/[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 767 - (show annotations)
Sat Nov 26 12:48:56 2011 UTC (3 years, 5 months ago) by zherczeg
File MIME type: text/plain
File size: 87901 byte(s)
Error occurred while calculating annotation data.
Make simple patterns work in PCRE16
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 /* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef
75 inline, and there are *still* stupid compilers about that don't like indented
76 pre-processor statements, or at least there were when I first wrote this. After
77 all, it had only been about 10 years then...
78
79 It turns out that the Mac Debugging.h header also defines the macro DPRINTF, so
80 be absolutely sure we get our version. */
81
82 #undef DPRINTF
83 #ifdef PCRE_DEBUG
84 #define DPRINTF(p) printf p
85 #else
86 #define DPRINTF(p) /* Nothing */
87 #endif
88
89
90 /* Standard C headers plus the external interface definition. The only time
91 setjmp and stdarg are used is when NO_RECURSE is set. */
92
93 #include <ctype.h>
94 #include <limits.h>
95 #include <stddef.h>
96 #include <stdio.h>
97 #include <stdlib.h>
98 #include <string.h>
99
100 /* When compiling a DLL for Windows, the exported symbols have to be declared
101 using some MS magic. I found some useful information on this web page:
102 http://msdn2.microsoft.com/en-us/library/y4h7bcy6(VS.80).aspx. According to the
103 information there, using __declspec(dllexport) without "extern" we have a
104 definition; with "extern" we have a declaration. The settings here override the
105 setting in pcre.h (which is included below); it defines only PCRE_EXP_DECL,
106 which is all that is needed for applications (they just import the symbols). We
107 use:
108
109 PCRE_EXP_DECL for declarations
110 PCRE_EXP_DEFN for definitions of exported functions
111 PCRE_EXP_DATA_DEFN for definitions of exported variables
112
113 The reason for the two DEFN macros is that in non-Windows environments, one
114 does not want to have "extern" before variable definitions because it leads to
115 compiler warnings. So we distinguish between functions and variables. In
116 Windows, the two should always be the same.
117
118 The reason for wrapping this in #ifndef PCRE_EXP_DECL is so that pcretest,
119 which is an application, but needs to import this file in order to "peek" at
120 internals, can #include pcre.h first to get an application's-eye view.
121
122 In principle, people compiling for non-Windows, non-Unix-like (i.e. uncommon,
123 special-purpose environments) might want to stick other stuff in front of
124 exported symbols. That's why, in the non-Windows case, we set PCRE_EXP_DEFN and
125 PCRE_EXP_DATA_DEFN only if they are not already set. */
126
127 #ifndef PCRE_EXP_DECL
128 # ifdef _WIN32
129 # ifndef PCRE_STATIC
130 # define PCRE_EXP_DECL extern __declspec(dllexport)
131 # define PCRE_EXP_DEFN __declspec(dllexport)
132 # define PCRE_EXP_DATA_DEFN __declspec(dllexport)
133 # else
134 # define PCRE_EXP_DECL extern
135 # define PCRE_EXP_DEFN
136 # define PCRE_EXP_DATA_DEFN
137 # endif
138 # else
139 # ifdef __cplusplus
140 # define PCRE_EXP_DECL extern "C"
141 # else
142 # define PCRE_EXP_DECL extern
143 # endif
144 # ifndef PCRE_EXP_DEFN
145 # define PCRE_EXP_DEFN PCRE_EXP_DECL
146 # endif
147 # ifndef PCRE_EXP_DATA_DEFN
148 # define PCRE_EXP_DATA_DEFN
149 # endif
150 # endif
151 #endif
152
153 /* When compiling with the MSVC compiler, it is sometimes necessary to include
154 a "calling convention" before exported function names. (This is secondhand
155 information; I know nothing about MSVC myself). For example, something like
156
157 void __cdecl function(....)
158
159 might be needed. In order so make this easy, all the exported functions have
160 PCRE_CALL_CONVENTION just before their names. It is rarely needed; if not
161 set, we ensure here that it has no effect. */
162
163 #ifndef PCRE_CALL_CONVENTION
164 #define PCRE_CALL_CONVENTION
165 #endif
166
167 /* We need to have types that specify unsigned 8, 16 and 32-bit integers. We
168 cannot determine these outside the compilation (e.g. by running a program as
169 part of "configure") because PCRE is often cross-compiled for use on other
170 systems. Instead we make use of the maximum sizes that are available at
171 preprocessor time in standard C environments. */
172
173 typedef unsigned char pcre_uint8;
174
175 #if USHRT_MAX == 65535
176 typedef unsigned short pcre_uint16;
177 typedef short pcre_int16;
178 #elif UINT_MAX == 65535
179 typedef unsigned int pcre_uint16;
180 typedef int pcre_int16;
181 #else
182 #error Cannot determine a type for 16-bit unsigned integers
183 #endif
184
185 #if UINT_MAX == 4294967295
186 typedef unsigned int pcre_uint32;
187 typedef int pcre_int32;
188 #elif ULONG_MAX == 4294967295
189 typedef unsigned long int pcre_uint32;
190 typedef long int pcre_int32;
191 #else
192 #error Cannot determine a type for 32-bit unsigned integers
193 #endif
194
195 /* When checking for integer overflow in pcre_compile(), we need to handle
196 large integers. If a 64-bit integer type is available, we can use that.
197 Otherwise we have to cast to double, which of course requires floating point
198 arithmetic. Handle this by defining a macro for the appropriate type. If
199 stdint.h is available, include it; it may define INT64_MAX. Systems that do not
200 have stdint.h (e.g. Solaris) may have inttypes.h. The macro int64_t may be set
201 by "configure". */
202
203 #if HAVE_STDINT_H
204 #include <stdint.h>
205 #elif HAVE_INTTYPES_H
206 #include <inttypes.h>
207 #endif
208
209 #if defined INT64_MAX || defined int64_t
210 #define INT64_OR_DOUBLE int64_t
211 #else
212 #define INT64_OR_DOUBLE double
213 #endif
214
215 /* All character handling must be done as unsigned characters. Otherwise there
216 are problems with top-bit-set characters and functions such as isspace().
217 However, we leave the interface to the outside world as char * or short *,
218 because that should make things easier for callers. We define a short type
219 for the current character representation (either 8 or 16 bit) to save lots
220 of typing. I tried "uchar", but it causes problems on Digital Unix, where
221 it is defined in sys/types, so use "uschar" instead. */
222
223 #ifdef COMPILE_PCRE8
224
225 typedef unsigned char pcre_uchar;
226 #define IN_UCHARS(x) (x)
227
228 #else
229
230 #ifdef COMPILE_PCRE16
231 #if USHRT_MAX != 65535
232 /* This is a warning message. Change PCRE_SCHAR16 to a 16 bit data type in
233 pcre.h(.in) and disable (comment out) this message. */
234 #error Warning: PCRE_SCHAR16 is not a 16 bit data type.
235 #endif
236 typedef pcre_uint16 pcre_uchar;
237 #define IN_UCHARS(x) ((x) << 1)
238
239 #else
240 #error Unsupported compiling mode
241 #endif /* COMPILE_PCRE16 */
242
243 #endif /* COMPILE_PCRE8 */
244
245 /* This is an unsigned int value that no character can ever have. UTF-8
246 characters only go up to 0x7fffffff (though Unicode doesn't go beyond
247 0x0010ffff). */
248
249 #define NOTACHAR 0xffffffff
250
251 /* PCRE is able to support several different kinds of newline (CR, LF, CRLF,
252 "any" and "anycrlf" at present). The following macros are used to package up
253 testing for newlines. NLBLOCK, PSSTART, and PSEND are defined in the various
254 modules to indicate in which datablock the parameters exist, and what the
255 start/end of string field names are. */
256
257 #define NLTYPE_FIXED 0 /* Newline is a fixed length string */
258 #define NLTYPE_ANY 1 /* Newline is any Unicode line ending */
259 #define NLTYPE_ANYCRLF 2 /* Newline is CR, LF, or CRLF */
260
261 /* This macro checks for a newline at the given position */
262
263 #define IS_NEWLINE(p) \
264 ((NLBLOCK->nltype != NLTYPE_FIXED)? \
265 ((p) < NLBLOCK->PSEND && \
266 PRIV(is_newline)((p), NLBLOCK->nltype, NLBLOCK->PSEND, &(NLBLOCK->nllen),\
267 utf8)) \
268 : \
269 ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \
270 (p)[0] == NLBLOCK->nl[0] && \
271 (NLBLOCK->nllen == 1 || (p)[1] == NLBLOCK->nl[1]) \
272 ) \
273 )
274
275 /* This macro checks for a newline immediately preceding the given position */
276
277 #define WAS_NEWLINE(p) \
278 ((NLBLOCK->nltype != NLTYPE_FIXED)? \
279 ((p) > NLBLOCK->PSSTART && \
280 PRIV(was_newline)((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \
281 &(NLBLOCK->nllen), utf8)) \
282 : \
283 ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \
284 (p)[-NLBLOCK->nllen] == NLBLOCK->nl[0] && \
285 (NLBLOCK->nllen == 1 || (p)[-NLBLOCK->nllen+1] == NLBLOCK->nl[1]) \
286 ) \
287 )
288
289 /* When PCRE is compiled as a C++ library, the subject pointer can be replaced
290 with a custom type. This makes it possible, for example, to allow pcre_exec()
291 to process subject strings that are discontinuous by using a smart pointer
292 class. It must always be possible to inspect all of the subject string in
293 pcre_exec() because of the way it backtracks. Two macros are required in the
294 normal case, for sign-unspecified and unsigned char pointers. The former is
295 used for the external interface and appears in pcre.h, which is why its name
296 must begin with PCRE_. */
297
298 #ifdef CUSTOM_SUBJECT_PTR
299 #define PCRE_PUCHAR CUSTOM_SUBJECT_PTR
300 #else
301 #define PCRE_PUCHAR const pcre_uchar *
302 #endif
303
304 /* Include the public PCRE header and the definitions of UCP character property
305 values. */
306
307 #include "pcre.h"
308 #include "ucp.h"
309
310 /* When compiling for use with the Virtual Pascal compiler, these functions
311 need to have their names changed. PCRE must be compiled with the -DVPCOMPAT
312 option on the command line. */
313
314 #ifdef VPCOMPAT
315 #define strlen(s) _strlen(s)
316 #define strncmp(s1,s2,m) _strncmp(s1,s2,m)
317 #define memcmp(s,c,n) _memcmp(s,c,n)
318 #define memcpy(d,s,n) _memcpy(d,s,n)
319 #define memmove(d,s,n) _memmove(d,s,n)
320 #define memset(s,c,n) _memset(s,c,n)
321 #else /* VPCOMPAT */
322
323 /* To cope with SunOS4 and other systems that lack memmove() but have bcopy(),
324 define a macro for memmove() if HAVE_MEMMOVE is false, provided that HAVE_BCOPY
325 is set. Otherwise, include an emulating function for those systems that have
326 neither (there some non-Unix environments where this is the case). */
327
328 #ifndef HAVE_MEMMOVE
329 #undef memmove /* some systems may have a macro */
330 #ifdef HAVE_BCOPY
331 #define memmove(a, b, c) bcopy(b, a, c)
332 #else /* HAVE_BCOPY */
333 static void *
334 pcre_memmove(void *d, const void *s, size_t n)
335 {
336 size_t i;
337 unsigned char *dest = (unsigned char *)d;
338 const unsigned char *src = (const unsigned char *)s;
339 if (dest > src)
340 {
341 dest += n;
342 src += n;
343 for (i = 0; i < n; ++i) *(--dest) = *(--src);
344 return (void *)dest;
345 }
346 else
347 {
348 for (i = 0; i < n; ++i) *dest++ = *src++;
349 return (void *)(dest - n);
350 }
351 }
352 #define memmove(a, b, c) pcre_memmove(a, b, c)
353 #endif /* not HAVE_BCOPY */
354 #endif /* not HAVE_MEMMOVE */
355 #endif /* not VPCOMPAT */
356
357
358 /* PCRE keeps offsets in its compiled code as 2-byte quantities (always stored
359 in big-endian order) by default. These are used, for example, to link from the
360 start of a subpattern to its alternatives and its end. The use of 2 bytes per
361 offset limits the size of the compiled regex to around 64K, which is big enough
362 for almost everybody. However, I received a request for an even bigger limit.
