advancing the pointer. This is called when we
607 know we are in UTF-8 mode. */
608
609 #define GETCHARINC(c, eptr) \
610 c = *eptr++; \
611 if (c >= 0xc0) GETUTF8INC(c, eptr);
612
613 /* Get the next character, testing for UTF-8 mode, and advancing the pointer.
614 This is called when we don't know if we are in UTF-8 mode. */
615
616 #define GETCHARINCTEST(c, eptr) \
617 c = *eptr++; \
618 if (utf8 && c >= 0xc0) GETUTF8INC(c, eptr);
619
620 /* Base macro to pick up the remaining bytes of a UTF-8 character, not
621 advancing the pointer, incrementing the length. */
622
623 #define GETUTF8LEN(c, eptr, len) \
624 { \
625 if ((c & 0x20) == 0) \
626 { \
627 c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
628 len++; \
629 } \
630 else if ((c & 0x10) == 0) \
631 { \
632 c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
633 len += 2; \
634 } \
635 else if ((c & 0x08) == 0) \
636 {\
637 c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
638 ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
639 len += 3; \
640 } \
641 else if ((c & 0x04) == 0) \
642 { \
643 c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
644 ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
645 (eptr[4] & 0x3f); \
646 len += 4; \
647 } \
648 else \
649 {\
650 c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
651 ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
652 ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
653 len += 5; \
654 } \
655 }
656
657 /* Get the next UTF-8 character, not advancing the pointer, incrementing length
658 if there are extra bytes. This is called when we know we are in UTF-8 mode. */
659
660 #define GETCHARLEN(c, eptr, len) \
661 c = *eptr; \
662 if (c >= 0xc0) GETUTF8LEN(c, eptr, len);
663
664 /* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the
665 pointer, incrementing length if there are extra bytes. This is called when we
666 do not know if we are in UTF-8 mode. */
667
668 #define GETCHARLENTEST(c, eptr, len) \
669 c = *eptr; \
670 if (utf8 && c >= 0xc0) GETUTF8LEN(c, eptr, len);
671
672 /* If the pointer is not at the start of a character, move it back until
673 it is. This is called only in UTF-8 mode - we don't put a test within the macro
674 because almost all calls are already within a block of UTF-8 only code. */
675
676 #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--
677
678 #endif /* SUPPORT_UTF8 */
679
680
681 /* In case there is no definition of offsetof() provided - though any proper
682 Standard C system should have one. */
683
684 #ifndef offsetof
685 #define offsetof(p_type,field) ((size_t)&(((p_type *)0)->field))
686 #endif
687
688
689 /* Private flags containing information about the compiled regex. They used to
690 live at the top end of the options word, but that got almost full, so now they
691 are in a 16-bit flags word. From release 8.00, PCRE_NOPARTIAL is unused, as
692 the restrictions on partial matching have been lifted. It remains for backwards
693 compatibility. */
694
695 #define PCRE_NOPARTIAL 0x0001 /* can't use partial with this regex */
696 #define PCRE_FIRSTSET 0x0002 /* first_byte is set */
697 #define PCRE_REQCHSET 0x0004 /* req_byte is set */
698 #define PCRE_STARTLINE 0x0008 /* start after \n for multiline */
699 #define PCRE_JCHANGED 0x0010 /* j option used in regex */
700 #define PCRE_HASCRORLF 0x0020 /* explicit \r or \n in pattern */
701 #define PCRE_HASTHEN 0x0040 /* pattern contains (*THEN) */
702
703 /* Flags for the "extra" block produced by pcre_study(). */
704
705 #define PCRE_STUDY_MAPPED 0x0001 /* a map of starting chars exists */
706 #define PCRE_STUDY_MINLEN 0x0002 /* a minimum length field exists */
707
708 /* Masks for identifying the public options that are permitted at compile
709 time, run time, or study time, respectively. */
710
711 #define PCRE_NEWLINE_BITS (PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_ANY| \
712 PCRE_NEWLINE_ANYCRLF)
713
714 #define PUBLIC_COMPILE_OPTIONS \
715 (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \
716 PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \
717 PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \
718 PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
719 PCRE_JAVASCRIPT_COMPAT|PCRE_UCP|PCRE_NO_START_OPTIMIZE)
720
721 #define PUBLIC_EXEC_OPTIONS \
722 (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
723 PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_NEWLINE_BITS| \
724 PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE|PCRE_NO_START_OPTIMIZE)
725
726 #define PUBLIC_DFA_EXEC_OPTIONS \
727 (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
728 PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_DFA_SHORTEST| \
729 PCRE_DFA_RESTART|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
730 PCRE_NO_START_OPTIMIZE)
731
732 #define PUBLIC_STUDY_OPTIONS \
733 PCRE_STUDY_JIT_COMPILE
734
735 /* Magic number to provide a small check against being handed junk. Also used
736 to detect whether a pattern was compiled on a host of different endianness. */
737
738 #define MAGIC_NUMBER 0x50435245UL /* 'PCRE' */
739
740 /* Negative values for the firstchar and reqchar variables */
741
742 #define REQ_UNSET (-2)
743 #define REQ_NONE (-1)
744
745 /* The maximum remaining length of subject we are prepared to search for a
746 req_byte match. */
747
748 #define REQ_BYTE_MAX 1000
749
750 /* Flags added to firstbyte or reqbyte; a "non-literal" item is either a
751 variable-length repeat, or a anything other than literal characters. */
752
753 #define REQ_CASELESS 0x0100 /* indicates caselessness */
754 #define REQ_VARY 0x0200 /* reqbyte followed non-literal item */
755
756 /* Miscellaneous definitions. The #ifndef is to pacify compiler warnings in
757 environments where these macros are defined elsewhere. Unfortunately, there
758 is no way to do the same for the typedef. */
759
760 typedef int BOOL;
761
762 #ifndef FALSE
763 #define FALSE 0
764 #define TRUE 1
765 #endif
766
767 /* If PCRE is to support UTF-8 on EBCDIC platforms, we cannot use normal
768 character constants like '*' because the compiler would emit their EBCDIC code,
769 which is different from their ASCII/UTF-8 code. Instead we define macros for
770 the characters so that they always use the ASCII/UTF-8 code when UTF-8 support
771 is enabled. When UTF-8 support is not enabled, the definitions use character
772 literals. Both character and string versions of each character are needed, and
773 there are some longer strings as well.
774
775 This means that, on EBCDIC platforms, the PCRE library can handle either
776 EBCDIC, or UTF-8, but not both. To support both in the same compiled library
777 would need different lookups depending on whether PCRE_UTF8 was set or not.
778 This would make it impossible to use characters in switch/case statements,
779 which would reduce performance. For a theoretical use (which nobody has asked
780 for) in a minority area (EBCDIC platforms), this is not sensible. Any
781 application that did need both could compile two versions of the library, using
782 macros to give the functions distinct names. */
783
784 #ifndef SUPPORT_UTF8
785
786 /* UTF-8 support is not enabled; use the platform-dependent character literals
787 so that PCRE works on both ASCII and EBCDIC platforms, in non-UTF-mode only. */
788
789 #define CHAR_HT '\t'
790 #define CHAR_VT '\v'
791 #define CHAR_FF '\f'
792 #define CHAR_CR '\r'
793 #define CHAR_NL '\n'
794 #define CHAR_BS '\b'
795 #define CHAR_BEL '\a'
796 #ifdef EBCDIC
797 #define CHAR_ESC '\047'
798 #define CHAR_DEL '\007'
799 #else
800 #define CHAR_ESC '\033'
801 #define CHAR_DEL '\177'
802 #endif
803
804 #define CHAR_SPACE ' '
805 #define CHAR_EXCLAMATION_MARK '!'
806 #define CHAR_QUOTATION_MARK '"'
807 #define CHAR_NUMBER_SIGN '#'
808 #define CHAR_DOLLAR_SIGN '$'
809 #define CHAR_PERCENT_SIGN '%'
810 #define CHAR_AMPERSAND '&'
811 #define CHAR_APOSTROPHE '\''
812 #define CHAR_LEFT_PARENTHESIS '('
813 #define CHAR_RIGHT_PARENTHESIS ')'
814 #define CHAR_ASTERISK '*'
815 #define CHAR_PLUS '+'
816 #define CHAR_COMMA ','
817 #define CHAR_MINUS '-'
818 #define CHAR_DOT '.'
819 #define CHAR_SLASH '/'
820 #define CHAR_0 '0'
821 #define CHAR_1 '1'
822 #define CHAR_2 '2'
823 #define CHAR_3 '3'
824 #define CHAR_4 '4'
825 #define CHAR_5 '5'
826 #define CHAR_6 '6'
827 #define CHAR_7 '7'
828 #define CHAR_8 '8'
829 #define CHAR_9 '9'
830 #define CHAR_COLON ':'
831 #define CHAR_SEMICOLON ';'
832 #define CHAR_LESS_THAN_SIGN '<'
833 #define CHAR_EQUALS_SIGN '='
834 #define CHAR_GREATER_THAN_SIGN '>'
835 #define CHAR_QUESTION_MARK '?'
