/[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 785 - (show annotations)
Mon Dec 5 20:12:24 2011 UTC (7 years, 9 months ago) by zherczeg
File MIME type: text/plain
File size: 92870 byte(s)
Improving UTF-16 support by fixing a lot of issues.
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. This character type is
234 called pcre_uchar.
235
236 The IN_UCHARS macro multiply its argument with the byte size of the current
237 pcre_uchar type. Useful for memcpy and such operations, whose require the
238 byte size of their input/output buffers.
239
240 The MAX_255 macro checks whether its pcre_uchar input is less than 256.
241
242 The TABLE_GET macro is designed for accessing elements of tables whose contain
243 exactly 256 items. When the character is able to contain more than 256
244 items, some check is needed before accessing these tables.
245 */
246
247 #ifdef COMPILE_PCRE8
248
249 typedef unsigned char pcre_uchar;
250 #define IN_UCHARS(x) (x)
251 #define MAX_255(c) 1
252 #define TABLE_GET(c, table, default) ((table)[c])
253
254 #else
255
256 #ifdef COMPILE_PCRE16
257 #if USHRT_MAX != 65535
258 /* This is a warning message. Change PCRE_SCHAR16 to a 16 bit data type in
259 pcre.h(.in) and disable (comment out) this message. */
260 #error Warning: PCRE_SCHAR16 is not a 16 bit data type.
261 #endif
262
263 typedef pcre_uint16 pcre_uchar;
264 #define IN_UCHARS(x) ((x) << 1)
265 #define MAX_255(c) ((c) <= 255u)
266 #define TABLE_GET(c, table, default) (MAX_255(c)? ((table)[c]):(default))
267
268 #else
269 #error Unsupported compiling mode
270 #endif /* COMPILE_PCRE16 */
271
272 #endif /* COMPILE_PCRE8 */
273
274 /* This is an unsigned int value that no character can ever have. UTF-8
275 characters only go up to 0x7fffffff (though Unicode doesn't go beyond
276 0x0010ffff). */
277
278 #define NOTACHAR 0xffffffff
279
280 /* PCRE is able to support several different kinds of newline (CR, LF, CRLF,
281 "any" and "anycrlf" at present). The following macros are used to package up
282 testing for newlines. NLBLOCK, PSSTART, and PSEND are defined in the various
283 modules to indicate in which datablock the parameters exist, and what the
284 start/end of string field names are. */
285
286 #define NLTYPE_FIXED 0 /* Newline is a fixed length string */
287 #define NLTYPE_ANY 1 /* Newline is any Unicode line ending */
288 #define NLTYPE_ANYCRLF 2 /* Newline is CR, LF, or CRLF */
289
290 /* This macro checks for a newline at the given position */
291
292 #define IS_NEWLINE(p) \
293 ((NLBLOCK->nltype != NLTYPE_FIXED)? \
294 ((p) < NLBLOCK->PSEND && \
295 PRIV(is_newline)((p), NLBLOCK->nltype, NLBLOCK->PSEND, \
296 &(NLBLOCK->nllen), utf)) \
297 : \
298 ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \
299 (p)[0] == NLBLOCK->nl[0] && \
300 (NLBLOCK->nllen == 1 || (p)[1] == NLBLOCK->nl[1]) \
301 ) \
302 )
303
304 /* This macro checks for a newline immediately preceding the given position */
305
306 #define WAS_NEWLINE(p) \
307 ((NLBLOCK->nltype != NLTYPE_FIXED)? \
308 ((p) > NLBLOCK->PSSTART && \
309 PRIV(was_newline)((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \
310 &(NLBLOCK->nllen), utf)) \
311 : \
312 ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \
313 (p)[-NLBLOCK->nllen] == NLBLOCK->nl[0] && \
314 (NLBLOCK->nllen == 1 || (p)[-NLBLOCK->nllen+1] == NLBLOCK->nl[1]) \
315 ) \
316 )
317
318 /* When PCRE is compiled as a C++ library, the subject pointer can be replaced
319 with a custom type. This makes it possible, for example, to allow pcre_exec()
320 to process subject strings that are discontinuous by using a smart pointer
321 class. It must always be possible to inspect all of the subject string in
322 pcre_exec() because of the way it backtracks. Two macros are required in the
323 normal case, for sign-unspecified and unsigned char pointers. The former is
324 used for the external interface and appears in pcre.h, which is why its name
325 must begin with PCRE_. */
326
327 #ifdef CUSTOM_SUBJECT_PTR
328 #define PCRE_PUCHAR CUSTOM_SUBJECT_PTR
329 #else
330 #define PCRE_PUCHAR const pcre_uchar *
331 #endif
332
333 /* Include the public PCRE header and the definitions of UCP character property
334 values. */
335
336 #include "pcre.h"
337 #include "ucp.h"
338
339 /* When compiling for use with the Virtual Pascal compiler, these functions
340 need to have their names changed. PCRE must be compiled with the -DVPCOMPAT
341 option on the command line. */
342
343 #ifdef VPCOMPAT
344 #define strlen(s) _strlen(s)
345 #define strncmp(s1,s2,m) _strncmp(s1,s2,m)
346 #define memcmp(s,c,n) _memcmp(s,c,n)
347 #define memcpy(d,s,n) _memcpy(d,s,n)
348 #define memmove(d,s,n) _memmove(d,s,n)
349 #define memset(s,c,n) _memset(s,c,n)
350 #else /* VPCOMPAT */
351
352 /* To cope with SunOS4 and other systems that lack memmove() but have bcopy(),
353 define a macro for memmove() if HAVE_MEMMOVE is false, provided that HAVE_BCOPY
354 is set. Otherwise, include an emulating function for those systems that have
355 neither (there some non-Unix environments where this is the case). */
356
357 #ifndef HAVE_MEMMOVE
358 #undef memmove /* some systems may have a macro */
359 #ifdef HAVE_BCOPY
360 #define memmove(a, b, c) bcopy(b, a, c)
361 #else /* HAVE_BCOPY */
362 static void *
363 pcre_memmove(void *d, const void *s, size_t n)
364 {
365 size_t i;
366 unsigned char *dest = (unsigned char *)d;
367 const unsigned char *src = (const unsigned char *)s;
368 if (dest > src)
369 {
370 dest += n;
371 src += n;
372 for (i = 0; i < n; ++i) *(--dest) = *(--src);
373 return (void *)dest;
374 }
375 else
376 {
377 for (i = 0; i < n; ++i) *dest++ = *src++;
378 return (void *)(dest - n);
379 }
380 }
381 #define memmove(a, b, c) pcre_memmove(a, b, c)
382 #endif /* not HAVE_BCOPY */
383 #endif /* not HAVE_MEMMOVE */
384 #endif /* not VPCOMPAT */
385
386
387 /* PCRE keeps offsets in its compiled code as 2-byte quantities (always stored
388 in big-endian order) by default. These are used, for example, to link from the
389 start of a subpattern to its alternatives and its end. The use of 2 bytes per
390 offset limits the size of the compiled regex to around 64K, which is big enough
391 for almost everybody. However, I received a request for an even bigger limit.
392 For this reason, and also to make the code easier to maintain, the storing and
393 loading of offsets from the byte string is now handled by the macros that are
394 defined here.