363 For this reason, and also to make the code easier to maintain, the storing and
364 loading of offsets from the byte string is now handled by the macros that are
365 defined here.
366
367 The macros are controlled by the value of LINK_SIZE. This defaults to 2 in
368 the config.h file, but can be overridden by using -D on the command line. This
369 is automated on Unix systems via the "configure" command. */
370
371 #ifdef COMPILE_PCRE8
372
373 #if LINK_SIZE == 2
374
375 #define PUT(a,n,d) \
376 (a[n] = (d) >> 8), \
377 (a[(n)+1] = (d) & 255)
378
379 #define GET(a,n) \
380 (((a)[n] << 8) | (a)[(n)+1])
381
382 #define MAX_PATTERN_SIZE (1 << 16)
383
384
385 #elif LINK_SIZE == 3
386
387 #define PUT(a,n,d) \
388 (a[n] = (d) >> 16), \
389 (a[(n)+1] = (d) >> 8), \
390 (a[(n)+2] = (d) & 255)
391
392 #define GET(a,n) \
393 (((a)[n] << 16) | ((a)[(n)+1] << 8) | (a)[(n)+2])
394
395 #define MAX_PATTERN_SIZE (1 << 24)
396
397
398 #elif LINK_SIZE == 4
399
400 #define PUT(a,n,d) \
401 (a[n] = (d) >> 24), \
402 (a[(n)+1] = (d) >> 16), \
403 (a[(n)+2] = (d) >> 8), \
404 (a[(n)+3] = (d) & 255)
405
406 #define GET(a,n) \
407 (((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3])
408
409 /* Keep it positive */
410 #define MAX_PATTERN_SIZE (1 << 30)
411
412 #else
413 #error LINK_SIZE must be either 2, 3, or 4
414 #endif
415
416 #else /* COMPILE_PCRE8 */
417
418 #ifdef COMPILE_PCRE16
419
420 #if LINK_SIZE == 2
421
422 #undef LINK_SIZE
423 #define LINK_SIZE 1
424
425 #define PUT(a,n,d) \
426 (a[n] = (d))
427
428 #define GET(a,n) \
429 (a[n])
430
431 #define MAX_PATTERN_SIZE (1 << 16)
432
433 #elif LINK_SIZE == 3 || LINK_SIZE == 4
434
435 #undef LINK_SIZE
436 #define LINK_SIZE 2
437
438 #define PUT(a,n,d) \
439 (a[n] = (d) >> 16), \
440 (a[(n)+1] = (d) & 65536)
441
442 #define GET(a,n) \
443 (((a)[n] << 16) | (a)[(n)+1])
444
445 /* Keep it positive */
446 #define MAX_PATTERN_SIZE (1 << 30)
447
448 #else
449 #error LINK_SIZE must be either 2, 3, or 4
450 #endif
451
452 #else
453 #error Unsupported compiling mode
454 #endif /* COMPILE_PCRE16 */
455
456 #endif /* COMPILE_PCRE8 */
457
458 /* Convenience macro defined in terms of the others */
459
460 #define PUTINC(a,n,d) PUT(a,n,d), a += LINK_SIZE
461
462
463 /* PCRE uses some other 2-byte quantities that do not change when the size of
464 offsets changes. There are used for repeat counts and for other things such as
465 capturing parenthesis numbers in back references. */
466
467 #ifdef COMPILE_PCRE8
468
469 #define IMM2_SIZE 2
470
471 #define PUT2(a,n,d) \
472 a[n] = (d) >> 8; \
473 a[(n)+1] = (d) & 255
474
475 #define GET2(a,n) \
476 (((a)[n] << 8) | (a)[(n)+1])
477
478 #else /* COMPILE_PCRE8 */
479
480 #ifdef COMPILE_PCRE16
481
482 #define IMM2_SIZE 1
483
484 #define PUT2(a,n,d) \
485 a[n] = d
486
487 #define GET2(a,n) \
488 a[n]
489
490 #else
491 #error Unsupported compiling mode
492 #endif /* COMPILE_PCRE16 */
493
494 #endif /* COMPILE_PCRE8 */
495
496 #define PUT2INC(a,n,d) PUT2(a,n,d), a += IMM2_SIZE
497
498 /* When UTF-8 encoding is being used, a character is no longer just a single
499 byte. The macros for character handling generate simple sequences when used in
500 byte-mode, and more complicated ones for UTF-8 characters. GETCHARLENTEST is
501 not used when UTF-8 is not supported, so it is not defined, and BACKCHAR should
502 never be called in byte mode. To make sure they can never even appear when
503 UTF-8 support is omitted, we don't even define them. */
504
505 #ifndef SUPPORT_UTF8
506 #define GETCHAR(c, eptr) c = *eptr;
507 #define GETCHARTEST(c, eptr) c = *eptr;
508 #define GETCHARINC(c, eptr) c = *eptr++;
509 #define GETCHARINCTEST(c, eptr) c = *eptr++;
510 #define GETCHARLEN(c, eptr, len) c = *eptr;
511 /* #define GETCHARLENTEST(c, eptr, len) */
512 /* #define BACKCHAR(eptr) */
513
514 #else /* SUPPORT_UTF8 */
515
516 /* These macros were originally written in the form of loops that used data
517 from the tables whose names start with PRIV(utf8_table). They were rewritten by
518 a user so as not to use loops, because in some environments this gives a
519 significant performance advantage, and it seems never to do any harm. */
520
521 /* Base macro to pick up the remaining bytes of a UTF-8 character, not
522 advancing the pointer. */
523
524 #define GETUTF8(c, eptr) \
525 { \
526 if ((c & 0x20) == 0) \
527 c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
528 else if ((c & 0x10) == 0) \
529 c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
530 else if ((c & 0x08) == 0) \
531 c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
532 ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
533 else if ((c & 0x04) == 0) \
534 c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
535 ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
536 (eptr[4] & 0x3f); \
537 else \
538 c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
539 ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
540 ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
541 }
542
543 /* Get the next UTF-8 character, not advancing the pointer. This is called when
544 we know we are in UTF-8 mode. */
545
546 #define GETCHAR(c, eptr) \
547 c = *eptr; \
548 if (c >= 0xc0) GETUTF8(c, eptr);
549
550 /* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the
551 pointer. */
552
553 #define GETCHARTEST(c, eptr) \
554 c = *eptr; \
555 if (utf8 && c >= 0xc0) GETUTF8(c, eptr);
556
557 /* Base macro to pick up the remaining bytes of a UTF-8 character, advancing
558 the pointer. */
559
560 #define GETUTF8INC(c, eptr) \
561 { \
562 if ((c & 0x20) == 0) \
563 c = ((c & 0x1f) << 6) | (*eptr++ & 0x3f); \
564 else if ((c & 0x10) == 0) \
565 { \
566 c = ((c & 0x0f) << 12) | ((*eptr & 0x3f) << 6) | (eptr[1] & 0x3f); \
567 eptr += 2; \
568 } \
569 else if ((c & 0x08) == 0) \
570 { \
571 c = ((c & 0x07) << 18) | ((*eptr & 0x3f) << 12) | \
572 ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
573 eptr += 3; \
574 } \
575 else if ((c & 0x04) == 0) \
576 { \
577 c = ((c & 0x03) << 24) | ((*eptr & 0x3f) << 18) | \
578 ((eptr[1] & 0x3f) << 12) | ((eptr[2] & 0x3f) << 6) | \
579 (eptr[3] & 0x3f); \
580 eptr += 4; \
581 } \
582 else \
583 { \
584 c = ((c & 0x01) << 30) | ((*eptr & 0x3f) << 24) | \
585 ((eptr[1] & 0x3f) << 18) | ((eptr[2] & 0x3f) << 12) | \
586 ((eptr[3] & 0x3f) << 6) | (eptr[4] & 0x3f); \
587 eptr += 5; \
588 } \
589 }
590
591 /* Get the next UTF-8 character, advancing the pointer. This is called when we
592 know we are in UTF-8 mode. */
593
594 #define GETCHARINC(c, eptr) \
595 c = *eptr++; \
596 if (c >= 0xc0) GETUTF8INC(c, eptr);
597
598 /* Get the next character, testing for UTF-8 mode, and advancing the pointer.