836 #define CHAR_COMMERCIAL_AT '@'
837 #define CHAR_A 'A'
838 #define CHAR_B 'B'
839 #define CHAR_C 'C'
840 #define CHAR_D 'D'
841 #define CHAR_E 'E'
842 #define CHAR_F 'F'
843 #define CHAR_G 'G'
844 #define CHAR_H 'H'
845 #define CHAR_I 'I'
846 #define CHAR_J 'J'
847 #define CHAR_K 'K'
848 #define CHAR_L 'L'
849 #define CHAR_M 'M'
850 #define CHAR_N 'N'
851 #define CHAR_O 'O'
852 #define CHAR_P 'P'
853 #define CHAR_Q 'Q'
854 #define CHAR_R 'R'
855 #define CHAR_S 'S'
856 #define CHAR_T 'T'
857 #define CHAR_U 'U'
858 #define CHAR_V 'V'
859 #define CHAR_W 'W'
860 #define CHAR_X 'X'
861 #define CHAR_Y 'Y'
862 #define CHAR_Z 'Z'
863 #define CHAR_LEFT_SQUARE_BRACKET '['
864 #define CHAR_BACKSLASH '\\'
865 #define CHAR_RIGHT_SQUARE_BRACKET ']'
866 #define CHAR_CIRCUMFLEX_ACCENT '^'
867 #define CHAR_UNDERSCORE '_'
868 #define CHAR_GRAVE_ACCENT '`'
869 #define CHAR_a 'a'
870 #define CHAR_b 'b'
871 #define CHAR_c 'c'
872 #define CHAR_d 'd'
873 #define CHAR_e 'e'
874 #define CHAR_f 'f'
875 #define CHAR_g 'g'
876 #define CHAR_h 'h'
877 #define CHAR_i 'i'
878 #define CHAR_j 'j'
879 #define CHAR_k 'k'
880 #define CHAR_l 'l'
881 #define CHAR_m 'm'
882 #define CHAR_n 'n'
883 #define CHAR_o 'o'
884 #define CHAR_p 'p'
885 #define CHAR_q 'q'
886 #define CHAR_r 'r'
887 #define CHAR_s 's'
888 #define CHAR_t 't'
889 #define CHAR_u 'u'
890 #define CHAR_v 'v'
891 #define CHAR_w 'w'
892 #define CHAR_x 'x'
893 #define CHAR_y 'y'
894 #define CHAR_z 'z'
895 #define CHAR_LEFT_CURLY_BRACKET '{'
896 #define CHAR_VERTICAL_LINE '|'
897 #define CHAR_RIGHT_CURLY_BRACKET '}'
898 #define CHAR_TILDE '~'
899
900 #define STR_HT "\t"
901 #define STR_VT "\v"
902 #define STR_FF "\f"
903 #define STR_CR "\r"
904 #define STR_NL "\n"
905 #define STR_BS "\b"
906 #define STR_BEL "\a"
907 #ifdef EBCDIC
908 #define STR_ESC "\047"
909 #define STR_DEL "\007"
910 #else
911 #define STR_ESC "\033"
912 #define STR_DEL "\177"
913 #endif
914
915 #define STR_SPACE " "
916 #define STR_EXCLAMATION_MARK "!"
917 #define STR_QUOTATION_MARK "\""
918 #define STR_NUMBER_SIGN "#"
919 #define STR_DOLLAR_SIGN "$"
920 #define STR_PERCENT_SIGN "%"
921 #define STR_AMPERSAND "&"
922 #define STR_APOSTROPHE "'"
923 #define STR_LEFT_PARENTHESIS "("
924 #define STR_RIGHT_PARENTHESIS ")"
925 #define STR_ASTERISK "*"
926 #define STR_PLUS "+"
927 #define STR_COMMA ","
928 #define STR_MINUS "-"
929 #define STR_DOT "."
930 #define STR_SLASH "/"
931 #define STR_0 "0"
932 #define STR_1 "1"
933 #define STR_2 "2"
934 #define STR_3 "3"
935 #define STR_4 "4"
936 #define STR_5 "5"
937 #define STR_6 "6"
938 #define STR_7 "7"
939 #define STR_8 "8"
940 #define STR_9 "9"
941 #define STR_COLON ":"
942 #define STR_SEMICOLON ";"
943 #define STR_LESS_THAN_SIGN "<"
944 #define STR_EQUALS_SIGN "="
945 #define STR_GREATER_THAN_SIGN ">"
946 #define STR_QUESTION_MARK "?"
947 #define STR_COMMERCIAL_AT "@"
948 #define STR_A "A"
949 #define STR_B "B"
950 #define STR_C "C"
951 #define STR_D "D"
952 #define STR_E "E"
953 #define STR_F "F"
954 #define STR_G "G"
955 #define STR_H "H"
956 #define STR_I "I"
957 #define STR_J "J"
958 #define STR_K "K"
959 #define STR_L "L"
960 #define STR_M "M"
961 #define STR_N "N"
962 #define STR_O "O"
963 #define STR_P "P"
964 #define STR_Q "Q"
965 #define STR_R "R"
966 #define STR_S "S"
967 #define STR_T "T"
968 #define STR_U "U"
969 #define STR_V "V"
970 #define STR_W "W"
971 #define STR_X "X"
972 #define STR_Y "Y"
973 #define STR_Z "Z"
974 #define STR_LEFT_SQUARE_BRACKET "["
975 #define STR_BACKSLASH "\\"
976 #define STR_RIGHT_SQUARE_BRACKET "]"
977 #define STR_CIRCUMFLEX_ACCENT "^"
978 #define STR_UNDERSCORE "_"
979 #define STR_GRAVE_ACCENT "`"
980 #define STR_a "a"
981 #define STR_b "b"
982 #define STR_c "c"
983 #define STR_d "d"
984 #define STR_e "e"
985 #define STR_f "f"
986 #define STR_g "g"
987 #define STR_h "h"
988 #define STR_i "i"
989 #define STR_j "j"
990 #define STR_k "k"
991 #define STR_l "l"
992 #define STR_m "m"
993 #define STR_n "n"
994 #define STR_o "o"
995 #define STR_p "p"
996 #define STR_q "q"
997 #define STR_r "r"
998 #define STR_s "s"
999 #define STR_t "t"
1000 #define STR_u "u"
1001 #define STR_v "v"
1002 #define STR_w "w"
1003 #define STR_x "x"
1004 #define STR_y "y"
1005 #define STR_z "z"
1006 #define STR_LEFT_CURLY_BRACKET "{"
1007 #define STR_VERTICAL_LINE "|"
1008 #define STR_RIGHT_CURLY_BRACKET "}"
1009 #define STR_TILDE "~"
1010
1011 #define STRING_ACCEPT0 "ACCEPT\0"
1012 #define STRING_COMMIT0 "COMMIT\0"
1013 #define STRING_F0 "F\0"
1014 #define STRING_FAIL0 "FAIL\0"
1015 #define STRING_MARK0 "MARK\0"
1016 #define STRING_PRUNE0 "PRUNE\0"
1017 #define STRING_SKIP0 "SKIP\0"
1018 #define STRING_THEN "THEN"
1019
1020 #define STRING_alpha0 "alpha\0"
1021 #define STRING_lower0 "lower\0"
1022 #define STRING_upper0 "upper\0"
1023 #define STRING_alnum0 "alnum\0"
1024 #define STRING_ascii0 "ascii\0"
1025 #define STRING_blank0 "blank\0"
1026 #define STRING_cntrl0 "cntrl\0"
1027 #define STRING_digit0 "digit\0"
1028 #define STRING_graph0 "graph\0"
1029 #define STRING_print0 "print\0"
1030 #define STRING_punct0 "punct\0"
1031 #define STRING_space0 "space\0"
1032 #define STRING_word0 "word\0"
1033 #define STRING_xdigit "xdigit"
1034
1035 #define STRING_DEFINE "DEFINE"
1036
1037 #define STRING_CR_RIGHTPAR "CR)"
1038 #define STRING_LF_RIGHTPAR "LF)"
1039 #define STRING_CRLF_RIGHTPAR "CRLF)"
1040 #define STRING_ANY_RIGHTPAR "ANY)"
1041 #define STRING_ANYCRLF_RIGHTPAR "ANYCRLF)"
1042 #define STRING_BSR_ANYCRLF_RIGHTPAR "BSR_ANYCRLF)"
1043 #define STRING_BSR_UNICODE_RIGHTPAR "BSR_UNICODE)"
1044 #define STRING_UTF8_RIGHTPAR "UTF8)"
1045 #define STRING_UCP_RIGHTPAR "UCP)"
1046 #define STRING_NO_START_OPT_RIGHTPAR "NO_START_OPT)"
1047
1048 #else /* SUPPORT_UTF8 */
1049
1050 /* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This
1051 works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode
1052 only. */
1053
1054 #define CHAR_HT '\011'
1055 #define CHAR_VT '\013'
1056 #define CHAR_FF '\014'
1057 #define CHAR_CR '\015'
1058 #define CHAR_NL '\012'
1059 #define CHAR_BS '\010'
1060 #define CHAR_BEL '\007'
1061 #define CHAR_ESC '\033'
1062 #define CHAR_DEL '\177'
1063
1064 #define CHAR_SPACE '\040'
1065 #define CHAR_EXCLAMATION_MARK '\041'
1066 #define CHAR_QUOTATION_MARK '\042'
1067 #define CHAR_NUMBER_SIGN '\043'
1068 #define CHAR_DOLLAR_SIGN '\044'
1069 #define CHAR_PERCENT_SIGN '\045'
1070 #define CHAR_AMPERSAND '\046'
1071 #define CHAR_APOSTROPHE '\047'
1072 #define CHAR_LEFT_PARENTHESIS '\050'
1073 #define CHAR_RIGHT_PARENTHESIS '\051'
1074 #define CHAR_ASTERISK '\052'
1075 #define CHAR_PLUS '\053'
1076 #define CHAR_COMMA '\054'
1077 #define CHAR_MINUS '\055'
1078 #define CHAR_DOT '\056'
1079 #define CHAR_SLASH '\057'
1080 #define CHAR_0 '\060'
1081 #define CHAR_1 '\061'
1082 #define CHAR_2 '\062'
1083 #define CHAR_3 '\063'
1084 #define CHAR_4 '\064'
1085 #define CHAR_5 '\065'
1086 #define CHAR_6 '\066'
1087 #define CHAR_7 '\067'
1088 #define CHAR_8 '\070'
1089 #define CHAR_9 '\071'
1090 #define CHAR_COLON '\072'
1091 #define CHAR_SEMICOLON '\073'