395
396 The macros are controlled by the value of LINK_SIZE. This defaults to 2 in
397 the config.h file, but can be overridden by using -D on the command line. This
398 is automated on Unix systems via the "configure" command. */
399
400 #ifdef COMPILE_PCRE8
401
402 #if LINK_SIZE == 2
403
404 #define PUT(a,n,d) \
405 (a[n] = (d) >> 8), \
406 (a[(n)+1] = (d) & 255)
407
408 #define GET(a,n) \
409 (((a)[n] << 8) | (a)[(n)+1])
410
411 #define MAX_PATTERN_SIZE (1 << 16)
412
413
414 #elif LINK_SIZE == 3
415
416 #define PUT(a,n,d) \
417 (a[n] = (d) >> 16), \
418 (a[(n)+1] = (d) >> 8), \
419 (a[(n)+2] = (d) & 255)
420
421 #define GET(a,n) \
422 (((a)[n] << 16) | ((a)[(n)+1] << 8) | (a)[(n)+2])
423
424 #define MAX_PATTERN_SIZE (1 << 24)
425
426
427 #elif LINK_SIZE == 4
428
429 #define PUT(a,n,d) \
430 (a[n] = (d) >> 24), \
431 (a[(n)+1] = (d) >> 16), \
432 (a[(n)+2] = (d) >> 8), \
433 (a[(n)+3] = (d) & 255)
434
435 #define GET(a,n) \
436 (((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3])
437
438 /* Keep it positive */
439 #define MAX_PATTERN_SIZE (1 << 30)
440
441 #else
442 #error LINK_SIZE must be either 2, 3, or 4
443 #endif
444
445 #else /* COMPILE_PCRE8 */
446
447 #ifdef COMPILE_PCRE16
448
449 #if LINK_SIZE == 2
450
451 #undef LINK_SIZE
452 #define LINK_SIZE 1
453
454 #define PUT(a,n,d) \
455 (a[n] = (d))
456
457 #define GET(a,n) \
458 (a[n])
459
460 #define MAX_PATTERN_SIZE (1 << 16)
461
462 #elif LINK_SIZE == 3 || LINK_SIZE == 4
463
464 #undef LINK_SIZE
465 #define LINK_SIZE 2
466
467 #define PUT(a,n,d) \
468 (a[n] = (d) >> 16), \
469 (a[(n)+1] = (d) & 65536)
470
471 #define GET(a,n) \
472 (((a)[n] << 16) | (a)[(n)+1])
473
474 /* Keep it positive */
475 #define MAX_PATTERN_SIZE (1 << 30)
476
477 #else
478 #error LINK_SIZE must be either 2, 3, or 4
479 #endif
480
481 #else
482 #error Unsupported compiling mode
483 #endif /* COMPILE_PCRE16 */
484
485 #endif /* COMPILE_PCRE8 */
486
487 /* Convenience macro defined in terms of the others */
488
489 #define PUTINC(a,n,d) PUT(a,n,d), a += LINK_SIZE
490
491
492 /* PCRE uses some other 2-byte quantities that do not change when the size of
493 offsets changes. There are used for repeat counts and for other things such as
494 capturing parenthesis numbers in back references. */
495
496 #ifdef COMPILE_PCRE8
497
498 #define IMM2_SIZE 2
499
500 #define PUT2(a,n,d) \
501 a[n] = (d) >> 8; \
502 a[(n)+1] = (d) & 255
503
504 #define GET2(a,n) \
505 (((a)[n] << 8) | (a)[(n)+1])
506
507 #else /* COMPILE_PCRE8 */
508
509 #ifdef COMPILE_PCRE16
510
511 #define IMM2_SIZE 1
512
513 #define PUT2(a,n,d) \
514 a[n] = d
515
516 #define GET2(a,n) \
517 a[n]
518
519 #else
520 #error Unsupported compiling mode
521 #endif /* COMPILE_PCRE16 */
522
523 #endif /* COMPILE_PCRE8 */
524
525 #define PUT2INC(a,n,d) PUT2(a,n,d), a += IMM2_SIZE
526
527 /* When UTF-8 encoding is being used, a character is no longer just a single
528 byte. The macros for character handling generate simple sequences when used in
529 byte-mode, and more complicated ones for UTF-8 characters. GETCHARLENTEST is
530 not used when UTF-8 is not supported, so it is not defined, and BACKCHAR should
531 never be called in byte mode. To make sure they can never even appear when
532 UTF-8 support is omitted, we don't even define them. */
533
534 /* #define HAS_EXTRALEN(c) */
535 /* #define GET_EXTRALEN(c) */
536 #ifndef SUPPORT_UTF
537 #define GETCHAR(c, eptr) c = *eptr;
538 #define GETCHARTEST(c, eptr) c = *eptr;
539 #define GETCHARINC(c, eptr) c = *eptr++;
540 #define GETCHARINCTEST(c, eptr) c = *eptr++;
541 #define GETCHARLEN(c, eptr, len) c = *eptr;
542 /* #define GETCHARLENTEST(c, eptr, len) */
543 /* #define BACKCHAR(eptr) */
544 /* #define FORWARDCHAR(eptr) */
545 /* #define ACROSSCHAR(condition, eptr, action) */
546
547 #else /* SUPPORT_UTF */
548
549 #ifdef COMPILE_PCRE8
550
551 /* These macros were originally written in the form of loops that used data
552 from the tables whose names start with PRIV(utf8_table). They were rewritten by
553 a user so as not to use loops, because in some environments this gives a
554 significant performance advantage, and it seems never to do any harm. */
555
556 /* Tests whether the code point needs extra characters to decode. */
557
558 #define HAS_EXTRALEN(c) ((c) >= 0xc0)
559
560 /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
561 Otherwise it has an undefined behaviour. */
562
563 #define GET_EXTRALEN(c) (PRIV(utf8_table4)[(c) & 0x3f])
564
565 /* Base macro to pick up the remaining bytes of a UTF-8 character, not
566 advancing the pointer. */
567
568 #define GETUTF8(c, eptr) \
569 { \
570 if ((c & 0x20) == 0) \
571 c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
572 else if ((c & 0x10) == 0) \
573 c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
574 else if ((c & 0x08) == 0) \
575 c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
576 ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
577 else if ((c & 0x04) == 0) \
578 c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
579 ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
580 (eptr[4] & 0x3f); \
581 else \
582 c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
583 ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
584 ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
585 }
586
587 /* Get the next UTF-8 character, not advancing the pointer. This is called when
588 we know we are in UTF-8 mode. */
589
590 #define GETCHAR(c, eptr) \
591 c = *eptr; \
592 if (c >= 0xc0) GETUTF8(c, eptr);
593
594 /* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the
595 pointer. */
596
597 #define GETCHARTEST(c, eptr) \
598 c = *eptr; \
599 if (utf && c >= 0xc0) GETUTF8(c, eptr);
600
601 /* Base macro to pick up the remaining bytes of a UTF-8 character, advancing
602 the pointer. */
603
604 #define GETUTF8INC(c, eptr) \
605 { \
606 if ((c & 0x20) == 0) \
607 c = ((c & 0x1f) << 6) | (*eptr++ & 0x3f); \
608 else if ((c & 0x10) == 0) \
609 { \
610 c = ((c & 0x0f) << 12) | ((*eptr & 0x3f) << 6) | (eptr[1] & 0x3f); \
611 eptr += 2; \
612 } \
613 else if ((c & 0x08) == 0) \
614 { \
615 c = ((c & 0x07) << 18) | ((*eptr & 0x3f) << 12) | \
616 ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
617 eptr += 3; \
618 } \
619 else if ((c & 0x04) == 0) \
620 { \
621 c = ((c & 0x03) << 24) | ((*eptr & 0x3f) << 18) | \
622 ((eptr[1] & 0x3f) << 12) | ((eptr[2] & 0x3f) << 6) | \
623 (eptr[3] & 0x3f); \
624 eptr += 4; \
625 } \
626 else \
627 { \
628 c = ((c & 0x01) << 30) | ((*eptr & 0x3f) << 24) | \
629 ((eptr[1] & 0x3f) << 18) | ((eptr[2] & 0x3f) << 12) | \
630 ((eptr[3] & 0x3f) << 6) | (eptr[4] & 0x3f); \
631 eptr += 5; \
632 } \
633 }
634
635 /* Get the next UTF-8 character, advancing the pointer. This is called when we
636 know we are in UTF-8 mode. */
637
638 #define GETCHARINC(c, eptr) \
639 c = *eptr++; \
640 if (c >= 0xc0) GETUTF8INC(c, eptr);
641
642 /* Get the next character, testing for UTF-8 mode, and advancing the pointer.