599 This is called when we don't know if we are in UTF-8 mode. */
600
601 #define GETCHARINCTEST(c, eptr) \
602 c = *eptr++; \
603 if (utf8 && c >= 0xc0) GETUTF8INC(c, eptr);
604
605 /* Base macro to pick up the remaining bytes of a UTF-8 character, not
606 advancing the pointer, incrementing the length. */
607
608 #define GETUTF8LEN(c, eptr, len) \
609 { \
610 if ((c & 0x20) == 0) \
611 { \
612 c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
613 len++; \
614 } \
615 else if ((c & 0x10) == 0) \
616 { \
617 c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
618 len += 2; \
619 } \
620 else if ((c & 0x08) == 0) \
621 {\
622 c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
623 ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
624 len += 3; \
625 } \
626 else if ((c & 0x04) == 0) \
627 { \
628 c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
629 ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
630 (eptr[4] & 0x3f); \
631 len += 4; \
632 } \
633 else \
634 {\
635 c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
636 ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
637 ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
638 len += 5; \
639 } \
640 }
641
642 /* Get the next UTF-8 character, not advancing the pointer, incrementing length
643 if there are extra bytes. This is called when we know we are in UTF-8 mode. */
644
645 #define GETCHARLEN(c, eptr, len) \
646 c = *eptr; \
647 if (c >= 0xc0) GETUTF8LEN(c, eptr, len);
648
649 /* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the
650 pointer, incrementing length if there are extra bytes. This is called when we
651 do not know if we are in UTF-8 mode. */
652
653 #define GETCHARLENTEST(c, eptr, len) \
654 c = *eptr; \
655 if (utf8 && c >= 0xc0) GETUTF8LEN(c, eptr, len);
656
657 /* If the pointer is not at the start of a character, move it back until
658 it is. This is called only in UTF-8 mode - we don't put a test within the macro
659 because almost all calls are already within a block of UTF-8 only code. */
660
661 #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--
662
663 #endif /* SUPPORT_UTF8 */
664
665
666 /* In case there is no definition of offsetof() provided - though any proper
667 Standard C system should have one. */
668
669 #ifndef offsetof
670 #define offsetof(p_type,field) ((size_t)&(((p_type *)0)->field))
671 #endif
672
673
674 /* Private flags containing information about the compiled regex. They used to
675 live at the top end of the options word, but that got almost full, so now they
676 are in a 16-bit flags word. From release 8.00, PCRE_NOPARTIAL is unused, as
677 the restrictions on partial matching have been lifted. It remains for backwards
678 compatibility. */
679
680 #define PCRE_NOPARTIAL 0x0001 /* can't use partial with this regex */
681 #define PCRE_FIRSTSET 0x0002 /* first_byte is set */
682 #define PCRE_REQCHSET 0x0004 /* req_byte is set */
683 #define PCRE_STARTLINE 0x0008 /* start after \n for multiline */
684 #define PCRE_JCHANGED 0x0010 /* j option used in regex */
685 #define PCRE_HASCRORLF 0x0020 /* explicit \r or \n in pattern */
686 #define PCRE_HASTHEN 0x0040 /* pattern contains (*THEN) */
687
688 /* Flags for the "extra" block produced by pcre_study(). */
689
690 #define PCRE_STUDY_MAPPED 0x0001 /* a map of starting chars exists */
691 #define PCRE_STUDY_MINLEN 0x0002 /* a minimum length field exists */
692
693 /* Masks for identifying the public options that are permitted at compile
694 time, run time, or study time, respectively. */
695
696 #define PCRE_NEWLINE_BITS (PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_ANY| \
697 PCRE_NEWLINE_ANYCRLF)
698
699 #define PUBLIC_COMPILE_OPTIONS \
700 (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \
701 PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \
702 PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \
703 PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
704 PCRE_JAVASCRIPT_COMPAT|PCRE_UCP|PCRE_NO_START_OPTIMIZE)
705
706 #define PUBLIC_EXEC_OPTIONS \
707 (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
708 PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_NEWLINE_BITS| \
709 PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE|PCRE_NO_START_OPTIMIZE)
710
711 #define PUBLIC_DFA_EXEC_OPTIONS \
712 (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
713 PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_DFA_SHORTEST| \
714 PCRE_DFA_RESTART|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
715 PCRE_NO_START_OPTIMIZE)
716
717 #define PUBLIC_STUDY_OPTIONS \
718 PCRE_STUDY_JIT_COMPILE
719
720 /* Magic number to provide a small check against being handed junk. Also used
721 to detect whether a pattern was compiled on a host of different endianness. */
722
723 #define MAGIC_NUMBER 0x50435245UL /* 'PCRE' */
724
725 /* Negative values for the firstchar and reqchar variables */
726
727 #define REQ_UNSET (-2)
728 #define REQ_NONE (-1)
729
730 /* The maximum remaining length of subject we are prepared to search for a
731 req_byte match. */
732
733 #define REQ_BYTE_MAX 1000
734
735 /* Flags added to firstbyte or reqbyte; a "non-literal" item is either a
736 variable-length repeat, or a anything other than literal characters. */
737
738 #define REQ_CASELESS 0x0100 /* indicates caselessness */
739 #define REQ_VARY 0x0200 /* reqbyte followed non-literal item */
740
741 /* Miscellaneous definitions. The #ifndef is to pacify compiler warnings in
742 environments where these macros are defined elsewhere. Unfortunately, there
743 is no way to do the same for the typedef. */
744
745 typedef int BOOL;
746
747 #ifndef FALSE
748 #define FALSE 0
749 #define TRUE 1
750 #endif
751
752 /* If PCRE is to support UTF-8 on EBCDIC platforms, we cannot use normal
753 character constants like '*' because the compiler would emit their EBCDIC code,
754 which is different from their ASCII/UTF-8 code. Instead we define macros for
755 the characters so that they always use the ASCII/UTF-8 code when UTF-8 support
756 is enabled. When UTF-8 support is not enabled, the definitions use character
757 literals. Both character and string versions of each character are needed, and
758 there are some longer strings as well.
759
760 This means that, on EBCDIC platforms, the PCRE library can handle either
761 EBCDIC, or UTF-8, but not both. To support both in the same compiled library
762 would need different lookups depending on whether PCRE_UTF8 was set or not.
763 This would make it impossible to use characters in switch/case statements,
764 which would reduce performance. For a theoretical use (which nobody has asked
765 for) in a minority area (EBCDIC platforms), this is not sensible. Any
766 application that did need both could compile two versions of the library, using
767 macros to give the functions distinct names. */
768
769 #ifndef SUPPORT_UTF8
770
771 /* UTF-8 support is not enabled; use the platform-dependent character literals
772 so that PCRE works on both ASCII and EBCDIC platforms, in non-UTF-mode only. */
773
774 #define CHAR_HT '\t'
775 #define CHAR_VT '\v'
776 #define CHAR_FF '\f'
777 #define CHAR_CR '\r'
778 #define CHAR_NL '\n'
779 #define CHAR_BS '\b'
780 #define CHAR_BEL '\a'
781 #ifdef EBCDIC
782 #define CHAR_ESC '\047'
783 #define CHAR_DEL '\007'
784 #else
785 #define CHAR_ESC '\033'
786 #define CHAR_DEL '\177'
787 #endif
788
789 #define CHAR_SPACE ' '
790 #define CHAR_EXCLAMATION_MARK '!'
791 #define CHAR_QUOTATION_MARK '"'
792 #define CHAR_NUMBER_SIGN '#'
793 #define CHAR_DOLLAR_SIGN '$'
794 #define CHAR_PERCENT_SIGN '%'
795 #define CHAR_AMPERSAND '&'
796 #define CHAR_APOSTROPHE '\''
797 #define CHAR_LEFT_PARENTHESIS '('
798 #define CHAR_RIGHT_PARENTHESIS ')'
799 #define CHAR_ASTERISK '*'
800 #define CHAR_PLUS '+'
801 #define CHAR_COMMA ','
802 #define CHAR_MINUS '-'
803 #define CHAR_DOT '.'
804 #define CHAR_SLASH '/'
805 #define CHAR_0 '0'
806 #define CHAR_1 '1'
807 #define CHAR_2 '2'
808 #define CHAR_3 '3'
809 #define CHAR_4 '4'
810 #define CHAR_5 '5'
811 #define CHAR_6 '6'
812 #define CHAR_7 '7'
813 #define CHAR_8 '8'
814 #define CHAR_9 '9'
815 #define CHAR_COLON ':'
816 #define CHAR_SEMICOLON ';'
817 #define CHAR_LESS_THAN_SIGN '<'
818 #define CHAR_EQUALS_SIGN '='
819 #define CHAR_GREATER_THAN_SIGN '>'
820 #define CHAR_QUESTION_MARK '?'
821 #define CHAR_COMMERCIAL_AT '@'
822 #define CHAR_A 'A'
823 #define CHAR_B 'B'
824 #define CHAR_C 'C'
825 #define CHAR_D 'D'
826 #define CHAR_E 'E'
827 #define CHAR_F 'F'
828 #define CHAR_G 'G'
829 #define CHAR_H 'H'
830 #define CHAR_I 'I'
831 #define CHAR_J 'J'
832 #define CHAR_K 'K'
833 #define CHAR_L 'L'
834 #define CHAR_M 'M'
835 #define CHAR_N 'N'
836 #define CHAR_O 'O'
837 #define CHAR_P 'P'
838 #define CHAR_Q 'Q'
839 #define CHAR_R 'R'
840 #define CHAR_S 'S'
841 #define CHAR_T 'T'
842 #define CHAR_U 'U'
843 #define CHAR_V 'V'
844 #define CHAR_W 'W'
845 #define CHAR_X 'X'
846 #define CHAR_Y 'Y'
847 #define CHAR_Z 'Z'
848 #define CHAR_LEFT_SQUARE_BRACKET '['
849 #define CHAR_BACKSLASH '\\'
850 #define CHAR_RIGHT_SQUARE_BRACKET ']'
851 #define CHAR_CIRCUMFLEX_ACCENT '^'
852 #define CHAR_UNDERSCORE '_'
853 #define CHAR_GRAVE_ACCENT '`'
854 #define CHAR_a 'a'
855 #define CHAR_b 'b'
856 #define CHAR_c 'c'
857 #define CHAR_d 'd'
858 #define CHAR_e 'e'
859 #define CHAR_f 'f'
860 #define CHAR_g 'g'
861 #define CHAR_h 'h'
862 #define CHAR_i 'i'
863 #define CHAR_j 'j'
864 #define CHAR_k 'k'
865 #define CHAR_l 'l'
866 #define CHAR_m 'm'
867 #define CHAR_n 'n'
868 #define CHAR_o 'o'
869 #define CHAR_p 'p'
870 #define CHAR_q 'q'
871 #define CHAR_r 'r'
872 #define CHAR_s 's'
873 #define CHAR_t 't'
874 #define CHAR_u 'u'
875 #define CHAR_v 'v'
876 #define CHAR_w 'w'
877 #define CHAR_x 'x'
878 #define CHAR_y 'y'
879 #define CHAR_z 'z'
880 #define CHAR_LEFT_CURLY_BRACKET '{'
881 #define CHAR_VERTICAL_LINE '|'
882 #define CHAR_RIGHT_CURLY_BRACKET '}'
883 #define CHAR_TILDE '~'
884
885 #define STR_HT "\t"
886 #define STR_VT "\v"
887 #define STR_FF "\f"
888 #define STR_CR "\r"
889 #define STR_NL "\n"
890 #define STR_BS "\b"
891 #define STR_BEL "\a"
892 #ifdef EBCDIC
893 #define STR_ESC "\047"
894 #define STR_DEL "\007"
895 #else
896 #define STR_ESC "\033"
897 #define STR_DEL "\177"
898 #endif
899
900 #define STR_SPACE " "
901 #define STR_EXCLAMATION_MARK "!"
902 #define STR_QUOTATION_MARK "\""
903 #define STR_NUMBER_SIGN "#"
904 #define STR_DOLLAR_SIGN "$"
905 #define STR_PERCENT_SIGN "%"
906 #define STR_AMPERSAND "&"
907 #define STR_APOSTROPHE "'"
908 #define STR_LEFT_PARENTHESIS "("
909 #define STR_RIGHT_PARENTHESIS ")"
910 #define STR_ASTERISK "*"
911 #define STR_PLUS "+"
912 #define STR_COMMA ","
913 #define STR_MINUS "-"
914 #define STR_DOT "."
915 #define STR_SLASH "/"
916 #define STR_0 "0"
917 #define STR_1 "1"
918 #define STR_2 "2"
919 #define STR_3 "3"
920 #define STR_4 "4"
921 #define STR_5 "5"
922 #define STR_6 "6"
923 #define STR_7 "7"
924 #define STR_8 "8"
925 #define STR_9 "9"
926 #define STR_COLON ":"
927 #define STR_SEMICOLON ";"
928 #define STR_LESS_THAN_SIGN "<"
929 #define STR_EQUALS_SIGN "="
930 #define STR_GREATER_THAN_SIGN ">"
931 #define STR_QUESTION_MARK "?"