1092 #define CHAR_LESS_THAN_SIGN '\074'
1093 #define CHAR_EQUALS_SIGN '\075'
1094 #define CHAR_GREATER_THAN_SIGN '\076'
1095 #define CHAR_QUESTION_MARK '\077'
1096 #define CHAR_COMMERCIAL_AT '\100'
1097 #define CHAR_A '\101'
1098 #define CHAR_B '\102'
1099 #define CHAR_C '\103'
1100 #define CHAR_D '\104'
1101 #define CHAR_E '\105'
1102 #define CHAR_F '\106'
1103 #define CHAR_G '\107'
1104 #define CHAR_H '\110'
1105 #define CHAR_I '\111'
1106 #define CHAR_J '\112'
1107 #define CHAR_K '\113'
1108 #define CHAR_L '\114'
1109 #define CHAR_M '\115'
1110 #define CHAR_N '\116'
1111 #define CHAR_O '\117'
1112 #define CHAR_P '\120'
1113 #define CHAR_Q '\121'
1114 #define CHAR_R '\122'
1115 #define CHAR_S '\123'
1116 #define CHAR_T '\124'
1117 #define CHAR_U '\125'
1118 #define CHAR_V '\126'
1119 #define CHAR_W '\127'
1120 #define CHAR_X '\130'
1121 #define CHAR_Y '\131'
1122 #define CHAR_Z '\132'
1123 #define CHAR_LEFT_SQUARE_BRACKET '\133'
1124 #define CHAR_BACKSLASH '\134'
1125 #define CHAR_RIGHT_SQUARE_BRACKET '\135'
1126 #define CHAR_CIRCUMFLEX_ACCENT '\136'
1127 #define CHAR_UNDERSCORE '\137'
1128 #define CHAR_GRAVE_ACCENT '\140'
1129 #define CHAR_a '\141'
1130 #define CHAR_b '\142'
1131 #define CHAR_c '\143'
1132 #define CHAR_d '\144'
1133 #define CHAR_e '\145'
1134 #define CHAR_f '\146'
1135 #define CHAR_g '\147'
1136 #define CHAR_h '\150'
1137 #define CHAR_i '\151'
1138 #define CHAR_j '\152'
1139 #define CHAR_k '\153'
1140 #define CHAR_l '\154'
1141 #define CHAR_m '\155'
1142 #define CHAR_n '\156'
1143 #define CHAR_o '\157'
1144 #define CHAR_p '\160'
1145 #define CHAR_q '\161'
1146 #define CHAR_r '\162'
1147 #define CHAR_s '\163'
1148 #define CHAR_t '\164'
1149 #define CHAR_u '\165'
1150 #define CHAR_v '\166'
1151 #define CHAR_w '\167'
1152 #define CHAR_x '\170'
1153 #define CHAR_y '\171'
1154 #define CHAR_z '\172'
1155 #define CHAR_LEFT_CURLY_BRACKET '\173'
1156 #define CHAR_VERTICAL_LINE '\174'
1157 #define CHAR_RIGHT_CURLY_BRACKET '\175'
1158 #define CHAR_TILDE '\176'
1159
1160 #define STR_HT "\011"
1161 #define STR_VT "\013"
1162 #define STR_FF "\014"
1163 #define STR_CR "\015"
1164 #define STR_NL "\012"
1165 #define STR_BS "\010"
1166 #define STR_BEL "\007"
1167 #define STR_ESC "\033"
1168 #define STR_DEL "\177"
1169
1170 #define STR_SPACE "\040"
1171 #define STR_EXCLAMATION_MARK "\041"
1172 #define STR_QUOTATION_MARK "\042"
1173 #define STR_NUMBER_SIGN "\043"
1174 #define STR_DOLLAR_SIGN "\044"
1175 #define STR_PERCENT_SIGN "\045"
1176 #define STR_AMPERSAND "\046"
1177 #define STR_APOSTROPHE "\047"
1178 #define STR_LEFT_PARENTHESIS "\050"
1179 #define STR_RIGHT_PARENTHESIS "\051"
1180 #define STR_ASTERISK "\052"
1181 #define STR_PLUS "\053"
1182 #define STR_COMMA "\054"
1183 #define STR_MINUS "\055"
1184 #define STR_DOT "\056"
1185 #define STR_SLASH "\057"
1186 #define STR_0 "\060"
1187 #define STR_1 "\061"
1188 #define STR_2 "\062"
1189 #define STR_3 "\063"
1190 #define STR_4 "\064"
1191 #define STR_5 "\065"
1192 #define STR_6 "\066"
1193 #define STR_7 "\067"
1194 #define STR_8 "\070"
1195 #define STR_9 "\071"
1196 #define STR_COLON "\072"
1197 #define STR_SEMICOLON "\073"
1198 #define STR_LESS_THAN_SIGN "\074"
1199 #define STR_EQUALS_SIGN "\075"
1200 #define STR_GREATER_THAN_SIGN "\076"
1201 #define STR_QUESTION_MARK "\077"
1202 #define STR_COMMERCIAL_AT "\100"
1203 #define STR_A "\101"
1204 #define STR_B "\102"
1205 #define STR_C "\103"
1206 #define STR_D "\104"
1207 #define STR_E "\105"
1208 #define STR_F "\106"
1209 #define STR_G "\107"
1210 #define STR_H "\110"
1211 #define STR_I "\111"
1212 #define STR_J "\112"
1213 #define STR_K "\113"
1214 #define STR_L "\114"
1215 #define STR_M "\115"
1216 #define STR_N "\116"
1217 #define STR_O "\117"
1218 #define STR_P "\120"
1219 #define STR_Q "\121"
1220 #define STR_R "\122"
1221 #define STR_S "\123"
1222 #define STR_T "\124"
1223 #define STR_U "\125"
1224 #define STR_V "\126"
1225 #define STR_W "\127"
1226 #define STR_X "\130"
1227 #define STR_Y "\131"
1228 #define STR_Z "\132"
1229 #define STR_LEFT_SQUARE_BRACKET "\133"
1230 #define STR_BACKSLASH "\134"
1231 #define STR_RIGHT_SQUARE_BRACKET "\135"
1232 #define STR_CIRCUMFLEX_ACCENT "\136"
1233 #define STR_UNDERSCORE "\137"
1234 #define STR_GRAVE_ACCENT "\140"
1235 #define STR_a "\141"
1236 #define STR_b "\142"
1237 #define STR_c "\143"
1238 #define STR_d "\144"
1239 #define STR_e "\145"
1240 #define STR_f "\146"
1241 #define STR_g "\147"
1242 #define STR_h "\150"
1243 #define STR_i "\151"
1244 #define STR_j "\152"
1245 #define STR_k "\153"
1246 #define STR_l "\154"
1247 #define STR_m "\155"
1248 #define STR_n "\156"
1249 #define STR_o "\157"
1250 #define STR_p "\160"
1251 #define STR_q "\161"
1252 #define STR_r "\162"
1253 #define STR_s "\163"
1254 #define STR_t "\164"
1255 #define STR_u "\165"
1256 #define STR_v "\166"
1257 #define STR_w "\167"
1258 #define STR_x "\170"
1259 #define STR_y "\171"
1260 #define STR_z "\172"
1261 #define STR_LEFT_CURLY_BRACKET "\173"
1262 #define STR_VERTICAL_LINE "\174"
1263 #define STR_RIGHT_CURLY_BRACKET "\175"
1264 #define STR_TILDE "\176"
1265
1266 #define STRING_ACCEPT0 STR_A STR_C STR_C STR_E STR_P STR_T "\0"
1267 #define STRING_COMMIT0 STR_C STR_O STR_M STR_M STR_I STR_T "\0"
1268 #define STRING_F0 STR_F "\0"
1269 #define STRING_FAIL0 STR_F STR_A STR_I STR_L "\0"
1270 #define STRING_MARK0 STR_M STR_A STR_R STR_K "\0"
1271 #define STRING_PRUNE0 STR_P STR_R STR_U STR_N STR_E "\0"
1272 #define STRING_SKIP0 STR_S STR_K STR_I STR_P "\0"
1273 #define STRING_THEN STR_T STR_H STR_E STR_N
1274
1275 #define STRING_alpha0 STR_a STR_l STR_p STR_h STR_a "\0"
1276 #define STRING_lower0 STR_l STR_o STR_w STR_e STR_r "\0"
1277 #define STRING_upper0 STR_u STR_p STR_p STR_e STR_r "\0"
1278 #define STRING_alnum0 STR_a STR_l STR_n STR_u STR_m "\0"
1279 #define STRING_ascii0 STR_a STR_s STR_c STR_i STR_i "\0"
1280 #define STRING_blank0 STR_b STR_l STR_a STR_n STR_k "\0"
1281 #define STRING_cntrl0 STR_c STR_n STR_t STR_r STR_l "\0"
1282 #define STRING_digit0 STR_d STR_i STR_g STR_i STR_t "\0"
1283 #define STRING_graph0 STR_g STR_r STR_a STR_p STR_h "\0"
1284 #define STRING_print0 STR_p STR_r STR_i STR_n STR_t "\0"
1285 #define STRING_punct0 STR_p STR_u STR_n STR_c STR_t "\0"
1286 #define STRING_space0 STR_s STR_p STR_a STR_c STR_e "\0"
1287 #define STRING_word0 STR_w STR_o STR_r STR_d "\0"
1288 #define STRING_xdigit STR_x STR_d STR_i STR_g STR_i STR_t
1289
1290 #define STRING_DEFINE STR_D STR_E STR_F STR_I STR_N STR_E
1291
1292 #define STRING_CR_RIGHTPAR STR_C STR_R STR_RIGHT_PARENTHESIS
1293 #define STRING_LF_RIGHTPAR STR_L STR_F STR_RIGHT_PARENTHESIS
1294 #define STRING_CRLF_RIGHTPAR STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1295 #define STRING_ANY_RIGHTPAR STR_A STR_N STR_Y STR_RIGHT_PARENTHESIS
1296 #define STRING_ANYCRLF_RIGHTPAR STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1297 #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
1298 #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