643 This is called when we don't know if we are in UTF-8 mode. */
644
645 #define GETCHARINCTEST(c, eptr) \
646 c = *eptr++; \
647 if (utf && c >= 0xc0) GETUTF8INC(c, eptr);
648
649 /* Base macro to pick up the remaining bytes of a UTF-8 character, not
650 advancing the pointer, incrementing the length. */
651
652 #define GETUTF8LEN(c, eptr, len) \
653 { \
654 if ((c & 0x20) == 0) \
655 { \
656 c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
657 len++; \
658 } \
659 else if ((c & 0x10) == 0) \
660 { \
661 c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
662 len += 2; \
663 } \
664 else if ((c & 0x08) == 0) \
665 {\
666 c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
667 ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
668 len += 3; \
669 } \
670 else if ((c & 0x04) == 0) \
671 { \
672 c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
673 ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
674 (eptr[4] & 0x3f); \
675 len += 4; \
676 } \
677 else \
678 {\
679 c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
680 ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
681 ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
682 len += 5; \
683 } \
684 }
685
686 /* Get the next UTF-8 character, not advancing the pointer, incrementing length
687 if there are extra bytes. This is called when we know we are in UTF-8 mode. */
688
689 #define GETCHARLEN(c, eptr, len) \
690 c = *eptr; \
691 if (c >= 0xc0) GETUTF8LEN(c, eptr, len);
692
693 /* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the
694 pointer, incrementing length if there are extra bytes. This is called when we
695 do not know if we are in UTF-8 mode. */
696
697 #define GETCHARLENTEST(c, eptr, len) \
698 c = *eptr; \
699 if (utf && c >= 0xc0) GETUTF8LEN(c, eptr, len);
700
701 /* If the pointer is not at the start of a character, move it back until
702 it is. This is called only in UTF-8 mode - we don't put a test within the macro
703 because almost all calls are already within a block of UTF-8 only code. */
704
705 #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--
706
707 /* Same as above, just in the other direction. */
708 #define FORWARDCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr++
709
710 /* Same as above, but it allows a fully customizable form. */
711 #define ACROSSCHAR(condition, eptr, action) \
712 while((condition) && ((eptr) & 0xc0) == 0x80) action
713
714 #else /* COMPILE_PCRE8 */
715
716 #ifdef COMPILE_PCRE16
717
718 /* Tests whether the code point needs extra characters to decode. */
719
720 #define HAS_EXTRALEN(c) (((c) & 0xfc00) == 0xd800)
721
722 /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
723 Otherwise it has an undefined behaviour. */
724
725 #define GET_EXTRALEN(c) 1
726
727 /* Base macro to pick up the low surrogate of a UTF-16 character, not
728 advancing the pointer. */
729
730 #define GETUTF16(c, eptr) \
731 { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; }
732
733 /* Get the next UTF-16 character, not advancing the pointer. This is called when
734 we know we are in UTF-16 mode. */
735
736 #define GETCHAR(c, eptr) \
737 c = *eptr; \
738 if ((c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
739
740 /* Get the next UTF-16 character, testing for UTF-16 mode, and not advancing the
741 pointer. */
742
743 #define GETCHARTEST(c, eptr) \
744 c = *eptr; \
745 if (utf && (c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
746
747 /* Base macro to pick up the low surrogate of a UTF-16 character, advancing
748 the pointer. */
749
750 #define GETUTF16INC(c, eptr) \
751 { c = (((c & 0x3ff) << 10) | (*eptr++ & 0x3ff)) + 0x10000; }
752
753 /* Get the next UTF-16 character, advancing the pointer. This is called when we
754 know we are in UTF-16 mode. */
755
756 #define GETCHARINC(c, eptr) \
757 c = *eptr++; \
758 if ((c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
759
760 /* Get the next character, testing for UTF-16 mode, and advancing the pointer.
761 This is called when we don't know if we are in UTF-16 mode. */
762
763 #define GETCHARINCTEST(c, eptr) \
764 c = *eptr++; \
765 if (utf && (c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
766
767 /* Base macro to pick up the low surrogate of a UTF-16 character, not
768 advancing the pointer, incrementing the length. */
769
770 #define GETUTF16LEN(c, eptr, len) \
771 { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; len++; }
772
773 /* Get the next UTF-16 character, not advancing the pointer, incrementing
774 length if there is a low surrogate. This is called when we know we are in
775 UTF-16 mode. */
776
777 #define GETCHARLEN(c, eptr, len) \
778 c = *eptr; \
779 if ((c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
780
781 /* Get the next UTF-816character, testing for UTF-16 mode, not advancing the
782 pointer, incrementing length if there is a low surrogate. This is called when
783 we do not know if we are in UTF-16 mode. */
784
785 #define GETCHARLENTEST(c, eptr, len) \
786 c = *eptr; \
787 if (utf && (c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
788
789 /* If the pointer is not at the start of a character, move it back until
790 it is. This is called only in UTF-16 mode - we don't put a test within the
791 macro because almost all calls are already within a block of UTF-16 only
792 code. */
793
794 #define BACKCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr--
795
796 /* Same as above, just in the other direction. */
797 #define FORWARDCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr++
798
799 /* Same as above, but it allows a fully customizable form. */
800 #define ACROSSCHAR(condition, eptr, action) \
801 if ((condition) && ((eptr) & 0xfc00) == 0xdc00) action
802
803 #endif
804
805 #endif /* COMPILE_PCRE8 */
806
807 #endif /* SUPPORT_UTF */
808
809
810 /* In case there is no definition of offsetof() provided - though any proper
811 Standard C system should have one. */
812
813 #ifndef offsetof
814 #define offsetof(p_type,field) ((size_t)&(((p_type *)0)->field))
815 #endif
816
817
818 /* Private flags containing information about the compiled regex. They used to
819 live at the top end of the options word, but that got almost full, so now they
820 are in a 16-bit flags word. From release 8.00, PCRE_NOPARTIAL is unused, as
821 the restrictions on partial matching have been lifted. It remains for backwards
822 compatibility. */
823
824 #define PCRE_NOPARTIAL 0x0001 /* can't use partial with this regex */
825 #define PCRE_FIRSTSET 0x0002 /* first_char is set */
826 #define PCRE_REQCHSET 0x0004 /* req_byte is set */
827 #define PCRE_STARTLINE 0x0008 /* start after \n for multiline */
828 #define PCRE_JCHANGED 0x0010 /* j option used in regex */
829 #define PCRE_HASCRORLF 0x0020 /* explicit \r or \n in pattern */
830 #define PCRE_HASTHEN 0x0040 /* pattern contains (*THEN) */
831 #define PCRE_FCH_CASELESS 0x0080 /* caseless first char */
832 #define PCRE_RCH_CASELESS 0x0100 /* caseless requested char */
833
834 /* Flags for the "extra" block produced by pcre_study(). */
835
836 #define PCRE_STUDY_MAPPED 0x0001 /* a map of starting chars exists */
837 #define PCRE_STUDY_MINLEN 0x0002 /* a minimum length field exists */
838
839 /* Masks for identifying the public options that are permitted at compile
840 time, run time, or study time, respectively. */
841
842 #define PCRE_NEWLINE_BITS (PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_ANY| \
843 PCRE_NEWLINE_ANYCRLF)
844
845 #define PUBLIC_COMPILE_OPTIONS \
846 (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \
847 PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \
848 PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \
849 PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
850 PCRE_JAVASCRIPT_COMPAT|PCRE_UCP|PCRE_NO_START_OPTIMIZE)
851
852 #define PUBLIC_EXEC_OPTIONS \
853 (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
854 PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_NEWLINE_BITS| \
855 PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE|PCRE_NO_START_OPTIMIZE)
856
857 #define PUBLIC_DFA_EXEC_OPTIONS \
858 (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
859 PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_DFA_SHORTEST| \
860 PCRE_DFA_RESTART|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
861 PCRE_NO_START_OPTIMIZE)
862
863 #define PUBLIC_STUDY_OPTIONS \
864 PCRE_STUDY_JIT_COMPILE
865
866 /* Magic number to provide a small check against being handed junk. Also used
867 to detect whether a pattern was compiled on a host of different endianness. */
868
869 #define MAGIC_NUMBER 0x50435245UL /* 'PCRE' */
870
871 /* Negative values for the firstchar and reqchar variables */
872
873 #define REQ_UNSET (-2)
874 #define REQ_NONE (-1)
875
876 /* The maximum remaining length of subject we are prepared to search for a
877 req_byte match. */
878
879 #define REQ_BYTE_MAX 1000
880
881 /* Miscellaneous definitions. The #ifndef is to pacify compiler warnings in
882 environments where these macros are defined elsewhere. Unfortunately, there
883 is no way to do the same for the typedef. */
884
885 typedef int BOOL;
886
887 #ifndef FALSE
888 #define FALSE 0
889 #define TRUE 1
890 #endif
891
892 /* If PCRE is to support UTF-8 on EBCDIC platforms, we cannot use normal
893 character constants like '*' because the compiler would emit their EBCDIC code,
894 which is different from their ASCII/UTF-8 code. Instead we define macros for
895 the characters so that they always use the ASCII/UTF-8 code when UTF-8 support
896 is enabled. When UTF-8 support is not enabled, the definitions use character
897 literals. Both character and string versions of each character are needed, and
898 there are some longer strings as well.