932 #define STR_COMMERCIAL_AT "@"
933 #define STR_A "A"
934 #define STR_B "B"
935 #define STR_C "C"
936 #define STR_D "D"
937 #define STR_E "E"
938 #define STR_F "F"
939 #define STR_G "G"
940 #define STR_H "H"
941 #define STR_I "I"
942 #define STR_J "J"
943 #define STR_K "K"
944 #define STR_L "L"
945 #define STR_M "M"
946 #define STR_N "N"
947 #define STR_O "O"
948 #define STR_P "P"
949 #define STR_Q "Q"
950 #define STR_R "R"
951 #define STR_S "S"
952 #define STR_T "T"
953 #define STR_U "U"
954 #define STR_V "V"
955 #define STR_W "W"
956 #define STR_X "X"
957 #define STR_Y "Y"
958 #define STR_Z "Z"
959 #define STR_LEFT_SQUARE_BRACKET "["
960 #define STR_BACKSLASH "\\"
961 #define STR_RIGHT_SQUARE_BRACKET "]"
962 #define STR_CIRCUMFLEX_ACCENT "^"
963 #define STR_UNDERSCORE "_"
964 #define STR_GRAVE_ACCENT "`"
965 #define STR_a "a"
966 #define STR_b "b"
967 #define STR_c "c"
968 #define STR_d "d"
969 #define STR_e "e"
970 #define STR_f "f"
971 #define STR_g "g"
972 #define STR_h "h"
973 #define STR_i "i"
974 #define STR_j "j"
975 #define STR_k "k"
976 #define STR_l "l"
977 #define STR_m "m"
978 #define STR_n "n"
979 #define STR_o "o"
980 #define STR_p "p"
981 #define STR_q "q"
982 #define STR_r "r"
983 #define STR_s "s"
984 #define STR_t "t"
985 #define STR_u "u"
986 #define STR_v "v"
987 #define STR_w "w"
988 #define STR_x "x"
989 #define STR_y "y"
990 #define STR_z "z"
991 #define STR_LEFT_CURLY_BRACKET "{"
992 #define STR_VERTICAL_LINE "|"
993 #define STR_RIGHT_CURLY_BRACKET "}"
994 #define STR_TILDE "~"
995
996 #define STRING_ACCEPT0 "ACCEPT\0"
997 #define STRING_COMMIT0 "COMMIT\0"
998 #define STRING_F0 "F\0"
999 #define STRING_FAIL0 "FAIL\0"
1000 #define STRING_MARK0 "MARK\0"
1001 #define STRING_PRUNE0 "PRUNE\0"
1002 #define STRING_SKIP0 "SKIP\0"
1003 #define STRING_THEN "THEN"
1004
1005 #define STRING_alpha0 "alpha\0"
1006 #define STRING_lower0 "lower\0"
1007 #define STRING_upper0 "upper\0"
1008 #define STRING_alnum0 "alnum\0"
1009 #define STRING_ascii0 "ascii\0"
1010 #define STRING_blank0 "blank\0"
1011 #define STRING_cntrl0 "cntrl\0"
1012 #define STRING_digit0 "digit\0"
1013 #define STRING_graph0 "graph\0"
1014 #define STRING_print0 "print\0"
1015 #define STRING_punct0 "punct\0"
1016 #define STRING_space0 "space\0"
1017 #define STRING_word0 "word\0"
1018 #define STRING_xdigit "xdigit"
1019
1020 #define STRING_DEFINE "DEFINE"
1021
1022 #define STRING_CR_RIGHTPAR "CR)"
1023 #define STRING_LF_RIGHTPAR "LF)"
1024 #define STRING_CRLF_RIGHTPAR "CRLF)"
1025 #define STRING_ANY_RIGHTPAR "ANY)"
1026 #define STRING_ANYCRLF_RIGHTPAR "ANYCRLF)"
1027 #define STRING_BSR_ANYCRLF_RIGHTPAR "BSR_ANYCRLF)"
1028 #define STRING_BSR_UNICODE_RIGHTPAR "BSR_UNICODE)"
1029 #define STRING_UTF8_RIGHTPAR "UTF8)"
1030 #define STRING_UCP_RIGHTPAR "UCP)"
1031 #define STRING_NO_START_OPT_RIGHTPAR "NO_START_OPT)"
1032
1033 #else /* SUPPORT_UTF8 */
1034
1035 /* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This
1036 works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode
1037 only. */
1038
1039 #define CHAR_HT '\011'
1040 #define CHAR_VT '\013'
1041 #define CHAR_FF '\014'
1042 #define CHAR_CR '\015'
1043 #define CHAR_NL '\012'
1044 #define CHAR_BS '\010'
1045 #define CHAR_BEL '\007'
1046 #define CHAR_ESC '\033'
1047 #define CHAR_DEL '\177'
1048
1049 #define CHAR_SPACE '\040'
1050 #define CHAR_EXCLAMATION_MARK '\041'
1051 #define CHAR_QUOTATION_MARK '\042'
1052 #define CHAR_NUMBER_SIGN '\043'
1053 #define CHAR_DOLLAR_SIGN '\044'
1054 #define CHAR_PERCENT_SIGN '\045'
1055 #define CHAR_AMPERSAND '\046'
1056 #define CHAR_APOSTROPHE '\047'
1057 #define CHAR_LEFT_PARENTHESIS '\050'
1058 #define CHAR_RIGHT_PARENTHESIS '\051'
1059 #define CHAR_ASTERISK '\052'
1060 #define CHAR_PLUS '\053'
1061 #define CHAR_COMMA '\054'
1062 #define CHAR_MINUS '\055'
1063 #define CHAR_DOT '\056'
1064 #define CHAR_SLASH '\057'
1065 #define CHAR_0 '\060'
1066 #define CHAR_1 '\061'
1067 #define CHAR_2 '\062'
1068 #define CHAR_3 '\063'
1069 #define CHAR_4 '\064'
1070 #define CHAR_5 '\065'
1071 #define CHAR_6 '\066'
1072 #define CHAR_7 '\067'
1073 #define CHAR_8 '\070'
1074 #define CHAR_9 '\071'
1075 #define CHAR_COLON '\072'
1076 #define CHAR_SEMICOLON '\073'
1077 #define CHAR_LESS_THAN_SIGN '\074'
1078 #define CHAR_EQUALS_SIGN '\075'
1079 #define CHAR_GREATER_THAN_SIGN '\076'
1080 #define CHAR_QUESTION_MARK '\077'
1081 #define CHAR_COMMERCIAL_AT '\100'
1082 #define CHAR_A '\101'
1083 #define CHAR_B '\102'
1084 #define CHAR_C '\103'
1085 #define CHAR_D '\104'
1086 #define CHAR_E '\105'
1087 #define CHAR_F '\106'
1088 #define CHAR_G '\107'
1089 #define CHAR_H '\110'
1090 #define CHAR_I '\111'
1091 #define CHAR_J '\112'
1092 #define CHAR_K '\113'
1093 #define CHAR_L '\114'
1094 #define CHAR_M '\115'
1095 #define CHAR_N '\116'
1096 #define CHAR_O '\117'
1097 #define CHAR_P '\120'
1098 #define CHAR_Q '\121'
1099 #define CHAR_R '\122'
1100 #define CHAR_S '\123'
1101 #define CHAR_T '\124'
1102 #define CHAR_U '\125'
1103 #define CHAR_V '\126'
1104 #define CHAR_W '\127'
1105 #define CHAR_X '\130'
1106 #define CHAR_Y '\131'
1107 #define CHAR_Z '\132'
1108 #define CHAR_LEFT_SQUARE_BRACKET '\133'
1109 #define CHAR_BACKSLASH '\134'
1110 #define CHAR_RIGHT_SQUARE_BRACKET '\135'
1111 #define CHAR_CIRCUMFLEX_ACCENT '\136'
1112 #define CHAR_UNDERSCORE '\137'
1113 #define CHAR_GRAVE_ACCENT '\140'
1114 #define CHAR_a '\141'
1115 #define CHAR_b '\142'
1116 #define CHAR_c '\143'
1117 #define CHAR_d '\144'
1118 #define CHAR_e '\145'
1119 #define CHAR_f '\146'
1120 #define CHAR_g '\147'
1121 #define CHAR_h '\150'
1122 #define CHAR_i '\151'
1123 #define CHAR_j '\152'
1124 #define CHAR_k '\153'
1125 #define CHAR_l '\154'
1126 #define CHAR_m '\155'
1127 #define CHAR_n '\156'
1128 #define CHAR_o '\157'
1129 #define CHAR_p '\160'
1130 #define CHAR_q '\161'
1131 #define CHAR_r '\162'
1132 #define CHAR_s '\163'
1133 #define CHAR_t '\164'
1134 #define CHAR_u '\165'
1135 #define CHAR_v '\166'
1136 #define CHAR_w '\167'
1137 #define CHAR_x '\170'
1138 #define CHAR_y '\171'
1139 #define CHAR_z '\172'
1140 #define CHAR_LEFT_CURLY_BRACKET '\173'
1141 #define CHAR_VERTICAL_LINE '\174'
1142 #define CHAR_RIGHT_CURLY_BRACKET '\175'
1143 #define CHAR_TILDE '\176'
1144
1145 #define STR_HT "\011"
1146 #define STR_VT "\013"
1147 #define STR_FF "\014"
1148 #define STR_CR "\015"
1149 #define STR_NL "\012"
1150 #define STR_BS "\010"
1151 #define STR_BEL "\007"
1152 #define STR_ESC "\033"
1153 #define STR_DEL "\177"
1154
1155 #define STR_SPACE "\040"
1156 #define STR_EXCLAMATION_MARK "\041"
1157 #define STR_QUOTATION_MARK "\042"
1158 #define STR_NUMBER_SIGN "\043"
1159 #define STR_DOLLAR_SIGN "\044"
1160 #define STR_PERCENT_SIGN "\045"
1161 #define STR_AMPERSAND "\046"
1162 #define STR_APOSTROPHE "\047"
1163 #define STR_LEFT_PARENTHESIS "\050"
1164 #define STR_RIGHT_PARENTHESIS "\051"
1165 #define STR_ASTERISK "\052"
1166 #define STR_PLUS "\053"
1167 #define STR_COMMA "\054"
1168 #define STR_MINUS "\055"
1169 #define STR_DOT "\056"
1170 #define STR_SLASH "\057"
1171 #define STR_0 "\060"
1172 #define STR_1 "\061"
1173 #define STR_2 "\062"
1174 #define STR_3 "\063"
1175 #define STR_4 "\064"
1176 #define STR_5 "\065"
1177 #define STR_6 "\066"
1178 #define STR_7 "\067"
1179 #define STR_8 "\070"
1180 #define STR_9 "\071"
1181 #define STR_COLON "\072"
1182 #define STR_SEMICOLON "\073"
1183 #define STR_LESS_THAN_SIGN "\074"
1184 #define STR_EQUALS_SIGN "\075"
1185 #define STR_GREATER_THAN_SIGN "\076"
1186 #define STR_QUESTION_MARK "\077"
1187 #define STR_COMMERCIAL_AT "\100"
1188 #define STR_A "\101"
1189 #define STR_B "\102"
1190 #define STR_C "\103"
1191 #define STR_D "\104"
1192 #define STR_E "\105"
1193 #define STR_F "\106"
1194 #define STR_G "\107"
1195 #define STR_H "\110"
1196 #define STR_I "\111"
1197 #define STR_J "\112"
1198 #define STR_K "\113"
1199 #define STR_L "\114"
1200 #define STR_M "\115"
1201 #define STR_N "\116"
1202 #define STR_O "\117"
1203 #define STR_P "\120"
1204 #define STR_Q "\121"
1205 #define STR_R "\122"
1206 #define STR_S "\123"
1207 #define STR_T "\124"
1208 #define STR_U "\125"
1209 #define STR_V "\126"
1210 #define STR_W "\127"
1211 #define STR_X "\130"
1212 #define STR_Y "\131"
1213 #define STR_Z "\132"
1214 #define STR_LEFT_SQUARE_BRACKET "\133"
1215 #define STR_BACKSLASH "\134"
1216 #define STR_RIGHT_SQUARE_BRACKET "\135"
1217 #define STR_CIRCUMFLEX_ACCENT "\136"
1218 #define STR_UNDERSCORE "\137"
1219 #define STR_GRAVE_ACCENT "\140"
1220 #define STR_a "\141"
1221 #define STR_b "\142"
1222 #define STR_c "\143"
1223 #define STR_d "\144"
1224 #define STR_e "\145"
1225 #define STR_f "\146"
1226 #define STR_g "\147"
1227 #define STR_h "\150"
1228 #define STR_i "\151"
1229 #define STR_j "\152"
1230 #define STR_k "\153"
1231 #define STR_l "\154"
1232 #define STR_m "\155"
1233 #define STR_n "\156"
1234 #define STR_o "\157"
1235 #define STR_p "\160"
1236 #define STR_q "\161"
1237 #define STR_r "\162"
1238 #define STR_s "\163"
1239 #define STR_t "\164"
1240 #define STR_u "\165"
1241 #define STR_v "\166"
1242 #define STR_w "\167"
1243 #define STR_x "\170"
1244 #define STR_y "\171"