1299 #define STRING_UTF8_RIGHTPAR STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS
1300 #define STRING_UCP_RIGHTPAR STR_U STR_C STR_P STR_RIGHT_PARENTHESIS
1301 #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
1302
1303 #endif /* SUPPORT_UTF8 */
1304
1305 /* Escape items that are just an encoding of a particular data value. */
1306
1307 #ifndef ESC_e
1308 #define ESC_e CHAR_ESC
1309 #endif
1310
1311 #ifndef ESC_f
1312 #define ESC_f CHAR_FF
1313 #endif
1314
1315 #ifndef ESC_n
1316 #define ESC_n CHAR_NL
1317 #endif
1318
1319 #ifndef ESC_r
1320 #define ESC_r CHAR_CR
1321 #endif
1322
1323 /* We can't officially use ESC_t because it is a POSIX reserved identifier
1324 (presumably because of all the others like size_t). */
1325
1326 #ifndef ESC_tee
1327 #define ESC_tee CHAR_HT
1328 #endif
1329
1330 /* Codes for different types of Unicode property */
1331
1332 #define PT_ANY 0 /* Any property - matches all chars */
1333 #define PT_LAMP 1 /* L& - the union of Lu, Ll, Lt */
1334 #define PT_GC 2 /* Specified general characteristic (e.g. L) */
1335 #define PT_PC 3 /* Specified particular characteristic (e.g. Lu) */
1336 #define PT_SC 4 /* Script (e.g. Han) */
1337 #define PT_ALNUM 5 /* Alphanumeric - the union of L and N */
1338 #define PT_SPACE 6 /* Perl space - Z plus 9,10,12,13 */
1339 #define PT_PXSPACE 7 /* POSIX space - Z plus 9,10,11,12,13 */
1340 #define PT_WORD 8 /* Word - L plus N plus underscore */
1341
1342 /* Flag bits and data types for the extended class (OP_XCLASS) for classes that
1343 contain characters with values greater than 255. */
1344
1345 #define XCL_NOT 0x01 /* Flag: this is a negative class */
1346 #define XCL_MAP 0x02 /* Flag: a 32-byte map is present */
1347
1348 #define XCL_END 0 /* Marks end of individual items */
1349 #define XCL_SINGLE 1 /* Single item (one multibyte char) follows */
1350 #define XCL_RANGE 2 /* A range (two multibyte chars) follows */
1351 #define XCL_PROP 3 /* Unicode property (2-byte property code follows) */
1352 #define XCL_NOTPROP 4 /* Unicode inverted property (ditto) */
1353
1354 /* These are escaped items that aren't just an encoding of a particular data
1355 value such as \n. They must have non-zero values, as check_escape() returns
1356 their negation. Also, they must appear in the same order as in the opcode
1357 definitions below, up to ESC_z. There's a dummy for OP_ALLANY because it
1358 corresponds to "." in DOTALL mode rather than an escape sequence. It is also
1359 used for [^] in JavaScript compatibility mode, and for \C in non-utf8 mode. In
1360 non-DOTALL mode, "." behaves like \N.
1361
1362 The special values ESC_DU, ESC_du, etc. are used instead of ESC_D, ESC_d, etc.
1363 when PCRE_UCP is set, when replacement of \d etc by \p sequences is required.
1364 They must be contiguous, and remain in order so that the replacements can be
1365 looked up from a table.
1366
1367 The final escape must be ESC_REF as subsequent values are used for
1368 backreferences (\1, \2, \3, etc). There are two tests in the code for an escape
1369 greater than ESC_b and less than ESC_Z to detect the types that may be
1370 repeated. These are the types that consume characters. If any new escapes are
1371 put in between that don't consume a character, that code will have to change.
1372 */
1373
1374 enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s,
1375 ESC_W, ESC_w, ESC_N, ESC_dum, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H,
1376 ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z,
1377 ESC_E, ESC_Q, ESC_g, ESC_k,
1378 ESC_DU, ESC_du, ESC_SU, ESC_su, ESC_WU, ESC_wu,
1379 ESC_REF };
1380
1381 /* Opcode table: Starting from 1 (i.e. after OP_END), the values up to
1382 OP_EOD must correspond in order to the list of escapes immediately above.
1383
1384 *** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions
1385 that follow must also be updated to match. There are also tables called
1386 "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
1387
1388 enum {
1389 OP_END, /* 0 End of pattern */
1390
1391 /* Values corresponding to backslashed metacharacters */
1392
1393 OP_SOD, /* 1 Start of data: \A */
1394 OP_SOM, /* 2 Start of match (subject + offset): \G */
1395 OP_SET_SOM, /* 3 Set start of match (\K) */
1396 OP_NOT_WORD_BOUNDARY, /* 4 \B */
1397 OP_WORD_BOUNDARY, /* 5 \b */
1398 OP_NOT_DIGIT, /* 6 \D */
1399 OP_DIGIT, /* 7 \d */
1400 OP_NOT_WHITESPACE, /* 8 \S */
1401 OP_WHITESPACE, /* 9 \s */
1402 OP_NOT_WORDCHAR, /* 10 \W */
1403 OP_WORDCHAR, /* 11 \w */
1404
1405 OP_ANY, /* 12 Match any character except newline */
1406 OP_ALLANY, /* 13 Match any character */
1407 OP_ANYBYTE, /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */
1408 OP_NOTPROP, /* 15 \P (not Unicode property) */
1409 OP_PROP, /* 16 \p (Unicode property) */
1410 OP_ANYNL, /* 17 \R (any newline sequence) */
1411 OP_NOT_HSPACE, /* 18 \H (not horizontal whitespace) */
1412 OP_HSPACE, /* 19 \h (horizontal whitespace) */
1413 OP_NOT_VSPACE, /* 20 \V (not vertical whitespace) */
1414 OP_VSPACE, /* 21 \v (vertical whitespace) */
1415 OP_EXTUNI, /* 22 \X (extended Unicode sequence */
1416 OP_EODN, /* 23 End of data or \n at end of data: \Z. */
1417 OP_EOD, /* 24 End of data: \z */
1418
1419 OP_CIRC, /* 25 Start of line - not multiline */
1420 OP_CIRCM, /* 26 Start of line - multiline */
1421 OP_DOLL, /* 27 End of line - not multiline */
1422 OP_DOLLM, /* 28 End of line - multiline */
1423 OP_CHAR, /* 29 Match one character, casefully */
1424 OP_CHARI, /* 30 Match one character, caselessly */
1425 OP_NOT, /* 31 Match one character, not the given one, casefully */
1426 OP_NOTI, /* 32 Match one character, not the given one, caselessly */
1427
1428 /* The following sets of 13 opcodes must always be kept in step because
1429 the offset from the first one is used to generate the others. */
1430
1431 /**** Single characters, caseful, must precede the caseless ones ****/
1432
1433 OP_STAR, /* 33 The maximizing and minimizing versions of */
1434 OP_MINSTAR, /* 34 these six opcodes must come in pairs, with */
1435 OP_PLUS, /* 35 the minimizing one second. */
1436 OP_MINPLUS, /* 36 */
1437 OP_QUERY, /* 37 */
1438 OP_MINQUERY, /* 38 */
1439
1440 OP_UPTO, /* 39 From 0 to n matches of one character, caseful*/
1441 OP_MINUPTO, /* 40 */
1442 OP_EXACT, /* 41 Exactly n matches */
1443
1444 OP_POSSTAR, /* 42 Possessified star, caseful */
1445 OP_POSPLUS, /* 43 Possessified plus, caseful */
1446 OP_POSQUERY, /* 44 Posesssified query, caseful */
1447 OP_POSUPTO, /* 45 Possessified upto, caseful */
1448
1449 /**** Single characters, caseless, must follow the caseful ones */
1450
1451 OP_STARI, /* 46 */
1452 OP_MINSTARI, /* 47 */
1453 OP_PLUSI, /* 48 */
1454 OP_MINPLUSI, /* 49 */
1455 OP_QUERYI, /* 50 */
1456 OP_MINQUERYI, /* 51 */
1457
1458 OP_UPTOI, /* 52 From 0 to n matches of one character, caseless */
1459 OP_MINUPTOI, /* 53 */
1460 OP_EXACTI, /* 54 */
1461
1462 OP_POSSTARI, /* 55 Possessified star, caseless */
1463 OP_POSPLUSI, /* 56 Possessified plus, caseless */
1464 OP_POSQUERYI, /* 57 Posesssified query, caseless */