899
900 This means that, on EBCDIC platforms, the PCRE library can handle either
901 EBCDIC, or UTF-8, but not both. To support both in the same compiled library
902 would need different lookups depending on whether PCRE_UTF8 was set or not.
903 This would make it impossible to use characters in switch/case statements,
904 which would reduce performance. For a theoretical use (which nobody has asked
905 for) in a minority area (EBCDIC platforms), this is not sensible. Any
906 application that did need both could compile two versions of the library, using
907 macros to give the functions distinct names. */
908
909 #ifndef SUPPORT_UTF8
910
911 /* UTF-8 support is not enabled; use the platform-dependent character literals
912 so that PCRE works on both ASCII and EBCDIC platforms, in non-UTF-mode only. */
913
914 #define CHAR_HT '\t'
915 #define CHAR_VT '\v'
916 #define CHAR_FF '\f'
917 #define CHAR_CR '\r'
918 #define CHAR_NL '\n'
919 #define CHAR_BS '\b'
920 #define CHAR_BEL '\a'
921 #ifdef EBCDIC
922 #define CHAR_ESC '\047'
923 #define CHAR_DEL '\007'
924 #else
925 #define CHAR_ESC '\033'
926 #define CHAR_DEL '\177'
927 #endif
928
929 #define CHAR_SPACE ' '
930 #define CHAR_EXCLAMATION_MARK '!'
931 #define CHAR_QUOTATION_MARK '"'
932 #define CHAR_NUMBER_SIGN '#'
933 #define CHAR_DOLLAR_SIGN '$'
934 #define CHAR_PERCENT_SIGN '%'
935 #define CHAR_AMPERSAND '&'
936 #define CHAR_APOSTROPHE '\''
937 #define CHAR_LEFT_PARENTHESIS '('
938 #define CHAR_RIGHT_PARENTHESIS ')'
939 #define CHAR_ASTERISK '*'
940 #define CHAR_PLUS '+'
941 #define CHAR_COMMA ','
942 #define CHAR_MINUS '-'
943 #define CHAR_DOT '.'
944 #define CHAR_SLASH '/'
945 #define CHAR_0 '0'
946 #define CHAR_1 '1'
947 #define CHAR_2 '2'
948 #define CHAR_3 '3'
949 #define CHAR_4 '4'
950 #define CHAR_5 '5'
951 #define CHAR_6 '6'
952 #define CHAR_7 '7'
953 #define CHAR_8 '8'
954 #define CHAR_9 '9'
955 #define CHAR_COLON ':'
956 #define CHAR_SEMICOLON ';'
957 #define CHAR_LESS_THAN_SIGN '<'
958 #define CHAR_EQUALS_SIGN '='
959 #define CHAR_GREATER_THAN_SIGN '>'
960 #define CHAR_QUESTION_MARK '?'
961 #define CHAR_COMMERCIAL_AT '@'
962 #define CHAR_A 'A'
963 #define CHAR_B 'B'
964 #define CHAR_C 'C'
965 #define CHAR_D 'D'
966 #define CHAR_E 'E'
967 #define CHAR_F 'F'
968 #define CHAR_G 'G'
969 #define CHAR_H 'H'
970 #define CHAR_I 'I'
971 #define CHAR_J 'J'
972 #define CHAR_K 'K'
973 #define CHAR_L 'L'
974 #define CHAR_M 'M'
975 #define CHAR_N 'N'
976 #define CHAR_O 'O'
977 #define CHAR_P 'P'
978 #define CHAR_Q 'Q'
979 #define CHAR_R 'R'
980 #define CHAR_S 'S'
981 #define CHAR_T 'T'
982 #define CHAR_U 'U'
983 #define CHAR_V 'V'
984 #define CHAR_W 'W'
985 #define CHAR_X 'X'
986 #define CHAR_Y 'Y'
987 #define CHAR_Z 'Z'
988 #define CHAR_LEFT_SQUARE_BRACKET '['
989 #define CHAR_BACKSLASH '\\'
990 #define CHAR_RIGHT_SQUARE_BRACKET ']'
991 #define CHAR_CIRCUMFLEX_ACCENT '^'
992 #define CHAR_UNDERSCORE '_'
993 #define CHAR_GRAVE_ACCENT '`'
994 #define CHAR_a 'a'
995 #define CHAR_b 'b'
996 #define CHAR_c 'c'
997 #define CHAR_d 'd'
998 #define CHAR_e 'e'
999 #define CHAR_f 'f'
1000 #define CHAR_g 'g'
1001 #define CHAR_h 'h'
1002 #define CHAR_i 'i'
1003 #define CHAR_j 'j'
1004 #define CHAR_k 'k'
1005 #define CHAR_l 'l'
1006 #define CHAR_m 'm'
1007 #define CHAR_n 'n'
1008 #define CHAR_o 'o'
1009 #define CHAR_p 'p'
1010 #define CHAR_q 'q'
1011 #define CHAR_r 'r'
1012 #define CHAR_s 's'
1013 #define CHAR_t 't'
1014 #define CHAR_u 'u'
1015 #define CHAR_v 'v'
1016 #define CHAR_w 'w'
1017 #define CHAR_x 'x'
1018 #define CHAR_y 'y'
1019 #define CHAR_z 'z'
1020 #define CHAR_LEFT_CURLY_BRACKET '{'
1021 #define CHAR_VERTICAL_LINE '|'
1022 #define CHAR_RIGHT_CURLY_BRACKET '}'
1023 #define CHAR_TILDE '~'
1024
1025 #define STR_HT "\t"
1026 #define STR_VT "\v"
1027 #define STR_FF "\f"
1028 #define STR_CR "\r"
1029 #define STR_NL "\n"
1030 #define STR_BS "\b"
1031 #define STR_BEL "\a"
1032 #ifdef EBCDIC
1033 #define STR_ESC "\047"
1034 #define STR_DEL "\007"
1035 #else
1036 #define STR_ESC "\033"
1037 #define STR_DEL "\177"
1038 #endif
1039
1040 #define STR_SPACE " "
1041 #define STR_EXCLAMATION_MARK "!"
1042 #define STR_QUOTATION_MARK "\""
1043 #define STR_NUMBER_SIGN "#"
1044 #define STR_DOLLAR_SIGN "$"
1045 #define STR_PERCENT_SIGN "%"
1046 #define STR_AMPERSAND "&"
1047 #define STR_APOSTROPHE "'"
1048 #define STR_LEFT_PARENTHESIS "("
1049 #define STR_RIGHT_PARENTHESIS ")"
1050 #define STR_ASTERISK "*"
1051 #define STR_PLUS "+"
1052 #define STR_COMMA ","
1053 #define STR_MINUS "-"
1054 #define STR_DOT "."