1245 #define STR_z "\172"
1246 #define STR_LEFT_CURLY_BRACKET "\173"
1247 #define STR_VERTICAL_LINE "\174"
1248 #define STR_RIGHT_CURLY_BRACKET "\175"
1249 #define STR_TILDE "\176"
1250
1251 #define STRING_ACCEPT0 STR_A STR_C STR_C STR_E STR_P STR_T "\0"
1252 #define STRING_COMMIT0 STR_C STR_O STR_M STR_M STR_I STR_T "\0"
1253 #define STRING_F0 STR_F "\0"
1254 #define STRING_FAIL0 STR_F STR_A STR_I STR_L "\0"
1255 #define STRING_MARK0 STR_M STR_A STR_R STR_K "\0"
1256 #define STRING_PRUNE0 STR_P STR_R STR_U STR_N STR_E "\0"
1257 #define STRING_SKIP0 STR_S STR_K STR_I STR_P "\0"
1258 #define STRING_THEN STR_T STR_H STR_E STR_N
1259
1260 #define STRING_alpha0 STR_a STR_l STR_p STR_h STR_a "\0"
1261 #define STRING_lower0 STR_l STR_o STR_w STR_e STR_r "\0"
1262 #define STRING_upper0 STR_u STR_p STR_p STR_e STR_r "\0"
1263 #define STRING_alnum0 STR_a STR_l STR_n STR_u STR_m "\0"
1264 #define STRING_ascii0 STR_a STR_s STR_c STR_i STR_i "\0"
1265 #define STRING_blank0 STR_b STR_l STR_a STR_n STR_k "\0"
1266 #define STRING_cntrl0 STR_c STR_n STR_t STR_r STR_l "\0"
1267 #define STRING_digit0 STR_d STR_i STR_g STR_i STR_t "\0"
1268 #define STRING_graph0 STR_g STR_r STR_a STR_p STR_h "\0"
1269 #define STRING_print0 STR_p STR_r STR_i STR_n STR_t "\0"
1270 #define STRING_punct0 STR_p STR_u STR_n STR_c STR_t "\0"
1271 #define STRING_space0 STR_s STR_p STR_a STR_c STR_e "\0"
1272 #define STRING_word0 STR_w STR_o STR_r STR_d "\0"
1273 #define STRING_xdigit STR_x STR_d STR_i STR_g STR_i STR_t
1274
1275 #define STRING_DEFINE STR_D STR_E STR_F STR_I STR_N STR_E
1276
1277 #define STRING_CR_RIGHTPAR STR_C STR_R STR_RIGHT_PARENTHESIS
1278 #define STRING_LF_RIGHTPAR STR_L STR_F STR_RIGHT_PARENTHESIS
1279 #define STRING_CRLF_RIGHTPAR STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1280 #define STRING_ANY_RIGHTPAR STR_A STR_N STR_Y STR_RIGHT_PARENTHESIS
1281 #define STRING_ANYCRLF_RIGHTPAR STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1282 #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
1283 #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
1284 #define STRING_UTF8_RIGHTPAR STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS
1285 #define STRING_UCP_RIGHTPAR STR_U STR_C STR_P STR_RIGHT_PARENTHESIS
1286 #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
1287
1288 #endif /* SUPPORT_UTF8 */
1289
1290 /* Escape items that are just an encoding of a particular data value. */
1291
1292 #ifndef ESC_e
1293 #define ESC_e CHAR_ESC
1294 #endif
1295
1296 #ifndef ESC_f
1297 #define ESC_f CHAR_FF
1298 #endif
1299
1300 #ifndef ESC_n
1301 #define ESC_n CHAR_NL
1302 #endif
1303
1304 #ifndef ESC_r
1305 #define ESC_r CHAR_CR
1306 #endif
1307
1308 /* We can't officially use ESC_t because it is a POSIX reserved identifier
1309 (presumably because of all the others like size_t). */
1310
1311 #ifndef ESC_tee
1312 #define ESC_tee CHAR_HT
1313 #endif
1314
1315 /* Codes for different types of Unicode property */
1316
1317 #define PT_ANY 0 /* Any property - matches all chars */
1318 #define PT_LAMP 1 /* L& - the union of Lu, Ll, Lt */
1319 #define PT_GC 2 /* Specified general characteristic (e.g. L) */
1320 #define PT_PC 3 /* Specified particular characteristic (e.g. Lu) */
1321 #define PT_SC 4 /* Script (e.g. Han) */
1322 #define PT_ALNUM 5 /* Alphanumeric - the union of L and N */
1323 #define PT_SPACE 6 /* Perl space - Z plus 9,10,12,13 */
1324 #define PT_PXSPACE 7 /* POSIX space - Z plus 9,10,11,12,13 */
1325 #define PT_WORD 8 /* Word - L plus N plus underscore */
1326
1327 /* Flag bits and data types for the extended class (OP_XCLASS) for classes that
1328 contain UTF-8 characters with values greater than 255. */
1329
1330 #define XCL_NOT 0x01 /* Flag: this is a negative class */
1331 #define XCL_MAP 0x02 /* Flag: a 32-byte map is present */
1332
1333 #define XCL_END 0 /* Marks end of individual items */
1334 #define XCL_SINGLE 1 /* Single item (one multibyte char) follows */
1335 #define XCL_RANGE 2 /* A range (two multibyte chars) follows */
1336 #define XCL_PROP 3 /* Unicode property (2-byte property code follows) */
1337 #define XCL_NOTPROP 4 /* Unicode inverted property (ditto) */
1338
1339 /* These are escaped items that aren't just an encoding of a particular data
1340 value such as \n. They must have non-zero values, as check_escape() returns
1341 their negation. Also, they must appear in the same order as in the opcode
1342 definitions below, up to ESC_z. There's a dummy for OP_ALLANY because it
1343 corresponds to "." in DOTALL mode rather than an escape sequence. It is also
1344 used for [^] in JavaScript compatibility mode, and for \C in non-utf8 mode. In
1345 non-DOTALL mode, "." behaves like \N.
1346
1347 The special values ESC_DU, ESC_du, etc. are used instead of ESC_D, ESC_d, etc.
1348 when PCRE_UCP is set, when replacement of \d etc by \p sequences is required.
1349 They must be contiguous, and remain in order so that the replacements can be
1350 looked up from a table.
1351
1352 The final escape must be ESC_REF as subsequent values are used for
1353 backreferences (\1, \2, \3, etc). There are two tests in the code for an escape
1354 greater than ESC_b and less than ESC_Z to detect the types that may be
1355 repeated. These are the types that consume characters. If any new escapes are
1356 put in between that don't consume a character, that code will have to change.
1357 */
1358
1359 enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s,
1360 ESC_W, ESC_w, ESC_N, ESC_dum, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H,
1361 ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z,
1362 ESC_E, ESC_Q, ESC_g, ESC_k,
1363 ESC_DU, ESC_du, ESC_SU, ESC_su, ESC_WU, ESC_wu,
1364 ESC_REF };
1365
1366 /* Opcode table: Starting from 1 (i.e. after OP_END), the values up to
1367 OP_EOD must correspond in order to the list of escapes immediately above.
1368
1369 *** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions
1370 that follow must also be updated to match. There are also tables called
1371 "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
1372
1373 enum {
1374 OP_END, /* 0 End of pattern */
1375
1376 /* Values corresponding to backslashed metacharacters */
1377
1378 OP_SOD, /* 1 Start of data: \A */
1379 OP_SOM, /* 2 Start of match (subject + offset): \G */
1380 OP_SET_SOM, /* 3 Set start of match (\K) */
1381 OP_NOT_WORD_BOUNDARY, /* 4 \B */
1382 OP_WORD_BOUNDARY, /* 5 \b */
1383 OP_NOT_DIGIT, /* 6 \D */
1384 OP_DIGIT, /* 7 \d */
1385 OP_NOT_WHITESPACE, /* 8 \S */
1386 OP_WHITESPACE, /* 9 \s */
1387 OP_NOT_WORDCHAR, /* 10 \W */
1388 OP_WORDCHAR, /* 11 \w */
1389
1390 OP_ANY, /* 12 Match any character except newline */
1391 OP_ALLANY, /* 13 Match any character */
1392 OP_ANYBYTE, /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */
1393 OP_NOTPROP, /* 15 \P (not Unicode property) */
1394 OP_PROP, /* 16 \p (Unicode property) */
1395 OP_ANYNL, /* 17 \R (any newline sequence) */
1396 OP_NOT_HSPACE, /* 18 \H (not horizontal whitespace) */
1397 OP_HSPACE, /* 19 \h (horizontal whitespace) */
1398 OP_NOT_VSPACE, /* 20 \V (not vertical whitespace) */
1399 OP_VSPACE, /* 21 \v (vertical whitespace) */
1400 OP_EXTUNI, /* 22 \X (extended Unicode sequence */
1401 OP_EODN, /* 23 End of data or \n at end of data: \Z. */
1402 OP_EOD, /* 24 End of data: \z */
1403
1404 OP_CIRC, /* 25 Start of line - not multiline */
1405 OP_CIRCM, /* 26 Start of line - multiline */
1406 OP_DOLL, /* 27 End of line - not multiline */
1407 OP_DOLLM, /* 28 End of line - multiline */
1408 OP_CHAR, /* 29 Match one character, casefully */
1409 OP_CHARI, /* 30 Match one character, caselessly */
1410 OP_NOT, /* 31 Match one character, not the given one, casefully */
1411 OP_NOTI, /* 32 Match one character, not the given one, caselessly */
1412
1413 /* The following sets of 13 opcodes must always be kept in step because
1414 the offset from the first one is used to generate the others. */
1415
1416 /**** Single characters, caseful, must precede the caseless ones ****/
1417
1418 OP_STAR, /* 33 The maximizing and minimizing versions of */
1419 OP_MINSTAR, /* 34 these six opcodes must come in pairs, with */
1420 OP_PLUS, /* 35 the minimizing one second. */
1421 OP_MINPLUS, /* 36 */
1422 OP_QUERY, /* 37 */
1423 OP_MINQUERY, /* 38 */
1424
1425 OP_UPTO, /* 39 From 0 to n matches of one character, caseful*/
1426 OP_MINUPTO, /* 40 */
1427 OP_EXACT, /* 41 Exactly n matches */
1428
1429 OP_POSSTAR, /* 42 Possessified star, caseful */
1430 OP_POSPLUS, /* 43 Possessified plus, caseful */
1431 OP_POSQUERY, /* 44 Posesssified query, caseful */
1432 OP_POSUPTO, /* 45 Possessified upto, caseful */
1433
1434 /**** Single characters, caseless, must follow the caseful ones */
1435
1436 OP_STARI, /* 46 */
1437 OP_MINSTARI, /* 47 */
1438 OP_PLUSI, /* 48 */
1439 OP_MINPLUSI, /* 49 */
1440 OP_QUERYI, /* 50 */
1441 OP_MINQUERYI, /* 51 */
1442
1443 OP_UPTOI, /* 52 From 0 to n matches of one character, caseless */
1444 OP_MINUPTOI, /* 53 */
1445 OP_EXACTI, /* 54 */
1446
1447 OP_POSSTARI, /* 55 Possessified star, caseless */
1448 OP_POSPLUSI, /* 56 Possessified plus, caseless */