1465 OP_POSUPTOI, /* 58 Possessified upto, caseless */
1466
1467 /**** The negated ones must follow the non-negated ones, and match them ****/
1468 /**** Negated single character, caseful; must precede the caseless ones ****/
1469
1470 OP_NOTSTAR, /* 59 The maximizing and minimizing versions of */
1471 OP_NOTMINSTAR, /* 60 these six opcodes must come in pairs, with */
1472 OP_NOTPLUS, /* 61 the minimizing one second. They must be in */
1473 OP_NOTMINPLUS, /* 62 exactly the same order as those above. */
1474 OP_NOTQUERY, /* 63 */
1475 OP_NOTMINQUERY, /* 64 */
1476
1477 OP_NOTUPTO, /* 65 From 0 to n matches, caseful */
1478 OP_NOTMINUPTO, /* 66 */
1479 OP_NOTEXACT, /* 67 Exactly n matches */
1480
1481 OP_NOTPOSSTAR, /* 68 Possessified versions, caseful */
1482 OP_NOTPOSPLUS, /* 69 */
1483 OP_NOTPOSQUERY, /* 70 */
1484 OP_NOTPOSUPTO, /* 71 */
1485
1486 /**** Negated single character, caseless; must follow the caseful ones ****/
1487
1488 OP_NOTSTARI, /* 72 */
1489 OP_NOTMINSTARI, /* 73 */
1490 OP_NOTPLUSI, /* 74 */
1491 OP_NOTMINPLUSI, /* 75 */
1492 OP_NOTQUERYI, /* 76 */
1493 OP_NOTMINQUERYI, /* 77 */
1494
1495 OP_NOTUPTOI, /* 78 From 0 to n matches, caseless */
1496 OP_NOTMINUPTOI, /* 79 */
1497 OP_NOTEXACTI, /* 80 Exactly n matches */
1498
1499 OP_NOTPOSSTARI, /* 81 Possessified versions, caseless */
1500 OP_NOTPOSPLUSI, /* 82 */
1501 OP_NOTPOSQUERYI, /* 83 */
1502 OP_NOTPOSUPTOI, /* 84 */
1503
1504 /**** Character types ****/
1505
1506 OP_TYPESTAR, /* 85 The maximizing and minimizing versions of */
1507 OP_TYPEMINSTAR, /* 86 these six opcodes must come in pairs, with */
1508 OP_TYPEPLUS, /* 87 the minimizing one second. These codes must */
1509 OP_TYPEMINPLUS, /* 88 be in exactly the same order as those above. */
1510 OP_TYPEQUERY, /* 89 */
1511 OP_TYPEMINQUERY, /* 90 */
1512
1513 OP_TYPEUPTO, /* 91 From 0 to n matches */
1514 OP_TYPEMINUPTO, /* 92 */
1515 OP_TYPEEXACT, /* 93 Exactly n matches */
1516
1517 OP_TYPEPOSSTAR, /* 94 Possessified versions */
1518 OP_TYPEPOSPLUS, /* 95 */
1519 OP_TYPEPOSQUERY, /* 96 */
1520 OP_TYPEPOSUPTO, /* 97 */
1521
1522 /* These are used for character classes and back references; only the
1523 first six are the same as the sets above. */
1524
1525 OP_CRSTAR, /* 98 The maximizing and minimizing versions of */
1526 OP_CRMINSTAR, /* 99 all these opcodes must come in pairs, with */
1527 OP_CRPLUS, /* 100 the minimizing one second. These codes must */
1528 OP_CRMINPLUS, /* 101 be in exactly the same order as those above. */
1529 OP_CRQUERY, /* 102 */
1530 OP_CRMINQUERY, /* 103 */
1531
1532 OP_CRRANGE, /* 104 These are different to the three sets above. */
1533 OP_CRMINRANGE, /* 105 */
1534
1535 /* End of quantifier opcodes */
1536
1537 OP_CLASS, /* 106 Match a character class, chars < 256 only */
1538 OP_NCLASS, /* 107 Same, but the bitmap was created from a negative
1539 class - the difference is relevant only when a
1540 character > 255 is encountered. */
1541 OP_XCLASS, /* 108 Extended class for handling > 255 chars within the
1542 class. This does both positive and negative. */
1543 OP_REF, /* 109 Match a back reference, casefully */
1544 OP_REFI, /* 110 Match a back reference, caselessly */
1545 OP_RECURSE, /* 111 Match a numbered subpattern (possibly recursive) */
1546 OP_CALLOUT, /* 112 Call out to external function if provided */
1547
1548 OP_ALT, /* 113 Start of alternation */
1549 OP_KET, /* 114 End of group that doesn't have an unbounded repeat */
1550 OP_KETRMAX, /* 115 These two must remain together and in this */
1551 OP_KETRMIN, /* 116 order. They are for groups the repeat for ever. */
1552 OP_KETRPOS, /* 117 Possessive unlimited repeat. */
1553
1554 /* The assertions must come before BRA, CBRA, ONCE, and COND, and the four
1555 asserts must remain in order. */
1556
1557 OP_REVERSE, /* 118 Move pointer back - used in lookbehind assertions */
1558 OP_ASSERT, /* 119 Positive lookahead */
1559 OP_ASSERT_NOT, /* 120 Negative lookahead */
1560 OP_ASSERTBACK, /* 121 Positive lookbehind */
1561 OP_ASSERTBACK_NOT, /* 122 Negative lookbehind */
1562
1563 /* ONCE, ONCE_NC, BRA, BRAPOS, CBRA, CBRAPOS, and COND must come immediately
1564 after the assertions, with ONCE first, as there's a test for >= ONCE for a
1565 subpattern that isn't an assertion. The POS versions must immediately follow
1566 the non-POS versions in each case. */
1567
1568 OP_ONCE, /* 123 Atomic group, contains captures */
1569 OP_ONCE_NC, /* 124 Atomic group containing no captures */
1570 OP_BRA, /* 125 Start of non-capturing bracket */
1571 OP_BRAPOS, /* 126 Ditto, with unlimited, possessive repeat */
1572 OP_CBRA, /* 127 Start of capturing bracket */
1573 OP_CBRAPOS, /* 128 Ditto, with unlimited, possessive repeat */
1574 OP_COND, /* 129 Conditional group */
1575
1576 /* These five must follow the previous five, in the same order. There's a
1577 check for >= SBRA to distinguish the two sets. */
1578
1579 OP_SBRA, /* 130 Start of non-capturing bracket, check empty */
1580 OP_SBRAPOS, /* 131 Ditto, with unlimited, possessive repeat */
1581 OP_SCBRA, /* 132 Start of capturing bracket, check empty */
1582 OP_SCBRAPOS, /* 133 Ditto, with unlimited, possessive repeat */
1583 OP_SCOND, /* 134 Conditional group, check empty */
1584
1585 /* The next two pairs must (respectively) be kept together. */
1586
1587 OP_CREF, /* 135 Used to hold a capture number as condition */
1588 OP_NCREF, /* 136 Same, but generated by a name reference*/
1589 OP_RREF, /* 137 Used to hold a recursion number as condition */
1590 OP_NRREF, /* 138 Same, but generated by a name reference*/
1591 OP_DEF, /* 139 The DEFINE condition */
1592
1593 OP_BRAZERO, /* 140 These two must remain together and in this */
1594 OP_BRAMINZERO, /* 141 order. */
1595 OP_BRAPOSZERO, /* 142 */
1596
1597 /* These are backtracking control verbs */
1598
1599 OP_MARK, /* 143 always has an argument */
1600 OP_PRUNE, /* 144 */
1601 OP_PRUNE_ARG, /* 145 same, but with argument */
1602 OP_SKIP, /* 146 */
1603 OP_SKIP_ARG, /* 147 same, but with argument */
1604 OP_THEN, /* 148 */
1605 OP_THEN_ARG, /* 149 same, but with argument */
1606 OP_COMMIT, /* 150 */
1607
1608 /* These are forced failure and success verbs */
1609
1610 OP_FAIL, /* 151 */
1611 OP_ACCEPT, /* 152 */
1612 OP_ASSERT_ACCEPT, /* 153 Used inside assertions */
1613 OP_CLOSE, /* 154 Used before OP_ACCEPT to close open captures */
1614
1615 /* This is used to skip a subpattern with a {0} quantifier */
1616
1617 OP_SKIPZERO, /* 155 */
1618
1619 /* This is not an opcode, but is used to check that tables indexed by opcode
1620 are the correct length, in order to catch updating errors - there have been
1621 some in the past. */
1622
1623 OP_TABLE_LENGTH
1624 };
1625
1626 /* *** NOTE NOTE NOTE *** Whenever the list above is updated, the two macro
1627 definitions that follow must also be updated to match. There are also tables
1628 called "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
1629
1630
1631 /* This macro defines textual names for all the opcodes. These are used only