1055 #define STR_SLASH "/"
1056 #define STR_0 "0"
1057 #define STR_1 "1"
1058 #define STR_2 "2"
1059 #define STR_3 "3"
1060 #define STR_4 "4"
1061 #define STR_5 "5"
1062 #define STR_6 "6"
1063 #define STR_7 "7"
1064 #define STR_8 "8"
1065 #define STR_9 "9"
1066 #define STR_COLON ":"
1067 #define STR_SEMICOLON ";"
1068 #define STR_LESS_THAN_SIGN "<"
1069 #define STR_EQUALS_SIGN "="
1070 #define STR_GREATER_THAN_SIGN ">"
1071 #define STR_QUESTION_MARK "?"
1072 #define STR_COMMERCIAL_AT "@"
1073 #define STR_A "A"
1074 #define STR_B "B"
1075 #define STR_C "C"
1076 #define STR_D "D"
1077 #define STR_E "E"
1078 #define STR_F "F"
1079 #define STR_G "G"
1080 #define STR_H "H"
1081 #define STR_I "I"
1082 #define STR_J "J"
1083 #define STR_K "K"
1084 #define STR_L "L"
1085 #define STR_M "M"
1086 #define STR_N "N"
1087 #define STR_O "O"
1088 #define STR_P "P"
1089 #define STR_Q "Q"
1090 #define STR_R "R"
1091 #define STR_S "S"
1092 #define STR_T "T"
1093 #define STR_U "U"
1094 #define STR_V "V"
1095 #define STR_W "W"
1096 #define STR_X "X"
1097 #define STR_Y "Y"
1098 #define STR_Z "Z"
1099 #define STR_LEFT_SQUARE_BRACKET "["
1100 #define STR_BACKSLASH "\\"
1101 #define STR_RIGHT_SQUARE_BRACKET "]"
1102 #define STR_CIRCUMFLEX_ACCENT "^"
1103 #define STR_UNDERSCORE "_"
1104 #define STR_GRAVE_ACCENT "`"
1105 #define STR_a "a"
1106 #define STR_b "b"
1107 #define STR_c "c"
1108 #define STR_d "d"
1109 #define STR_e "e"
1110 #define STR_f "f"
1111 #define STR_g "g"
1112 #define STR_h "h"
1113 #define STR_i "i"
1114 #define STR_j "j"
1115 #define STR_k "k"
1116 #define STR_l "l"
1117 #define STR_m "m"
1118 #define STR_n "n"
1119 #define STR_o "o"
1120 #define STR_p "p"
1121 #define STR_q "q"
1122 #define STR_r "r"
1123 #define STR_s "s"
1124 #define STR_t "t"
1125 #define STR_u "u"
1126 #define STR_v "v"
1127 #define STR_w "w"
1128 #define STR_x "x"
1129 #define STR_y "y"
1130 #define STR_z "z"
1131 #define STR_LEFT_CURLY_BRACKET "{"
1132 #define STR_VERTICAL_LINE "|"
1133 #define STR_RIGHT_CURLY_BRACKET "}"
1134 #define STR_TILDE "~"
1135
1136 #define STRING_ACCEPT0 "ACCEPT\0"
1137 #define STRING_COMMIT0 "COMMIT\0"
1138 #define STRING_F0 "F\0"
1139 #define STRING_FAIL0 "FAIL\0"
1140 #define STRING_MARK0 "MARK\0"
1141 #define STRING_PRUNE0 "PRUNE\0"
1142 #define STRING_SKIP0 "SKIP\0"
1143 #define STRING_THEN "THEN"
1144
1145 #define STRING_alpha0 "alpha\0"
1146 #define STRING_lower0 "lower\0"
1147 #define STRING_upper0 "upper\0"
1148 #define STRING_alnum0 "alnum\0"
1149 #define STRING_ascii0 "ascii\0"
1150 #define STRING_blank0 "blank\0"
1151 #define STRING_cntrl0 "cntrl\0"
1152 #define STRING_digit0 "digit\0"
1153 #define STRING_graph0 "graph\0"
1154 #define STRING_print0 "print\0"
1155 #define STRING_punct0 "punct\0"
1156 #define STRING_space0 "space\0"
1157 #define STRING_word0 "word\0"
1158 #define STRING_xdigit "xdigit"
1159
1160 #define STRING_DEFINE "DEFINE"
1161
1162 #define STRING_CR_RIGHTPAR "CR)"
1163 #define STRING_LF_RIGHTPAR "LF)"
1164 #define STRING_CRLF_RIGHTPAR "CRLF)"
1165 #define STRING_ANY_RIGHTPAR "ANY)"
1166 #define STRING_ANYCRLF_RIGHTPAR "ANYCRLF)"
1167 #define STRING_BSR_ANYCRLF_RIGHTPAR "BSR_ANYCRLF)"
1168 #define STRING_BSR_UNICODE_RIGHTPAR "BSR_UNICODE)"
1169 #define STRING_UTF8_RIGHTPAR "UTF8)"
1170 #define STRING_UCP_RIGHTPAR "UCP)"
1171 #define STRING_NO_START_OPT_RIGHTPAR "NO_START_OPT)"
1172
1173 #else /* SUPPORT_UTF8 */
1174
1175 /* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This
1176 works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode
1177 only. */
1178
1179 #define CHAR_HT '\011'
1180 #define CHAR_VT '\013'
1181 #define CHAR_FF '\014'
1182 #define CHAR_CR '\015'
1183 #define CHAR_NL '\012'
1184 #define CHAR_BS '\010'
1185 #define CHAR_BEL '\007'
1186 #define CHAR_ESC '\033'
1187 #define CHAR_DEL '\177'
1188
1189 #define CHAR_SPACE '\040'
1190 #define CHAR_EXCLAMATION_MARK '\041'
1191 #define CHAR_QUOTATION_MARK '\042'
1192 #define CHAR_NUMBER_SIGN '\043'
1193 #define CHAR_DOLLAR_SIGN '\044'
1194 #define CHAR_PERCENT_SIGN '\045'
1195 #define CHAR_AMPERSAND '\046'
1196 #define CHAR_APOSTROPHE '\047'
1197 #define CHAR_LEFT_PARENTHESIS '\050'
1198 #define CHAR_RIGHT_PARENTHESIS '\051'
1199 #define CHAR_ASTERISK '\052'
1200 #define CHAR_PLUS '\053'
1201 #define CHAR_COMMA '\054'
1202 #define CHAR_MINUS '\055'
1203 #define CHAR_DOT '\056'
1204 #define CHAR_SLASH '\057'
1205 #define CHAR_0 '\060'
1206 #define CHAR_1 '\061'
1207 #define CHAR_2 '\062'
1208 #define CHAR_3 '\063'
1209 #define CHAR_4 '\064'
1210 #define CHAR_5 '\065'
1211 #define CHAR_6 '\066'
1212 #define CHAR_7 '\067'
1213 #define CHAR_8 '\070'
1214 #define CHAR_9 '\071'
1215 #define CHAR_COLON '\072'
1216 #define CHAR_SEMICOLON '\073'
1217 #define CHAR_LESS_THAN_SIGN '\074'
1218 #define CHAR_EQUALS_SIGN '\075'
1219 #define CHAR_GREATER_THAN_SIGN '\076'
1220 #define CHAR_QUESTION_MARK '\077'
1221 #define CHAR_COMMERCIAL_AT '\100'
1222 #define CHAR_A '\101'
1223 #define CHAR_B '\102'
1224 #define CHAR_C '\103'
1225 #define CHAR_D '\104'
1226 #define CHAR_E '\105'
1227 #define CHAR_F '\106'
1228 #define CHAR_G '\107'
1229 #define CHAR_H '\110'
1230 #define CHAR_I '\111'
1231 #define CHAR_J '\112'
1232 #define CHAR_K '\113'
1233 #define CHAR_L '\114'
1234 #define CHAR_M '\115'
1235 #define CHAR_N '\116'
1236 #define CHAR_O '\117'
1237 #define CHAR_P '\120'
1238 #define CHAR_Q '\121'
1239 #define CHAR_R '\122'
1240 #define CHAR_S '\123'
1241 #define CHAR_T '\124'
1242 #define CHAR_U '\125'
1243 #define CHAR_V '\126'
1244 #define CHAR_W '\127'
1245 #define CHAR_X '\130'
1246 #define CHAR_Y '\131'
1247 #define CHAR_Z '\132'
1248 #define CHAR_LEFT_SQUARE_BRACKET '\133'
1249 #define CHAR_BACKSLASH '\134'
1250 #define CHAR_RIGHT_SQUARE_BRACKET '\135'
1251 #define CHAR_CIRCUMFLEX_ACCENT '\136'
1252 #define CHAR_UNDERSCORE '\137'
1253 #define CHAR_GRAVE_ACCENT '\140'
1254 #define CHAR_a '\141'
1255 #define CHAR_b '\142'
1256 #define CHAR_c '\143'
1257 #define CHAR_d '\144'
1258 #define CHAR_e '\145'
1259 #define CHAR_f '\146'