1449 OP_POSQUERYI, /* 57 Posesssified query, caseless */
1450 OP_POSUPTOI, /* 58 Possessified upto, caseless */
1451
1452 /**** The negated ones must follow the non-negated ones, and match them ****/
1453 /**** Negated single character, caseful; must precede the caseless ones ****/
1454
1455 OP_NOTSTAR, /* 59 The maximizing and minimizing versions of */
1456 OP_NOTMINSTAR, /* 60 these six opcodes must come in pairs, with */
1457 OP_NOTPLUS, /* 61 the minimizing one second. They must be in */
1458 OP_NOTMINPLUS, /* 62 exactly the same order as those above. */
1459 OP_NOTQUERY, /* 63 */
1460 OP_NOTMINQUERY, /* 64 */
1461
1462 OP_NOTUPTO, /* 65 From 0 to n matches, caseful */
1463 OP_NOTMINUPTO, /* 66 */
1464 OP_NOTEXACT, /* 67 Exactly n matches */
1465
1466 OP_NOTPOSSTAR, /* 68 Possessified versions, caseful */
1467 OP_NOTPOSPLUS, /* 69 */
1468 OP_NOTPOSQUERY, /* 70 */
1469 OP_NOTPOSUPTO, /* 71 */
1470
1471 /**** Negated single character, caseless; must follow the caseful ones ****/
1472
1473 OP_NOTSTARI, /* 72 */
1474 OP_NOTMINSTARI, /* 73 */
1475 OP_NOTPLUSI, /* 74 */
1476 OP_NOTMINPLUSI, /* 75 */
1477 OP_NOTQUERYI, /* 76 */
1478 OP_NOTMINQUERYI, /* 77 */
1479
1480 OP_NOTUPTOI, /* 78 From 0 to n matches, caseless */
1481 OP_NOTMINUPTOI, /* 79 */
1482 OP_NOTEXACTI, /* 80 Exactly n matches */
1483
1484 OP_NOTPOSSTARI, /* 81 Possessified versions, caseless */
1485 OP_NOTPOSPLUSI, /* 82 */
1486 OP_NOTPOSQUERYI, /* 83 */
1487 OP_NOTPOSUPTOI, /* 84 */
1488
1489 /**** Character types ****/
1490
1491 OP_TYPESTAR, /* 85 The maximizing and minimizing versions of */
1492 OP_TYPEMINSTAR, /* 86 these six opcodes must come in pairs, with */
1493 OP_TYPEPLUS, /* 87 the minimizing one second. These codes must */
1494 OP_TYPEMINPLUS, /* 88 be in exactly the same order as those above. */
1495 OP_TYPEQUERY, /* 89 */
1496 OP_TYPEMINQUERY, /* 90 */
1497
1498 OP_TYPEUPTO, /* 91 From 0 to n matches */
1499 OP_TYPEMINUPTO, /* 92 */
1500 OP_TYPEEXACT, /* 93 Exactly n matches */
1501
1502 OP_TYPEPOSSTAR, /* 94 Possessified versions */
1503 OP_TYPEPOSPLUS, /* 95 */
1504 OP_TYPEPOSQUERY, /* 96 */
1505 OP_TYPEPOSUPTO, /* 97 */
1506
1507 /* These are used for character classes and back references; only the
1508 first six are the same as the sets above. */
1509
1510 OP_CRSTAR, /* 98 The maximizing and minimizing versions of */
1511 OP_CRMINSTAR, /* 99 all these opcodes must come in pairs, with */
1512 OP_CRPLUS, /* 100 the minimizing one second. These codes must */
1513 OP_CRMINPLUS, /* 101 be in exactly the same order as those above. */
1514 OP_CRQUERY, /* 102 */
1515 OP_CRMINQUERY, /* 103 */
1516
1517 OP_CRRANGE, /* 104 These are different to the three sets above. */
1518 OP_CRMINRANGE, /* 105 */
1519
1520 /* End of quantifier opcodes */
1521
1522 OP_CLASS, /* 106 Match a character class, chars < 256 only */
1523 OP_NCLASS, /* 107 Same, but the bitmap was created from a negative
1524 class - the difference is relevant only when a
1525 UTF-8 character > 255 is encountered. */
1526 OP_XCLASS, /* 108 Extended class for handling UTF-8 chars within the
1527 class. This does both positive and negative. */
1528 OP_REF, /* 109 Match a back reference, casefully */
1529 OP_REFI, /* 110 Match a back reference, caselessly */
1530 OP_RECURSE, /* 111 Match a numbered subpattern (possibly recursive) */
1531 OP_CALLOUT, /* 112 Call out to external function if provided */
1532
1533 OP_ALT, /* 113 Start of alternation */
1534 OP_KET, /* 114 End of group that doesn't have an unbounded repeat */
1535 OP_KETRMAX, /* 115 These two must remain together and in this */
1536 OP_KETRMIN, /* 116 order. They are for groups the repeat for ever. */
1537 OP_KETRPOS, /* 117 Possessive unlimited repeat. */
1538
1539 /* The assertions must come before BRA, CBRA, ONCE, and COND, and the four
1540 asserts must remain in order. */
1541
1542 OP_REVERSE, /* 118 Move pointer back - used in lookbehind assertions */
1543 OP_ASSERT, /* 119 Positive lookahead */
1544 OP_ASSERT_NOT, /* 120 Negative lookahead */
1545 OP_ASSERTBACK, /* 121 Positive lookbehind */
1546 OP_ASSERTBACK_NOT, /* 122 Negative lookbehind */
1547
1548 /* ONCE, ONCE_NC, BRA, BRAPOS, CBRA, CBRAPOS, and COND must come immediately
1549 after the assertions, with ONCE first, as there's a test for >= ONCE for a
1550 subpattern that isn't an assertion. The POS versions must immediately follow
1551 the non-POS versions in each case. */
1552
1553 OP_ONCE, /* 123 Atomic group, contains captures */
1554 OP_ONCE_NC, /* 124 Atomic group containing no captures */
1555 OP_BRA, /* 125 Start of non-capturing bracket */
1556 OP_BRAPOS, /* 126 Ditto, with unlimited, possessive repeat */
1557 OP_CBRA, /* 127 Start of capturing bracket */
1558 OP_CBRAPOS, /* 128 Ditto, with unlimited, possessive repeat */
1559 OP_COND, /* 129 Conditional group */
1560
1561 /* These five must follow the previous five, in the same order. There's a
1562 check for >= SBRA to distinguish the two sets. */
1563
1564 OP_SBRA, /* 130 Start of non-capturing bracket, check empty */
1565 OP_SBRAPOS, /* 131 Ditto, with unlimited, possessive repeat */
1566 OP_SCBRA, /* 132 Start of capturing bracket, check empty */
1567 OP_SCBRAPOS, /* 133 Ditto, with unlimited, possessive repeat */
1568 OP_SCOND, /* 134 Conditional group, check empty */
1569
1570 /* The next two pairs must (respectively) be kept together. */
1571
1572 OP_CREF, /* 135 Used to hold a capture number as condition */
1573 OP_NCREF, /* 136 Same, but generated by a name reference*/
1574 OP_RREF, /* 137 Used to hold a recursion number as condition */
1575 OP_NRREF, /* 138 Same, but generated by a name reference*/
1576 OP_DEF, /* 139 The DEFINE condition */
1577
1578 OP_BRAZERO, /* 140 These two must remain together and in this */
1579 OP_BRAMINZERO, /* 141 order. */
1580 OP_BRAPOSZERO, /* 142 */
1581
1582 /* These are backtracking control verbs */
1583
1584 OP_MARK, /* 143 always has an argument */
1585 OP_PRUNE, /* 144 */
1586 OP_PRUNE_ARG, /* 145 same, but with argument */
1587 OP_SKIP, /* 146 */
1588 OP_SKIP_ARG, /* 147 same, but with argument */
1589 OP_THEN, /* 148 */
1590 OP_THEN_ARG, /* 149 same, but with argument */
1591 OP_COMMIT, /* 150 */
1592
1593 /* These are forced failure and success verbs */
1594
1595 OP_FAIL, /* 151 */
1596 OP_ACCEPT, /* 152 */
1597 OP_ASSERT_ACCEPT, /* 153 Used inside assertions */
1598 OP_CLOSE, /* 154 Used before OP_ACCEPT to close open captures */
1599
1600 /* This is used to skip a subpattern with a {0} quantifier */
1601
1602 OP_SKIPZERO, /* 155 */
1603
1604 /* This is not an opcode, but is used to check that tables indexed by opcode
1605 are the correct length, in order to catch updating errors - there have been
1606 some in the past. */
1607
1608 OP_TABLE_LENGTH
1609 };
1610
1611 /* *** NOTE NOTE NOTE *** Whenever the list above is updated, the two macro
1612 definitions that follow must also be updated to match. There are also tables
1613 called "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
1614
1615
1616 /* This macro defines textual names for all the opcodes. These are used only
1617 for debugging, and some of them are only partial names. The macro is referenced
1618 only in pcre_printint.c, which fills out the full names in many cases (and in
1619 some cases doesn't actually use these names at all). */
1620
1621 #define OP_NAME_LIST \
1622 "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d", \
1623 "\\S", "\\s", "\\W", "\\w", "Any", "AllAny", "Anybyte", \
1624 "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v", \
1625 "extuni", "\\Z", "\\z", \
1626 "^", "^", "$", "$", "char", "chari", "not", "noti", \
1627 "*", "*?", "+", "+?", "?", "??", \
1628 "{", "{", "{", \
1629 "*+","++", "?+", "{", \
1630 "*", "*?", "+", "+?", "?", "??", \
1631 "{", "{", "{", \
1632 "*+","++", "?+", "{", \
1633 "*", "*?", "+", "+?", "?", "??", \
1634 "{", "{", "{", \
1635 "*+","++", "?+", "{", \
1636 "*", "*?", "+", "+?", "?", "??", \
1637 "{", "{", "{", \
1638 "*+","++", "?+", "{", \
1639 "*", "*?", "+", "+?", "?", "??", "{", "{", "{", \
1640 "*+","++", "?+", "{", \
1641 "*", "*?", "+", "+?", "?", "??", "{", "{", \
1642 "class", "nclass", "xclass", "Ref", "Refi", \
1643 "Recurse", "Callout", \
1644 "Alt", "Ket", "KetRmax", "KetRmin", "KetRpos", \
1645 "Reverse", "Assert", "Assert not", "AssertB", "AssertB not", \
1646 "Once", "Once_NC", \
1647 "Bra", "BraPos", "CBra", "CBraPos", \
1648 "Cond", \
1649 "SBra", "SBraPos", "SCBra", "SCBraPos", \
1650 "SCond", \
1651 "Cond ref", "Cond nref", "Cond rec", "Cond nrec", "Cond def", \
1652 "Brazero", "Braminzero", "Braposzero", \
1653 "*MARK", "*PRUNE", "*PRUNE", "*SKIP", "*SKIP", \
1654 "*THEN", "*THEN", "*COMMIT", "*FAIL", \
1655 "*ACCEPT", "*ASSERT_ACCEPT", \
1656 "Close", "Skip zero"
1657
1658
1659 /* This macro defines the length of fixed length operations in the compiled
1660 regex. The lengths are used when searching for specific things, and also in the
1661 debugging printing of a compiled regex. We use a macro so that it can be
1662 defined close to the definitions of the opcodes themselves.