1632 for debugging, and some of them are only partial names. The macro is referenced
1633 only in pcre_printint.c, which fills out the full names in many cases (and in
1634 some cases doesn't actually use these names at all). */
1635
1636 #define OP_NAME_LIST \
1637 "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d", \
1638 "\\S", "\\s", "\\W", "\\w", "Any", "AllAny", "Anybyte", \
1639 "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v", \
1640 "extuni", "\\Z", "\\z", \
1641 "^", "^", "$", "$", "char", "chari", "not", "noti", \
1642 "*", "*?", "+", "+?", "?", "??", \
1643 "{", "{", "{", \
1644 "*+","++", "?+", "{", \
1645 "*", "*?", "+", "+?", "?", "??", \
1646 "{", "{", "{", \
1647 "*+","++", "?+", "{", \
1648 "*", "*?", "+", "+?", "?", "??", \
1649 "{", "{", "{", \
1650 "*+","++", "?+", "{", \
1651 "*", "*?", "+", "+?", "?", "??", \
1652 "{", "{", "{", \
1653 "*+","++", "?+", "{", \
1654 "*", "*?", "+", "+?", "?", "??", "{", "{", "{", \
1655 "*+","++", "?+", "{", \
1656 "*", "*?", "+", "+?", "?", "??", "{", "{", \
1657 "class", "nclass", "xclass", "Ref", "Refi", \
1658 "Recurse", "Callout", \
1659 "Alt", "Ket", "KetRmax", "KetRmin", "KetRpos", \
1660 "Reverse", "Assert", "Assert not", "AssertB", "AssertB not", \
1661 "Once", "Once_NC", \
1662 "Bra", "BraPos", "CBra", "CBraPos", \
1663 "Cond", \
1664 "SBra", "SBraPos", "SCBra", "SCBraPos", \
1665 "SCond", \
1666 "Cond ref", "Cond nref", "Cond rec", "Cond nrec", "Cond def", \
1667 "Brazero", "Braminzero", "Braposzero", \
1668 "*MARK", "*PRUNE", "*PRUNE", "*SKIP", "*SKIP", \
1669 "*THEN", "*THEN", "*COMMIT", "*FAIL", \
1670 "*ACCEPT", "*ASSERT_ACCEPT", \
1671 "Close", "Skip zero"
1672
1673
1674 /* This macro defines the length of fixed length operations in the compiled
1675 regex. The lengths are used when searching for specific things, and also in the
1676 debugging printing of a compiled regex. We use a macro so that it can be
1677 defined close to the definitions of the opcodes themselves.
1678
1679 As things have been extended, some of these are no longer fixed lenths, but are
1680 minima instead. For example, the length of a single-character repeat may vary
1681 in UTF-8 mode. The code that uses this table must know about such things. */
1682
1683 #define OP_LENGTHS \
1684 1, /* End */ \
1685 1, 1, 1, 1, 1, /* \A, \G, \K, \B, \b */ \
1686 1, 1, 1, 1, 1, 1, /* \D, \d, \S, \s, \W, \w */ \
1687 1, 1, 1, /* Any, AllAny, Anybyte */ \
1688 3, 3, /* \P, \p */ \
1689 1, 1, 1, 1, 1, /* \R, \H, \h, \V, \v */ \
1690 1, /* \X */ \
1691 1, 1, 1, 1, 1, 1, /* \Z, \z, ^, ^M, $, $M */ \
1692 2, /* Char - the minimum length */ \
1693 2, /* Chari - the minimum length */ \
1694 2, /* not */ \
1695 2, /* noti */ \
1696 /* Positive single-char repeats ** These are */ \
1697 2, 2, 2, 2, 2, 2, /* *, *?, +, +?, ?, ?? ** minima in */ \
1698 2+IMM2_SIZE, 2+IMM2_SIZE, /* upto, minupto ** mode */ \
1699 2+IMM2_SIZE, /* exact */ \
1700 2, 2, 2, 2+IMM2_SIZE, /* *+, ++, ?+, upto+ */ \
1701 2, 2, 2, 2, 2, 2, /* *I, *?I, +I, +?I, ?I, ??I ** UTF-8 */ \
1702 2+IMM2_SIZE, 2+IMM2_SIZE, /* upto I, minupto I */ \
1703 2+IMM2_SIZE, /* exact I */ \
1704 2, 2, 2, 2+IMM2_SIZE, /* *+I, ++I, ?+I, upto+I */ \
1705 /* Negative single-char repeats - only for chars < 256 */ \
1706 2, 2, 2, 2, 2, 2, /* NOT *, *?, +, +?, ?, ?? */ \
1707 2+IMM2_SIZE, 2+IMM2_SIZE, /* NOT upto, minupto */ \
1708 2+IMM2_SIZE, /* NOT exact */ \
1709 2, 2, 2, 2+IMM2_SIZE, /* Possessive NOT *, +, ?, upto */ \
1710 2, 2, 2, 2, 2, 2, /* NOT *I, *?I, +I, +?I, ?I, ??I */ \
1711 2+IMM2_SIZE, 2+IMM2_SIZE, /* NOT upto I, minupto I */ \
1712 2+IMM2_SIZE, /* NOT exact I */ \
1713 2, 2, 2, 2+IMM2_SIZE, /* Possessive NOT *I, +I, ?I, upto I */ \
1714 /* Positive type repeats */ \
1715 2, 2, 2, 2, 2, 2, /* Type *, *?, +, +?, ?, ?? */ \
1716 2+IMM2_SIZE, 2+IMM2_SIZE, /* Type upto, minupto */ \
1717 2+IMM2_SIZE, /* Type exact */ \
1718 2, 2, 2, 2+IMM2_SIZE, /* Possessive *+, ++, ?+, upto+ */ \
1719 /* Character class & ref repeats */ \
1720 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ \
1721 1+2*IMM2_SIZE, 1+2*IMM2_SIZE, /* CRRANGE, CRMINRANGE */ \
1722 1+(32/sizeof(pcre_uchar)), /* CLASS */ \
1723 1+(32/sizeof(pcre_uchar)), /* NCLASS */ \
1724 0, /* XCLASS - variable length */ \
1725 1+IMM2_SIZE, /* REF */ \
1726 1+IMM2_SIZE, /* REFI */ \
1727 1+LINK_SIZE, /* RECURSE */ \
1728 2+2*LINK_SIZE, /* CALLOUT */ \
1729 1+LINK_SIZE, /* Alt */ \
1730 1+LINK_SIZE, /* Ket */ \
1731 1+LINK_SIZE, /* KetRmax */ \
1732 1+LINK_SIZE, /* KetRmin */ \
1733 1+LINK_SIZE, /* KetRpos */ \
1734 1+LINK_SIZE, /* Reverse */ \
1735 1+LINK_SIZE, /* Assert */ \
1736 1+LINK_SIZE, /* Assert not */ \
1737 1+LINK_SIZE, /* Assert behind */ \
1738 1+LINK_SIZE, /* Assert behind not */ \
1739 1+LINK_SIZE, /* ONCE */ \
1740 1+LINK_SIZE, /* ONCE_NC */ \
1741 1+LINK_SIZE, /* BRA */ \
1742 1+LINK_SIZE, /* BRAPOS */ \
1743 1+LINK_SIZE+IMM2_SIZE, /* CBRA */ \
1744 1+LINK_SIZE+IMM2_SIZE, /* CBRAPOS */ \
1745 1+LINK_SIZE, /* COND */ \
1746 1+LINK_SIZE, /* SBRA */ \
1747 1+LINK_SIZE, /* SBRAPOS */ \
1748 1+LINK_SIZE+IMM2_SIZE, /* SCBRA */ \
1749 1+LINK_SIZE+IMM2_SIZE, /* SCBRAPOS */ \
1750 1+LINK_SIZE, /* SCOND */ \
1751 1+IMM2_SIZE, 1+IMM2_SIZE, /* CREF, NCREF */ \
1752 1+IMM2_SIZE, 1+IMM2_SIZE, /* RREF, NRREF */ \
1753 1, /* DEF */ \
1754 1, 1, 1, /* BRAZERO, BRAMINZERO, BRAPOSZERO */ \
1755 3, 1, 3, /* MARK, PRUNE, PRUNE_ARG */ \
1756 1, 3, /* SKIP, SKIP_ARG */ \
1757 1, 3, /* THEN, THEN_ARG */ \
1758 1, 1, 1, 1, /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT */ \
1759 1+IMM2_SIZE, 1 /* CLOSE, SKIPZERO */
1760
1761 /* A magic value for OP_RREF and OP_NRREF to indicate the "any recursion"
1762 condition. */
1763
1764 #define RREF_ANY 0xffff
1765
1766 /* Compile time error code numbers. They are given names so that they can more
1767 easily be tracked. When a new number is added, the table called eint in
1768 pcreposix.c must be updated. */
1769
1770 enum { ERR0, ERR1, ERR2, ERR3, ERR4, ERR5, ERR6, ERR7, ERR8, ERR9,
1771 ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,
1772 ERR20, ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR28, ERR29,
1773 ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,
1774 ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,
1775 ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,
1776 ERR60, ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERR68, ERR69,
1777 ERR70, ERRCOUNT };
1778
1779 /* The real format of the start of the pcre block; the index of names and the
1780 code vector run on as long as necessary after the end. We store an explicit
1781 offset to the name table so that if a regex is compiled on one host, saved, and
1782 then run on another where the size of pointers is different, all might still
1783 be well. For the case of compiled-on-4 and run-on-8, we include an extra
1784 pointer that is always NULL. For future-proofing, a few dummy fields were
1785 originally included - even though you can never get this planning right - but
1786 there is only one left now.