1260 #define CHAR_g '\147'
1261 #define CHAR_h '\150'
1262 #define CHAR_i '\151'
1263 #define CHAR_j '\152'
1264 #define CHAR_k '\153'
1265 #define CHAR_l '\154'
1266 #define CHAR_m '\155'
1267 #define CHAR_n '\156'
1268 #define CHAR_o '\157'
1269 #define CHAR_p '\160'
1270 #define CHAR_q '\161'
1271 #define CHAR_r '\162'
1272 #define CHAR_s '\163'
1273 #define CHAR_t '\164'
1274 #define CHAR_u '\165'
1275 #define CHAR_v '\166'
1276 #define CHAR_w '\167'
1277 #define CHAR_x '\170'
1278 #define CHAR_y '\171'
1279 #define CHAR_z '\172'
1280 #define CHAR_LEFT_CURLY_BRACKET '\173'
1281 #define CHAR_VERTICAL_LINE '\174'
1282 #define CHAR_RIGHT_CURLY_BRACKET '\175'
1283 #define CHAR_TILDE '\176'
1284
1285 #define STR_HT "\011"
1286 #define STR_VT "\013"
1287 #define STR_FF "\014"
1288 #define STR_CR "\015"
1289 #define STR_NL "\012"
1290 #define STR_BS "\010"
1291 #define STR_BEL "\007"
1292 #define STR_ESC "\033"
1293 #define STR_DEL "\177"
1294
1295 #define STR_SPACE "\040"
1296 #define STR_EXCLAMATION_MARK "\041"
1297 #define STR_QUOTATION_MARK "\042"
1298 #define STR_NUMBER_SIGN "\043"
1299 #define STR_DOLLAR_SIGN "\044"
1300 #define STR_PERCENT_SIGN "\045"
1301 #define STR_AMPERSAND "\046"
1302 #define STR_APOSTROPHE "\047"
1303 #define STR_LEFT_PARENTHESIS "\050"
1304 #define STR_RIGHT_PARENTHESIS "\051"
1305 #define STR_ASTERISK "\052"
1306 #define STR_PLUS "\053"
1307 #define STR_COMMA "\054"
1308 #define STR_MINUS "\055"
1309 #define STR_DOT "\056"
1310 #define STR_SLASH "\057"
1311 #define STR_0 "\060"
1312 #define STR_1 "\061"
1313 #define STR_2 "\062"
1314 #define STR_3 "\063"
1315 #define STR_4 "\064"
1316 #define STR_5 "\065"
1317 #define STR_6 "\066"
1318 #define STR_7 "\067"
1319 #define STR_8 "\070"
1320 #define STR_9 "\071"
1321 #define STR_COLON "\072"
1322 #define STR_SEMICOLON "\073"
1323 #define STR_LESS_THAN_SIGN "\074"
1324 #define STR_EQUALS_SIGN "\075"
1325 #define STR_GREATER_THAN_SIGN "\076"
1326 #define STR_QUESTION_MARK "\077"
1327 #define STR_COMMERCIAL_AT "\100"
1328 #define STR_A "\101"
1329 #define STR_B "\102"
1330 #define STR_C "\103"
1331 #define STR_D "\104"
1332 #define STR_E "\105"
1333 #define STR_F "\106"
1334 #define STR_G "\107"
1335 #define STR_H "\110"
1336 #define STR_I "\111"
1337 #define STR_J "\112"
1338 #define STR_K "\113"
1339 #define STR_L "\114"
1340 #define STR_M "\115"
1341 #define STR_N "\116"
1342 #define STR_O "\117"
1343 #define STR_P "\120"
1344 #define STR_Q "\121"
1345 #define STR_R "\122"
1346 #define STR_S "\123"
1347 #define STR_T "\124"
1348 #define STR_U "\125"
1349 #define STR_V "\126"
1350 #define STR_W "\127"
1351 #define STR_X "\130"
1352 #define STR_Y "\131"
1353 #define STR_Z "\132"
1354 #define STR_LEFT_SQUARE_BRACKET "\133"
1355 #define STR_BACKSLASH "\134"
1356 #define STR_RIGHT_SQUARE_BRACKET "\135"
1357 #define STR_CIRCUMFLEX_ACCENT "\136"
1358 #define STR_UNDERSCORE "\137"
1359 #define STR_GRAVE_ACCENT "\140"
1360 #define STR_a "\141"
1361 #define STR_b "\142"
1362 #define STR_c "\143"
1363 #define STR_d "\144"
1364 #define STR_e "\145"
1365 #define STR_f "\146"
1366 #define STR_g "\147"
1367 #define STR_h "\150"
1368 #define STR_i "\151"
1369 #define STR_j "\152"
1370 #define STR_k "\153"
1371 #define STR_l "\154"
1372 #define STR_m "\155"
1373 #define STR_n "\156"
1374 #define STR_o "\157"
1375 #define STR_p "\160"
1376 #define STR_q "\161"
1377 #define STR_r "\162"
1378 #define STR_s "\163"
1379 #define STR_t "\164"
1380 #define STR_u "\165"
1381 #define STR_v "\166"
1382 #define STR_w "\167"
1383 #define STR_x "\170"
1384 #define STR_y "\171"
1385 #define STR_z "\172"
1386 #define STR_LEFT_CURLY_BRACKET "\173"
1387 #define STR_VERTICAL_LINE "\174"
1388 #define STR_RIGHT_CURLY_BRACKET "\175"
1389 #define STR_TILDE "\176"
1390
1391 #define STRING_ACCEPT0 STR_A STR_C STR_C STR_E STR_P STR_T "\0"
1392 #define STRING_COMMIT0 STR_C STR_O STR_M STR_M STR_I STR_T "\0"
1393 #define STRING_F0 STR_F "\0"
1394 #define STRING_FAIL0 STR_F STR_A STR_I STR_L "\0"
1395 #define STRING_MARK0 STR_M STR_A STR_R STR_K "\0"
1396 #define STRING_PRUNE0 STR_P STR_R STR_U STR_N STR_E "\0"
1397 #define STRING_SKIP0 STR_S STR_K STR_I STR_P "\0"
1398 #define STRING_THEN STR_T STR_H STR_E STR_N
1399
1400 #define STRING_alpha0 STR_a STR_l STR_p STR_h STR_a "\0"
1401 #define STRING_lower0 STR_l STR_o STR_w STR_e STR_r "\0"
1402 #define STRING_upper0 STR_u STR_p STR_p STR_e STR_r "\0"
1403 #define STRING_alnum0 STR_a STR_l STR_n STR_u STR_m "\0"
1404 #define STRING_ascii0 STR_a STR_s STR_c STR_i STR_i "\0"
1405 #define STRING_blank0 STR_b STR_l STR_a STR_n STR_k "\0"
1406 #define STRING_cntrl0 STR_c STR_n STR_t STR_r STR_l "\0"
1407 #define STRING_digit0 STR_d STR_i STR_g STR_i STR_t "\0"
1408 #define STRING_graph0 STR_g STR_r STR_a STR_p STR_h "\0"
1409 #define STRING_print0 STR_p STR_r STR_i STR_n STR_t "\0"
1410 #define STRING_punct0 STR_p STR_u STR_n STR_c STR_t "\0"
1411 #define STRING_space0 STR_s STR_p STR_a STR_c STR_e "\0"
1412 #define STRING_word0 STR_w STR_o STR_r STR_d "\0"
1413 #define STRING_xdigit STR_x STR_d STR_i STR_g STR_i STR_t
1414
1415 #define STRING_DEFINE STR_D STR_E STR_F STR_I STR_N STR_E
1416
1417 #define STRING_CR_RIGHTPAR STR_C STR_R STR_RIGHT_PARENTHESIS
1418 #define STRING_LF_RIGHTPAR STR_L STR_F STR_RIGHT_PARENTHESIS
1419 #define STRING_CRLF_RIGHTPAR STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1420 #define STRING_ANY_RIGHTPAR STR_A STR_N STR_Y STR_RIGHT_PARENTHESIS
1421 #define STRING_ANYCRLF_RIGHTPAR STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1422 #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
1423 #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
1424 #define STRING_UTF8_RIGHTPAR STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS
1425 #define STRING_UCP_RIGHTPAR STR_U STR_C STR_P STR_RIGHT_PARENTHESIS
1426 #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
1427
1428 #endif /* SUPPORT_UTF8 */
1429
1430 /* Escape items that are just an encoding of a particular data value. */
1431