1663
1664 As things have been extended, some of these are no longer fixed lenths, but are
1665 minima instead. For example, the length of a single-character repeat may vary
1666 in UTF-8 mode. The code that uses this table must know about such things. */
1667
1668 #define OP_LENGTHS \
1669 1, /* End */ \
1670 1, 1, 1, 1, 1, /* \A, \G, \K, \B, \b */ \
1671 1, 1, 1, 1, 1, 1, /* \D, \d, \S, \s, \W, \w */ \
1672 1, 1, 1, /* Any, AllAny, Anybyte */ \
1673 3, 3, /* \P, \p */ \
1674 1, 1, 1, 1, 1, /* \R, \H, \h, \V, \v */ \
1675 1, /* \X */ \
1676 1, 1, 1, 1, 1, 1, /* \Z, \z, ^, ^M, $, $M */ \
1677 2, /* Char - the minimum length */ \
1678 2, /* Chari - the minimum length */ \
1679 2, /* not */ \
1680 2, /* noti */ \
1681 /* Positive single-char repeats ** These are */ \
1682 2, 2, 2, 2, 2, 2, /* *, *?, +, +?, ?, ?? ** minima in */ \
1683 4, 4, 4, /* upto, minupto, exact ** mode */ \
1684 2, 2, 2, 4, /* *+, ++, ?+, upto+ */ \
1685 2, 2, 2, 2, 2, 2, /* *I, *?I, +I, +?I, ?I, ??I ** UTF-8 */ \
1686 4, 4, 4, /* upto I, minupto I, exact I */ \
1687 2, 2, 2, 4, /* *+I, ++I, ?+I, upto+I */ \
1688 /* Negative single-char repeats - only for chars < 256 */ \
1689 2, 2, 2, 2, 2, 2, /* NOT *, *?, +, +?, ?, ?? */ \
1690 4, 4, 4, /* NOT upto, minupto, exact */ \
1691 2, 2, 2, 4, /* Possessive NOT *, +, ?, upto */ \
1692 2, 2, 2, 2, 2, 2, /* NOT *I, *?I, +I, +?I, ?I, ??I */ \
1693 4, 4, 4, /* NOT upto I, minupto I, exact I */ \
1694 2, 2, 2, 4, /* Possessive NOT *I, +I, ?I, upto I */ \
1695 /* Positive type repeats */ \
1696 2, 2, 2, 2, 2, 2, /* Type *, *?, +, +?, ?, ?? */ \
1697 4, 4, 4, /* Type upto, minupto, exact */ \
1698 2, 2, 2, 4, /* Possessive *+, ++, ?+, upto+ */ \
1699 /* Character class & ref repeats */ \
1700 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ \
1701 5, 5, /* CRRANGE, CRMINRANGE */ \
1702 33, /* CLASS */ \
1703 33, /* NCLASS */ \
1704 0, /* XCLASS - variable length */ \
1705 3, /* REF */ \
1706 3, /* REFI */ \
1707 1+LINK_SIZE, /* RECURSE */ \
1708 2+2*LINK_SIZE, /* CALLOUT */ \
1709 1+LINK_SIZE, /* Alt */ \
1710 1+LINK_SIZE, /* Ket */ \
1711 1+LINK_SIZE, /* KetRmax */ \
1712 1+LINK_SIZE, /* KetRmin */ \
1713 1+LINK_SIZE, /* KetRpos */ \
1714 1+LINK_SIZE, /* Reverse */ \
1715 1+LINK_SIZE, /* Assert */ \
1716 1+LINK_SIZE, /* Assert not */ \
1717 1+LINK_SIZE, /* Assert behind */ \
1718 1+LINK_SIZE, /* Assert behind not */ \
1719 1+LINK_SIZE, /* ONCE */ \
1720 1+LINK_SIZE, /* ONCE_NC */ \
1721 1+LINK_SIZE, /* BRA */ \
1722 1+LINK_SIZE, /* BRAPOS */ \
1723 3+LINK_SIZE, /* CBRA */ \
1724 3+LINK_SIZE, /* CBRAPOS */ \
1725 1+LINK_SIZE, /* COND */ \
1726 1+LINK_SIZE, /* SBRA */ \
1727 1+LINK_SIZE, /* SBRAPOS */ \
1728 3+LINK_SIZE, /* SCBRA */ \
1729 3+LINK_SIZE, /* SCBRAPOS */ \
1730 1+LINK_SIZE, /* SCOND */ \
1731 3, 3, /* CREF, NCREF */ \
1732 3, 3, /* RREF, NRREF */ \
1733 1, /* DEF */ \
1734 1, 1, 1, /* BRAZERO, BRAMINZERO, BRAPOSZERO */ \
1735 3, 1, 3, /* MARK, PRUNE, PRUNE_ARG */ \
1736 1, 3, /* SKIP, SKIP_ARG */ \
1737 1, 3, /* THEN, THEN_ARG */ \
1738 1, 1, 1, 1, /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT */ \
1739 3, 1 /* CLOSE, SKIPZERO */
1740
1741 /* A magic value for OP_RREF and OP_NRREF to indicate the "any recursion"
1742 condition. */
1743
1744 #define RREF_ANY 0xffff
1745
1746 /* Compile time error code numbers. They are given names so that they can more
1747 easily be tracked. When a new number is added, the table called eint in
1748 pcreposix.c must be updated. */
1749
1750 enum { ERR0, ERR1, ERR2, ERR3, ERR4, ERR5, ERR6, ERR7, ERR8, ERR9,
1751 ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,
1752 ERR20, ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR28, ERR29,
1753 ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,
1754 ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,
1755 ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,
1756 ERR60, ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERR68, ERR69,
1757 ERR70, ERRCOUNT };
1758
1759 /* The real format of the start of the pcre block; the index of names and the
1760 code vector run on as long as necessary after the end. We store an explicit
1761 offset to the name table so that if a regex is compiled on one host, saved, and
1762 then run on another where the size of pointers is different, all might still
1763 be well. For the case of compiled-on-4 and run-on-8, we include an extra
1764 pointer that is always NULL. For future-proofing, a few dummy fields were
1765 originally included - even though you can never get this planning right - but
1766 there is only one left now.
1767
1768 NOTE NOTE NOTE:
1769 Because people can now save and re-use compiled patterns, any additions to this
1770 structure should be made at the end, and something earlier (e.g. a new
1771 flag in the options or one of the dummy fields) should indicate that the new
1772 fields are present. Currently PCRE always sets the dummy fields to zero.
1773 NOTE NOTE NOTE
1774 */
1775
1776 typedef struct real_pcre {
1777 pcre_uint32 magic_number;
1778 pcre_uint32 size; /* Total that was malloced */
1779 pcre_uint32 options; /* Public options */
1780 pcre_uint16 flags; /* Private flags */
1781 pcre_uint16 dummy1; /* For future use */
1782 pcre_uint16 top_bracket;
1783 pcre_uint16 top_backref;
1784 pcre_uint16 first_byte;
1785 pcre_uint16 req_byte;
1786 pcre_uint16 name_table_offset; /* Offset to name table that follows */
1787 pcre_uint16 name_entry_size; /* Size of any name items */
1788 pcre_uint16 name_count; /* Number of name items */
1789 pcre_uint16 ref_count; /* Reference count */
1790
1791 const pcre_uint8 *tables; /* Pointer to tables or NULL for std */
1792 const pcre_uint8 *nullpad; /* NULL padding */
1793 } real_pcre;
1794
1795 /* The format of the block used to store data from pcre_study(). The same
1796 remark (see NOTE above) about extending this structure applies. */
1797
1798 typedef struct pcre_study_data {
1799 pcre_uint32 size; /* Total that was malloced */
1800 pcre_uint32 flags; /* Private flags */
1801 pcre_uint8 start_bits[32]; /* Starting char bits */
1802 pcre_uint32 minlength; /* Minimum subject length */
1803 } pcre_study_data;
1804
1805 /* Structure for building a chain of open capturing subpatterns during
1806 compiling, so that instructions to close them can be compiled when (*ACCEPT) is
1807 encountered. This is also used to identify subpatterns that contain recursive
1808 back references to themselves, so that they can be made atomic. */
1809
1810 typedef struct open_capitem {
1811 struct open_capitem *next; /* Chain link */
1812 pcre_uint16 number; /* Capture number */
1813 pcre_uint16 flag; /* Set TRUE if recursive back ref */
1814 } open_capitem;
1815
1816 /* Structure for passing "static" information around between the functions
1817 doing the compiling, so that they are thread-safe. */
1818
1819 typedef struct compile_data {
1820 const pcre_uint8 *lcc; /* Points to lower casing table */
1821 const pcre_uint8 *fcc; /* Points to case-flipping table */
1822 const pcre_uint8 *cbits; /* Points to character type table */
1823 const pcre_uint8 *ctypes; /* Points to table of type maps */
1824 const pcre_uchar *start_workspace;/* The start of working space */
1825 const pcre_uchar *start_code; /* The start of the compiled code */
1826 const pcre_uchar *start_pattern; /* The start of the pattern */
1827 const pcre_uchar *end_pattern; /* The end of the pattern */
1828 open_capitem *open_caps; /* Chain of open capture items */
1829 pcre_uchar *hwm; /* High watermark of workspace */
1830 pcre_uchar *name_table; /* The name/number table */
1831 int names_found; /* Number of entries so far */
1832 int name_entry_size; /* Size of each entry */
1833 int bracount; /* Count of capturing parens as we compile */
1834 int final_bracount; /* Saved value after first pass */
1835 int top_backref; /* Maximum back reference */
1836 unsigned int backref_map; /* Bitmap of low back refs */
1837 int assert_depth; /* Depth of nested assertions */
1838 int external_options; /* External (initial) options */
1839 int external_flags; /* External flag bits to be set */
1840 int req_varyopt; /* "After variable item" flag for reqbyte */
1841 BOOL had_accept; /* (*ACCEPT) encountered */
1842 BOOL check_lookbehind; /* Lookbehinds need later checking */
1843 int nltype; /* Newline type */
1844 int nllen; /* Newline string length */
1845 pcre_uchar nl[4]; /* Newline string when fixed length */
1846 } compile_data;
1847
1848 /* Structure for maintaining a chain of pointers to the currently incomplete
1849 branches, for testing for left recursion while compiling. */
1850
1851 typedef struct branch_chain {
1852 struct branch_chain *outer;
1853 pcre_uchar *current_branch;
1854 } branch_chain;
1855
1856 /* Structure for items in a linked list that represents an explicit recursive
1857 call within the pattern; used by pcre_exec(). */
1858
1859 typedef struct recursion_info {
1860 struct recursion_info *prevrec; /* Previous recursion record (or NULL) */
1861 int group_num; /* Number of group that was called */
1862 int *offset_save; /* Pointer to start of saved offsets */
1863 int saved_max; /* Number of saved offsets */
1864 PCRE_PUCHAR subject_position; /* Position at start of recursion */
1865 } recursion_info;
1866
1867 /* A similar structure for pcre_dfa_exec(). */
1868
1869 typedef struct dfa_recursion_info {
1870 struct dfa_recursion_info *prevrec;
1871 int group_num;
1872 PCRE_PUCHAR subject_position;
1873 } dfa_recursion_info;
1874
1875 /* Structure for building a chain of data for holding the values of the subject
1876 pointer at the start of each subpattern, so as to detect when an empty string
1877 has been matched by a subpattern - to break infinite loops; used by
1878 pcre_exec(). */
1879
1880 typedef struct eptrblock {
1881 struct eptrblock *epb_prev;
1882 PCRE_PUCHAR epb_saved_eptr;
1883 } eptrblock;