1787
1788 NOTE NOTE NOTE:
1789 Because people can now save and re-use compiled patterns, any additions to this
1790 structure should be made at the end, and something earlier (e.g. a new
1791 flag in the options or one of the dummy fields) should indicate that the new
1792 fields are present. Currently PCRE always sets the dummy fields to zero.
1793 NOTE NOTE NOTE
1794 */
1795
1796 typedef struct real_pcre {
1797 pcre_uint32 magic_number;
1798 pcre_uint32 size; /* Total that was malloced */
1799 pcre_uint32 options; /* Public options */
1800 pcre_uint16 flags; /* Private flags */
1801 pcre_uint16 dummy1; /* For future use */
1802 pcre_uint16 top_bracket;
1803 pcre_uint16 top_backref;
1804 pcre_uint16 first_byte;
1805 pcre_uint16 req_byte;
1806 pcre_uint16 name_table_offset; /* Offset to name table that follows */
1807 pcre_uint16 name_entry_size; /* Size of any name items */
1808 pcre_uint16 name_count; /* Number of name items */
1809 pcre_uint16 ref_count; /* Reference count */
1810
1811 const pcre_uint8 *tables; /* Pointer to tables or NULL for std */
1812 const pcre_uint8 *nullpad; /* NULL padding */
1813 } real_pcre;
1814
1815 /* The format of the block used to store data from pcre_study(). The same
1816 remark (see NOTE above) about extending this structure applies. */
1817
1818 typedef struct pcre_study_data {
1819 pcre_uint32 size; /* Total that was malloced */
1820 pcre_uint32 flags; /* Private flags */
1821 pcre_uint8 start_bits[32]; /* Starting char bits */
1822 pcre_uint32 minlength; /* Minimum subject length */
1823 } pcre_study_data;
1824
1825 /* Structure for building a chain of open capturing subpatterns during
1826 compiling, so that instructions to close them can be compiled when (*ACCEPT) is
1827 encountered. This is also used to identify subpatterns that contain recursive
1828 back references to themselves, so that they can be made atomic. */
1829
1830 typedef struct open_capitem {
1831 struct open_capitem *next; /* Chain link */
1832 pcre_uint16 number; /* Capture number */
1833 pcre_uint16 flag; /* Set TRUE if recursive back ref */
1834 } open_capitem;
1835
1836 /* Structure for passing "static" information around between the functions
1837 doing the compiling, so that they are thread-safe. */
1838
1839 typedef struct compile_data {
1840 const pcre_uint8 *lcc; /* Points to lower casing table */
1841 const pcre_uint8 *fcc; /* Points to case-flipping table */
1842 const pcre_uint8 *cbits; /* Points to character type table */
1843 const pcre_uint8 *ctypes; /* Points to table of type maps */
1844 const pcre_uchar *start_workspace;/* The start of working space */
1845 const pcre_uchar *start_code; /* The start of the compiled code */
1846 const pcre_uchar *start_pattern; /* The start of the pattern */
1847 const pcre_uchar *end_pattern; /* The end of the pattern */
1848 open_capitem *open_caps; /* Chain of open capture items */
1849 pcre_uchar *hwm; /* High watermark of workspace */
1850 pcre_uchar *name_table; /* The name/number table */
1851 int names_found; /* Number of entries so far */
1852 int name_entry_size; /* Size of each entry */
1853 int bracount; /* Count of capturing parens as we compile */
1854 int final_bracount; /* Saved value after first pass */
1855 int top_backref; /* Maximum back reference */
1856 unsigned int backref_map; /* Bitmap of low back refs */
1857 int assert_depth; /* Depth of nested assertions */
1858 int external_options; /* External (initial) options */
1859 int external_flags; /* External flag bits to be set */
1860 int req_varyopt; /* "After variable item" flag for reqbyte */
1861 BOOL had_accept; /* (*ACCEPT) encountered */
1862 BOOL check_lookbehind; /* Lookbehinds need later checking */
1863 int nltype; /* Newline type */
1864 int nllen; /* Newline string length */
1865 pcre_uchar nl[4]; /* Newline string when fixed length */
1866 } compile_data;
1867
1868 /* Structure for maintaining a chain of pointers to the currently incomplete
1869 branches, for testing for left recursion while compiling. */
1870
1871 typedef struct branch_chain {
1872 struct branch_chain *outer;
1873 pcre_uchar *current_branch;
1874 } branch_chain;
1875
1876 /* Structure for items in a linked list that represents an explicit recursive
1877 call within the pattern; used by pcre_exec(). */
1878
1879 typedef struct recursion_info {
1880 struct recursion_info *prevrec; /* Previous recursion record (or NULL) */
1881 int group_num; /* Number of group that was called */
1882 int *offset_save; /* Pointer to start of saved offsets */
1883 int saved_max; /* Number of saved offsets */
1884 PCRE_PUCHAR subject_position; /* Position at start of recursion */
1885 } recursion_info;
1886
1887 /* A similar structure for pcre_dfa_exec(). */
1888
1889 typedef struct dfa_recursion_info {
1890 struct dfa_recursion_info *prevrec;
1891 int group_num;
1892 PCRE_PUCHAR subject_position;
1893 } dfa_recursion_info;
1894
1895 /* Structure for building a chain of data for holding the values of the subject
1896 pointer at the start of each subpattern, so as to detect when an empty string
1897 has been matched by a subpattern - to break infinite loops; used by
1898 pcre_exec(). */
1899
1900 typedef struct eptrblock {
1901 struct eptrblock *epb_prev;
1902 PCRE_PUCHAR epb_saved_eptr;
1903 } eptrblock;
1904
1905
1906 /* Structure for passing "static" information around between the functions
1907 doing traditional NFA matching, so that they are thread-safe. */
1908
1909 typedef struct match_data {
1910 unsigned long int match_call_count; /* As it says */
1911 unsigned long int match_limit; /* As it says */
1912 unsigned long int match_limit_recursion; /* As it says */
1913 int *offset_vector; /* Offset vector */
1914 int offset_end; /* One past the end */
1915 int offset_max; /* The maximum usable for return data */
1916 int nltype; /* Newline type */
1917 int nllen; /* Newline string length */
1918 int name_count; /* Number of names in name table */
1919 int name_entry_size; /* Size of entry in names table */
1920 pcre_uchar *name_table; /* Table of names */
1921 pcre_uchar nl[4]; /* Newline string when fixed */
1922 const pcre_uint8 *lcc; /* Points to lower casing table */
1923 const pcre_uint8 *ctypes; /* Points to table of type maps */
1924 BOOL offset_overflow; /* Set if too many extractions */
1925 BOOL notbol; /* NOTBOL flag */
1926 BOOL noteol; /* NOTEOL flag */
1927 BOOL utf8; /* UTF8 flag */
1928 BOOL jscript_compat; /* JAVASCRIPT_COMPAT flag */
1929 BOOL use_ucp; /* PCRE_UCP flag */
1930 BOOL endonly; /* Dollar not before final \n */
1931 BOOL notempty; /* Empty string match not wanted */
1932 BOOL notempty_atstart; /* Empty string match at start not wanted */
1933 BOOL hitend; /* Hit the end of the subject at some point */
1934 BOOL bsr_anycrlf; /* \R is just any CRLF, not full Unicode */
1935 BOOL hasthen; /* Pattern contains (*THEN) */
1936 const pcre_uchar *start_code; /* For use when recursing */
1937 PCRE_PUCHAR start_subject; /* Start of the subject string */
1938 PCRE_PUCHAR end_subject; /* End of the subject string */
1939 PCRE_PUCHAR start_match_ptr; /* Start of matched string */
1940 PCRE_PUCHAR end_match_ptr; /* Subject position at end match */
1941 PCRE_PUCHAR start_used_ptr; /* Earliest consulted character */
1942 int partial; /* PARTIAL options */
1943 int end_offset_top; /* Highwater mark at end of match */
1944 int capture_last; /* Most recent capture number */
1945 int start_offset; /* The start offset value */
1946 int match_function_type; /* Set for certain special calls of MATCH() */