1432 #ifndef ESC_e
1433 #define ESC_e CHAR_ESC
1434 #endif
1435
1436 #ifndef ESC_f
1437 #define ESC_f CHAR_FF
1438 #endif
1439
1440 #ifndef ESC_n
1441 #define ESC_n CHAR_NL
1442 #endif
1443
1444 #ifndef ESC_r
1445 #define ESC_r CHAR_CR
1446 #endif
1447
1448 /* We can't officially use ESC_t because it is a POSIX reserved identifier
1449 (presumably because of all the others like size_t). */
1450
1451 #ifndef ESC_tee
1452 #define ESC_tee CHAR_HT
1453 #endif
1454
1455 /* Codes for different types of Unicode property */
1456
1457 #define PT_ANY 0 /* Any property - matches all chars */
1458 #define PT_LAMP 1 /* L& - the union of Lu, Ll, Lt */
1459 #define PT_GC 2 /* Specified general characteristic (e.g. L) */
1460 #define PT_PC 3 /* Specified particular characteristic (e.g. Lu) */
1461 #define PT_SC 4 /* Script (e.g. Han) */
1462 #define PT_ALNUM 5 /* Alphanumeric - the union of L and N */
1463 #define PT_SPACE 6 /* Perl space - Z plus 9,10,12,13 */
1464 #define PT_PXSPACE 7 /* POSIX space - Z plus 9,10,11,12,13 */
1465 #define PT_WORD 8 /* Word - L plus N plus underscore */
1466
1467 /* Flag bits and data types for the extended class (OP_XCLASS) for classes that
1468 contain characters with values greater than 255. */
1469
1470 #define XCL_NOT 0x01 /* Flag: this is a negative class */
1471 #define XCL_MAP 0x02 /* Flag: a 32-byte map is present */
1472
1473 #define XCL_END 0 /* Marks end of individual items */
1474 #define XCL_SINGLE 1 /* Single item (one multibyte char) follows */
1475 #define XCL_RANGE 2 /* A range (two multibyte chars) follows */
1476 #define XCL_PROP 3 /* Unicode property (2-byte property code follows) */
1477 #define XCL_NOTPROP 4 /* Unicode inverted property (ditto) */
1478
1479 /* These are escaped items that aren't just an encoding of a particular data
1480 value such as \n. They must have non-zero values, as check_escape() returns
1481 their negation. Also, they must appear in the same order as in the opcode
1482 definitions below, up to ESC_z. There's a dummy for OP_ALLANY because it
1483 corresponds to "." in DOTALL mode rather than an escape sequence. It is also
1484 used for [^] in JavaScript compatibility mode, and for \C in non-utf mode. In
1485 non-DOTALL mode, "." behaves like \N.
1486
1487 The special values ESC_DU, ESC_du, etc. are used instead of ESC_D, ESC_d, etc.
1488 when PCRE_UCP is set, when replacement of \d etc by \p sequences is required.
1489 They must be contiguous, and remain in order so that the replacements can be
1490 looked up from a table.
1491
1492 The final escape must be ESC_REF as subsequent values are used for
1493 backreferences (\1, \2, \3, etc). There are two tests in the code for an escape
1494 greater than ESC_b and less than ESC_Z to detect the types that may be
1495 repeated. These are the types that consume characters. If any new escapes are
1496 put in between that don't consume a character, that code will have to change.
1497 */
1498
1499 enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s,
1500 ESC_W, ESC_w, ESC_N, ESC_dum, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H,
1501 ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z,
1502 ESC_E, ESC_Q, ESC_g, ESC_k,
1503 ESC_DU, ESC_du, ESC_SU, ESC_su, ESC_WU, ESC_wu,
1504 ESC_REF };
1505
1506 /* Opcode table: Starting from 1 (i.e. after OP_END), the values up to
1507 OP_EOD must correspond in order to the list of escapes immediately above.
1508
1509 *** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions
1510 that follow must also be updated to match. There are also tables called
1511 "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
1512
1513 enum {
1514 OP_END, /* 0 End of pattern */
1515
1516 /* Values corresponding to backslashed metacharacters */
1517
1518 OP_SOD, /* 1 Start of data: \A */
1519 OP_SOM, /* 2 Start of match (subject + offset): \G */
1520 OP_SET_SOM, /* 3 Set start of match (\K) */
1521 OP_NOT_WORD_BOUNDARY, /* 4 \B */
1522 OP_WORD_BOUNDARY, /* 5 \b */
1523 OP_NOT_DIGIT, /* 6 \D */
1524 OP_DIGIT, /* 7 \d */
1525 OP_NOT_WHITESPACE, /* 8 \S */
1526 OP_WHITESPACE, /* 9 \s */
1527 OP_NOT_WORDCHAR, /* 10 \W */
1528 OP_WORDCHAR, /* 11 \w */
1529
1530 OP_ANY, /* 12 Match any character except newline */
1531 OP_ALLANY, /* 13 Match any character */
1532 OP_ANYBYTE, /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */
1533 OP_NOTPROP, /* 15 \P (not Unicode property) */
1534 OP_PROP, /* 16 \p (Unicode property) */
1535 OP_ANYNL, /* 17 \R (any newline sequence) */
1536 OP_NOT_HSPACE, /* 18 \H (not horizontal whitespace) */
1537 OP_HSPACE, /* 19 \h (horizontal whitespace) */
1538 OP_NOT_VSPACE, /* 20 \V (not vertical whitespace) */
1539 OP_VSPACE, /* 21 \v (vertical whitespace) */
1540 OP_EXTUNI, /* 22 \X (extended Unicode sequence */
1541 OP_EODN, /* 23 End of data or \n at end of data: \Z. */
1542 OP_EOD, /* 24 End of data: \z */
1543
1544 OP_CIRC, /* 25 Start of line - not multiline */
1545 OP_CIRCM, /* 26 Start of line - multiline */
1546 OP_DOLL, /* 27 End of line - not multiline */
1547 OP_DOLLM, /* 28 End of line - multiline */
1548 OP_CHAR, /* 29 Match one character, casefully */
1549 OP_CHARI, /* 30 Match one character, caselessly */
1550 OP_NOT, /* 31 Match one character, not the given one, casefully */
1551 OP_NOTI, /* 32 Match one character, not the given one, caselessly */
1552
1553 /* The following sets of 13 opcodes must always be kept in step because
1554 the offset from the first one is used to generate the others. */
1555
1556 /**** Single characters, caseful, must precede the caseless ones ****/
1557
1558 OP_STAR, /* 33 The maximizing and minimizing versions of */
1559 OP_MINSTAR, /* 34 these six opcodes must come in pairs, with */
1560 OP_PLUS, /* 35 the minimizing one second. */
1561 OP_MINPLUS, /* 36 */
1562 OP_QUERY, /* 37 */
1563 OP_MINQUERY, /* 38 */
1564
1565 OP_UPTO, /* 39 From 0 to n matches of one character, caseful*/
1566 OP_MINUPTO, /* 40 */
1567 OP_EXACT, /* 41 Exactly n matches */
1568
1569 OP_POSSTAR, /* 42 Possessified star, caseful */