1884
1885
1886 /* Structure for passing "static" information around between the functions
1887 doing traditional NFA matching, so that they are thread-safe. */
1888
1889 typedef struct match_data {
1890 unsigned long int match_call_count; /* As it says */
1891 unsigned long int match_limit; /* As it says */
1892 unsigned long int match_limit_recursion; /* As it says */
1893 int *offset_vector; /* Offset vector */
1894 int offset_end; /* One past the end */
1895 int offset_max; /* The maximum usable for return data */
1896 int nltype; /* Newline type */
1897 int nllen; /* Newline string length */
1898 int name_count; /* Number of names in name table */
1899 int name_entry_size; /* Size of entry in names table */
1900 pcre_uchar *name_table; /* Table of names */
1901 pcre_uchar nl[4]; /* Newline string when fixed */
1902 const pcre_uint8 *lcc; /* Points to lower casing table */
1903 const pcre_uint8 *ctypes; /* Points to table of type maps */
1904 BOOL offset_overflow; /* Set if too many extractions */
1905 BOOL notbol; /* NOTBOL flag */
1906 BOOL noteol; /* NOTEOL flag */
1907 BOOL utf8; /* UTF8 flag */
1908 BOOL jscript_compat; /* JAVASCRIPT_COMPAT flag */
1909 BOOL use_ucp; /* PCRE_UCP flag */
1910 BOOL endonly; /* Dollar not before final \n */
1911 BOOL notempty; /* Empty string match not wanted */
1912 BOOL notempty_atstart; /* Empty string match at start not wanted */
1913 BOOL hitend; /* Hit the end of the subject at some point */
1914 BOOL bsr_anycrlf; /* \R is just any CRLF, not full Unicode */
1915 BOOL hasthen; /* Pattern contains (*THEN) */
1916 const pcre_uchar *start_code; /* For use when recursing */
1917 PCRE_PUCHAR start_subject; /* Start of the subject string */
1918 PCRE_PUCHAR end_subject; /* End of the subject string */
1919 PCRE_PUCHAR start_match_ptr; /* Start of matched string */
1920 PCRE_PUCHAR end_match_ptr; /* Subject position at end match */
1921 PCRE_PUCHAR start_used_ptr; /* Earliest consulted character */
1922 int partial; /* PARTIAL options */
1923 int end_offset_top; /* Highwater mark at end of match */
1924 int capture_last; /* Most recent capture number */
1925 int start_offset; /* The start offset value */
1926 int match_function_type; /* Set for certain special calls of MATCH() */
1927 eptrblock *eptrchain; /* Chain of eptrblocks for tail recursions */
1928 int eptrn; /* Next free eptrblock */
1929 recursion_info *recursive; /* Linked list of recursion data */
1930 void *callout_data; /* To pass back to callouts */
1931 const pcre_uchar *mark; /* Mark pointer to pass back */
1932 const pcre_uchar *once_target; /* Where to back up to for atomic groups */
1933 } match_data;
1934
1935 /* A similar structure is used for the same purpose by the DFA matching
1936 functions. */
1937
1938 typedef struct dfa_match_data {
1939 const pcre_uchar *start_code; /* Start of the compiled pattern */
1940 const pcre_uchar *start_subject ; /* Start of the subject string */
1941 const pcre_uchar *end_subject; /* End of subject string */
1942 const pcre_uchar *start_used_ptr; /* Earliest consulted character */
1943 const pcre_uint8 *tables; /* Character tables */
1944 int start_offset; /* The start offset value */
1945 int moptions; /* Match options */
1946 int poptions; /* Pattern options */
1947 int nltype; /* Newline type */
1948 int nllen; /* Newline string length */
1949 pcre_uchar nl[4]; /* Newline string when fixed */
1950 void *callout_data; /* To pass back to callouts */
1951 dfa_recursion_info *recursive; /* Linked list of recursion data */
1952 } dfa_match_data;
1953
1954 /* Bit definitions for entries in the pcre_ctypes table. */
1955
1956 #define ctype_space 0x01
1957 #define ctype_letter 0x02
1958 #define ctype_digit 0x04
1959 #define ctype_xdigit 0x08
1960 #define ctype_word 0x10 /* alphanumeric or '_' */
1961 #define ctype_meta 0x80 /* regexp meta char or zero (end pattern) */
1962
1963 /* Offsets for the bitmap tables in pcre_cbits. Each table contains a set
1964 of bits for a class map. Some classes are built by combining these tables. */
1965
1966 #define cbit_space 0 /* [:space:] or \s */
1967 #define cbit_xdigit 32 /* [:xdigit:] */
1968 #define cbit_digit 64 /* [:digit:] or \d */
1969 #define cbit_upper 96 /* [:upper:] */
1970 #define cbit_lower 128 /* [:lower:] */
1971 #define cbit_word 160 /* [:word:] or \w */
1972 #define cbit_graph 192 /* [:graph:] */
1973 #define cbit_print 224 /* [:print:] */
1974 #define cbit_punct 256 /* [:punct:] */
1975 #define cbit_cntrl 288 /* [:cntrl:] */
1976 #define cbit_length 320 /* Length of the cbits table */
1977
1978 /* Offsets of the various tables from the base tables pointer, and
1979 total length. */
1980
1981 #define lcc_offset 0
1982 #define fcc_offset 256
1983 #define cbits_offset 512
1984 #define ctypes_offset (cbits_offset + cbit_length)
1985 #define tables_length (ctypes_offset + 256)
1986
1987 /* Internal function prefix */
1988
1989 #ifdef COMPILE_PCRE8
1990 #define PRIV(name) _pcre_##name
1991 #else
1992 #ifdef COMPILE_PCRE16
1993 #define PRIV(name) _pcre16_##name
1994 #else
1995 #error Unsupported compiling mode
1996 #endif /* COMPILE_PCRE16 */
1997 #endif /* COMPILE_PCRE8 */
1998
1999 /* Layout of the UCP type table that translates property names into types and
2000 codes. Each entry used to point directly to a name, but to reduce the number of
2001 relocations in shared libraries, it now has an offset into a single string
2002 instead. */
2003
2004 typedef struct {
2005 pcre_uint16 name_offset;
2006 pcre_uint16 type;
2007 pcre_uint16 value;
2008 } ucp_type_table;
2009
2010
2011 /* Internal shared data tables. These are tables that are used by more than one
2012 of the exported public functions. They have to be "external" in the C sense,
2013 but are not part of the PCRE public API. The data for these tables is in the
2014 pcre_tables.c module. */
2015
2016 extern const int PRIV(utf8_table1)[];
2017 extern const int PRIV(utf8_table2)[];
2018 extern const int PRIV(utf8_table3)[];
2019 extern const pcre_uint8 PRIV(utf8_table4)[];
2020
2021 extern const int PRIV(utf8_table1_size);
2022
2023 extern const char PRIV(utt_names)[];
2024 extern const ucp_type_table PRIV(utt)[];
2025 extern const int PRIV(utt_size);
2026
2027 extern const pcre_uint8 PRIV(default_tables)[];
2028
2029 extern const pcre_uint8 PRIV(OP_lengths)[];
2030
2031
2032 /* Internal shared functions. These are functions that are used by more than
2033 one of the exported public functions. They have to be "external" in the C
2034 sense, but are not part of the PCRE public API. */
2035
2036 /* String comparison functions. */
2037 #ifdef COMPILE_PCRE8
2038
2039 #define STRCMP_UC_UC(str1, str2) \
2040 strcmp((char *)(str1), (char *)(str2))
2041 #define STRCMP_UC_C8(str1, str2) \
2042 strcmp((char *)(str1), (str2))
2043 #define STRNCMP_UC_UC(str1, str2, num) \
2044 strncmp((char *)(str1), (char *)(str2), (num))
2045 #define STRNCMP_UC_C8(str1, str2, num) \
2046 strncmp((char *)(str1), (str2), (num))
2047 #define STRLEN_UC(str) strlen((const char *)str)
2048
2049 #else
2050
2051 extern int PRIV(strcmp_uc_uc)(const pcre_uchar *,
2052 const pcre_uchar *);
2053 extern int PRIV(strcmp_uc_c8)(const pcre_uchar *,
2054 const char *);
2055 extern int PRIV(strncmp_uc_uc)(const pcre_uchar *,
2056 const pcre_uchar *, unsigned int num);
2057 extern int PRIV(strncmp_uc_c8)(const pcre_uchar *,
2058 const char *, unsigned int num);
2059 extern unsigned int PRIV(strlen_uc)(const pcre_uchar *str);
2060
2061 #define STRCMP_UC_UC(str1, str2) \
2062 PRIV(strcmp_uc_uc)((str1), (str2))
2063 #define STRCMP_UC_C8(str1, str2) \
2064 PRIV(strcmp_uc_c8)((str1), (str2))
2065 #define STRNCMP_UC_UC(str1, str2, num) \
2066 PRIV(strncmp_uc_uc)((str1), (str2), (num))
2067 #define STRNCMP_UC_C8(str1, str2, num) \
2068 PRIV(strncmp_uc_c8)((str1), (str2), (num))
2069 #define STRLEN_UC(str) PRIV(strlen_uc)(str)
2070
2071 #endif /* COMPILE_PCRE8 */
2072
2073 extern const pcre_uchar *PRIV(find_bracket)(const pcre_uchar *, BOOL, int);
2074 extern BOOL PRIV(is_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2075 int *, BOOL);
2076 extern int PRIV(ord2utf8)(int, pcre_uint8 *);
2077 extern real_pcre *PRIV(try_flipped)(const real_pcre *, real_pcre *,
2078 const pcre_study_data *, pcre_study_data *);
2079 #ifndef COMPILE_PCRE16
2080 extern int PRIV(valid_utf8)(PCRE_PUCHAR, int, int *);
2081 #else
2082 extern int PRIV(valid_utf16)(PCRE_PUCHAR, int, int *);
2083 #endif
2084 extern BOOL PRIV(was_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2085 int *, BOOL);
2086 extern BOOL PRIV(xclass)(int, const pcre_uchar *);
2087
2088 #ifdef SUPPORT_JIT
2089 extern void PRIV(jit_compile)(const real_pcre *, pcre_extra *);
2090 extern int PRIV(jit_exec)(const real_pcre *, void *,
2091 const pcre_uchar *, int, int, int, int, int *, int);
2092 extern void PRIV(jit_free)(void *);
2093 #endif
2094
2095 /* Unicode character database (UCD) */
2096
2097 typedef struct {
2098 pcre_uint8 script;
2099 pcre_uint8 chartype;
2100 pcre_int32 other_case;
2101 } ucd_record;
2102
2103 extern const ucd_record PRIV(ucd_records)[];
2104 extern const pcre_uint8 PRIV(ucd_stage1)[];
2105 extern const pcre_uint16 PRIV(ucd_stage2)[];
2106 extern const int PRIV(ucp_gentype)[];
2107 #ifdef SUPPORT_JIT
2108 extern const int PRIV(ucp_typerange)[];
2109 #endif
2110
2111 /* UCD access macros */
2112
2113 #define UCD_BLOCK_SIZE 128
2114 #define GET_UCD(ch) (PRIV(ucd_records) + \
2115 PRIV(ucd_stage2)[PRIV(ucd_stage1)[(ch) / UCD_BLOCK_SIZE] * \
2116 UCD_BLOCK_SIZE + (ch) % UCD_BLOCK_SIZE])
2117
2118 #define UCD_CHARTYPE(ch) GET_UCD(ch)->chartype
2119 #define UCD_SCRIPT(ch) GET_UCD(ch)->script
2120 #define UCD_CATEGORY(ch) PRIV(ucp_gentype)[UCD_CHARTYPE(ch)]
2121 #define UCD_OTHERCASE(ch) (ch + GET_UCD(ch)->other_case)
2122
2123 #endif
2124
2125 /* 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