1947 eptrblock *eptrchain; /* Chain of eptrblocks for tail recursions */
1948 int eptrn; /* Next free eptrblock */
1949 recursion_info *recursive; /* Linked list of recursion data */
1950 void *callout_data; /* To pass back to callouts */
1951 const pcre_uchar *mark; /* Mark pointer to pass back */
1952 const pcre_uchar *once_target; /* Where to back up to for atomic groups */
1953 } match_data;
1954
1955 /* A similar structure is used for the same purpose by the DFA matching
1956 functions. */
1957
1958 typedef struct dfa_match_data {
1959 const pcre_uchar *start_code; /* Start of the compiled pattern */
1960 const pcre_uchar *start_subject ; /* Start of the subject string */
1961 const pcre_uchar *end_subject; /* End of subject string */
1962 const pcre_uchar *start_used_ptr; /* Earliest consulted character */
1963 const pcre_uint8 *tables; /* Character tables */
1964 int start_offset; /* The start offset value */
1965 int moptions; /* Match options */
1966 int poptions; /* Pattern options */
1967 int nltype; /* Newline type */
1968 int nllen; /* Newline string length */
1969 pcre_uchar nl[4]; /* Newline string when fixed */
1970 void *callout_data; /* To pass back to callouts */
1971 dfa_recursion_info *recursive; /* Linked list of recursion data */
1972 } dfa_match_data;
1973
1974 /* Bit definitions for entries in the pcre_ctypes table. */
1975
1976 #define ctype_space 0x01
1977 #define ctype_letter 0x02
1978 #define ctype_digit 0x04
1979 #define ctype_xdigit 0x08
1980 #define ctype_word 0x10 /* alphanumeric or '_' */
1981 #define ctype_meta 0x80 /* regexp meta char or zero (end pattern) */
1982
1983 /* Offsets for the bitmap tables in pcre_cbits. Each table contains a set
1984 of bits for a class map. Some classes are built by combining these tables. */
1985
1986 #define cbit_space 0 /* [:space:] or \s */
1987 #define cbit_xdigit 32 /* [:xdigit:] */
1988 #define cbit_digit 64 /* [:digit:] or \d */
1989 #define cbit_upper 96 /* [:upper:] */
1990 #define cbit_lower 128 /* [:lower:] */
1991 #define cbit_word 160 /* [:word:] or \w */
1992 #define cbit_graph 192 /* [:graph:] */
1993 #define cbit_print 224 /* [:print:] */
1994 #define cbit_punct 256 /* [:punct:] */
1995 #define cbit_cntrl 288 /* [:cntrl:] */
1996 #define cbit_length 320 /* Length of the cbits table */
1997
1998 /* Offsets of the various tables from the base tables pointer, and
1999 total length. */
2000
2001 #define lcc_offset 0
2002 #define fcc_offset 256
2003 #define cbits_offset 512
2004 #define ctypes_offset (cbits_offset + cbit_length)
2005 #define tables_length (ctypes_offset + 256)
2006
2007 /* Internal function prefix */
2008
2009 #ifdef COMPILE_PCRE8
2010 #define PRIV(name) _pcre_##name
2011 #else
2012 #ifdef COMPILE_PCRE16
2013 #define PRIV(name) _pcre16_##name
2014 #else
2015 #error Unsupported compiling mode
2016 #endif /* COMPILE_PCRE16 */
2017 #endif /* COMPILE_PCRE8 */
2018
2019 /* Layout of the UCP type table that translates property names into types and
2020 codes. Each entry used to point directly to a name, but to reduce the number of
2021 relocations in shared libraries, it now has an offset into a single string
2022 instead. */
2023
2024 typedef struct {
2025 pcre_uint16 name_offset;
2026 pcre_uint16 type;
2027 pcre_uint16 value;
2028 } ucp_type_table;
2029
2030
2031 /* Internal shared data tables. These are tables that are used by more than one
2032 of the exported public functions. They have to be "external" in the C sense,
2033 but are not part of the PCRE public API. The data for these tables is in the
2034 pcre_tables.c module. */
2035
2036 extern const int PRIV(utf8_table1)[];
2037 extern const int PRIV(utf8_table2)[];
2038 extern const int PRIV(utf8_table3)[];
2039 extern const pcre_uint8 PRIV(utf8_table4)[];
2040
2041 extern const int PRIV(utf8_table1_size);
2042
2043 extern const char PRIV(utt_names)[];
2044 extern const ucp_type_table PRIV(utt)[];
2045 extern const int PRIV(utt_size);
2046
2047 extern const pcre_uint8 PRIV(default_tables)[];
2048
2049 extern const pcre_uint8 PRIV(OP_lengths)[];
2050
2051
2052 /* Internal shared functions. These are functions that are used by more than
2053 one of the exported public functions. They have to be "external" in the C
2054 sense, but are not part of the PCRE public API. */
2055
2056 /* String comparison functions. */
2057 #ifdef COMPILE_PCRE8
2058
2059 #define STRCMP_UC_UC(str1, str2) \
2060 strcmp((char *)(str1), (char *)(str2))
2061 #define STRCMP_UC_C8(str1, str2) \
2062 strcmp((char *)(str1), (str2))
2063 #define STRNCMP_UC_UC(str1, str2, num) \
2064 strncmp((char *)(str1), (char *)(str2), (num))
2065 #define STRNCMP_UC_C8(str1, str2, num) \
2066 strncmp((char *)(str1), (str2), (num))
2067 #define STRLEN_UC(str) strlen((const char *)str)
2068
2069 #else
2070
2071 extern int PRIV(strcmp_uc_uc)(const pcre_uchar *,
2072 const pcre_uchar *);
2073 extern int PRIV(strcmp_uc_c8)(const pcre_uchar *,
2074 const char *);
2075 extern int PRIV(strncmp_uc_uc)(const pcre_uchar *,
2076 const pcre_uchar *, unsigned int num);
2077 extern int PRIV(strncmp_uc_c8)(const pcre_uchar *,
2078 const char *, unsigned int num);
2079 extern unsigned int PRIV(strlen_uc)(const pcre_uchar *str);
2080
2081 #define STRCMP_UC_UC(str1, str2) \
2082 PRIV(strcmp_uc_uc)((str1), (str2))
2083 #define STRCMP_UC_C8(str1, str2) \
2084 PRIV(strcmp_uc_c8)((str1), (str2))
2085 #define STRNCMP_UC_UC(str1, str2, num) \
2086 PRIV(strncmp_uc_uc)((str1), (str2), (num))
2087 #define STRNCMP_UC_C8(str1, str2, num) \
2088 PRIV(strncmp_uc_c8)((str1), (str2), (num))
2089 #define STRLEN_UC(str) PRIV(strlen_uc)(str)
2090
2091 #endif /* COMPILE_PCRE8 */
2092
2093 extern const pcre_uchar *PRIV(find_bracket)(const pcre_uchar *, BOOL, int);
2094 extern BOOL PRIV(is_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2095 int *, BOOL);
2096 extern int PRIV(ord2utf8)(int, pcre_uint8 *);
2097 extern real_pcre *PRIV(try_flipped)(const real_pcre *, real_pcre *,
2098 const pcre_study_data *, pcre_study_data *);
2099 #ifndef COMPILE_PCRE16
2100 extern int PRIV(valid_utf8)(PCRE_PUCHAR, int, int *);
2101 #else
2102 extern int PRIV(valid_utf16)(PCRE_PUCHAR, int, int *);
2103 #endif
2104 extern BOOL PRIV(was_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2105 int *, BOOL);
2106 extern BOOL PRIV(xclass)(int, const pcre_uchar *);
2107
2108 #ifdef SUPPORT_JIT
2109 extern void PRIV(jit_compile)(const real_pcre *, pcre_extra *);
2110 extern int PRIV(jit_exec)(const real_pcre *, void *,
2111 const pcre_uchar *, int, int, int, int, int *, int);
2112 extern void PRIV(jit_free)(void *);
2113 #endif
2114
2115 /* Unicode character database (UCD) */
2116
2117 typedef struct {
2118 pcre_uint8 script;
2119 pcre_uint8 chartype;
2120 pcre_int32 other_case;
2121 } ucd_record;
2122
2123 extern const ucd_record PRIV(ucd_records)[];
2124 extern const pcre_uint8 PRIV(ucd_stage1)[];
2125 extern const pcre_uint16 PRIV(ucd_stage2)[];
2126 extern const int PRIV(ucp_gentype)[];
2127 #ifdef SUPPORT_JIT
2128 extern const int PRIV(ucp_typerange)[];
2129 #endif
2130
2131 /* UCD access macros */
2132
2133 #define UCD_BLOCK_SIZE 128
2134 #define GET_UCD(ch) (PRIV(ucd_records) + \
2135 PRIV(ucd_stage2)[PRIV(ucd_stage1)[(ch) / UCD_BLOCK_SIZE] * \
2136 UCD_BLOCK_SIZE + (ch) % UCD_BLOCK_SIZE])
2137
2138 #define UCD_CHARTYPE(ch) GET_UCD(ch)->chartype
2139 #define UCD_SCRIPT(ch) GET_UCD(ch)->script
2140 #define UCD_CATEGORY(ch) PRIV(ucp_gentype)[UCD_CHARTYPE(ch)]
2141 #define UCD_OTHERCASE(ch) (ch + GET_UCD(ch)->other_case)
2142
2143 #endif
2144
2145 /* 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