1570 OP_POSPLUS, /* 43 Possessified plus, caseful */
1571 OP_POSQUERY, /* 44 Posesssified query, caseful */
1572 OP_POSUPTO, /* 45 Possessified upto, caseful */
1573
1574 /**** Single characters, caseless, must follow the caseful ones */
1575
1576 OP_STARI, /* 46 */
1577 OP_MINSTARI, /* 47 */
1578 OP_PLUSI, /* 48 */
1579 OP_MINPLUSI, /* 49 */
1580 OP_QUERYI, /* 50 */
1581 OP_MINQUERYI, /* 51 */
1582
1583 OP_UPTOI, /* 52 From 0 to n matches of one character, caseless */
1584 OP_MINUPTOI, /* 53 */
1585 OP_EXACTI, /* 54 */
1586
1587 OP_POSSTARI, /* 55 Possessified star, caseless */
1588 OP_POSPLUSI, /* 56 Possessified plus, caseless */
1589 OP_POSQUERYI, /* 57 Posesssified query, caseless */
1590 OP_POSUPTOI, /* 58 Possessified upto, caseless */
1591
1592 /**** The negated ones must follow the non-negated ones, and match them ****/
1593 /**** Negated single character, caseful; must precede the caseless ones ****/
1594
1595 OP_NOTSTAR, /* 59 The maximizing and minimizing versions of */
1596 OP_NOTMINSTAR, /* 60 these six opcodes must come in pairs, with */
1597 OP_NOTPLUS, /* 61 the minimizing one second. They must be in */
1598 OP_NOTMINPLUS, /* 62 exactly the same order as those above. */
1599 OP_NOTQUERY, /* 63 */
1600 OP_NOTMINQUERY, /* 64 */
1601
1602 OP_NOTUPTO, /* 65 From 0 to n matches, caseful */
1603 OP_NOTMINUPTO, /* 66 */
1604 OP_NOTEXACT, /* 67 Exactly n matches */
1605
1606 OP_NOTPOSSTAR, /* 68 Possessified versions, caseful */
1607 OP_NOTPOSPLUS, /* 69 */
1608 OP_NOTPOSQUERY, /* 70 */
1609 OP_NOTPOSUPTO, /* 71 */
1610
1611 /**** Negated single character, caseless; must follow the caseful ones ****/
1612
1613 OP_NOTSTARI, /* 72 */
1614 OP_NOTMINSTARI, /* 73 */
1615 OP_NOTPLUSI, /* 74 */
1616 OP_NOTMINPLUSI, /* 75 */
1617 OP_NOTQUERYI, /* 76 */
1618 OP_NOTMINQUERYI, /* 77 */
1619
1620 OP_NOTUPTOI, /* 78 From 0 to n matches, caseless */
1621 OP_NOTMINUPTOI, /* 79 */
1622 OP_NOTEXACTI, /* 80 Exactly n matches */
1623
1624 OP_NOTPOSSTARI, /* 81 Possessified versions, caseless */
1625 OP_NOTPOSPLUSI, /* 82 */
1626 OP_NOTPOSQUERYI, /* 83 */
1627 OP_NOTPOSUPTOI, /* 84 */
1628
1629 /**** Character types ****/
1630
1631 OP_TYPESTAR, /* 85 The maximizing and minimizing versions of */
1632 OP_TYPEMINSTAR, /* 86 these six opcodes must come in pairs, with */
1633 OP_TYPEPLUS, /* 87 the minimizing one second. These codes must */
1634 OP_TYPEMINPLUS, /* 88 be in exactly the same order as those above. */
1635 OP_TYPEQUERY, /* 89 */
1636 OP_TYPEMINQUERY, /* 90 */
1637
1638 OP_TYPEUPTO, /* 91 From 0 to n matches */
1639 OP_TYPEMINUPTO, /* 92 */
1640 OP_TYPEEXACT, /* 93 Exactly n matches */
1641
1642 OP_TYPEPOSSTAR, /* 94 Possessified versions */
1643 OP_TYPEPOSPLUS, /* 95 */
1644 OP_TYPEPOSQUERY, /* 96 */
1645 OP_TYPEPOSUPTO, /* 97 */
1646
1647 /* These are used for character classes and back references; only the
1648 first six are the same as the sets above. */
1649
1650 OP_CRSTAR, /* 98 The maximizing and minimizing versions of */
1651 OP_CRMINSTAR, /* 99 all these opcodes must come in pairs, with */
1652 OP_CRPLUS, /* 100 the minimizing one second. These codes must */
1653 OP_CRMINPLUS, /* 101 be in exactly the same order as those above. */
1654 OP_CRQUERY, /* 102 */
1655 OP_CRMINQUERY, /* 103 */
1656
1657 OP_CRRANGE, /* 104 These are different to the three sets above. */
1658 OP_CRMINRANGE, /* 105 */
1659
1660 /* End of quantifier opcodes */
1661
1662 OP_CLASS, /* 106 Match a character class, chars < 256 only */
1663 OP_NCLASS, /* 107 Same, but the bitmap was created from a negative
1664 class - the difference is relevant only when a
1665 character > 255 is encountered. */
1666 OP_XCLASS, /* 108 Extended class for handling > 255 chars within the
1667 class. This does both positive and negative. */
1668 OP_REF, /* 109 Match a back reference, casefully */
1669 OP_REFI, /* 110 Match a back reference, caselessly */
1670 OP_RECURSE, /* 111 Match a numbered subpattern (possibly recursive) */
1671 OP_CALLOUT, /* 112 Call out to external function if provided */
1672
1673 OP_ALT, /* 113 Start of alternation */
1674 OP_KET, /* 114 End of group that doesn't have an unbounded repeat */
1675 OP_KETRMAX, /* 115 These two must remain together and in this */
1676 OP_KETRMIN, /* 116 order. They are for groups the repeat for ever. */
1677 OP_KETRPOS, /* 117 Possessive unlimited repeat. */
1678
1679 /* The assertions must come before BRA, CBRA, ONCE, and COND, and the four
1680 asserts must remain in order. */
1681
1682 OP_REVERSE, /* 118 Move pointer back - used in lookbehind assertions */
1683 OP_ASSERT, /* 119 Positive lookahead */
1684 OP_ASSERT_NOT, /* 120 Negative lookahead */
1685 OP_ASSERTBACK, /* 121 Positive lookbehind */
1686 OP_ASSERTBACK_NOT, /* 122 Negative lookbehind */
1687
1688 /* ONCE, ONCE_NC, BRA, BRAPOS, CBRA, CBRAPOS, and COND must come immediately
1689 after the assertions, with ONCE first, as there's a test for >= ONCE for a
1690 subpattern that isn't an assertion. The POS versions must immediately follow
1691 the non-POS versions in each case. */
1692
1693 OP_ONCE, /* 123 Atomic group, contains captures */
1694 OP_ONCE_NC, /* 124 Atomic group containing no captures */
1695 OP_BRA, /* 125 Start of non-capturing bracket */
1696 OP_BRAPOS, /* 126 Ditto, with unlimited, possessive repeat */
1697 OP_CBRA, /* 127 Start of capturing bracket */
1698 OP_CBRAPOS, /* 128 Ditto, with unlimited, possessive repeat */
1699 OP_COND, /* 129 Conditional group */
1700
1701 /* These five must follow the previous five, in the same order. There's a
1702 check for >= SBRA to distinguish the two sets. */
1703
1704 OP_SBRA, /* 130 Start of non-capturing bracket, check empty */
1705 OP_SBRAPOS, /* 131 Ditto, with unlimited, possessive repeat */
1706 OP_SCBRA, /* 132 Start of capturing bracket, check empty */
1707 OP_SCBRAPOS, /* 133 Ditto, with unlimited, possessive repeat */
1708 OP_SCOND, /* 134 Conditional group, check empty */
1709
1710 /* The next two pairs must (respectively) be kept together. */
1711
1712 OP_CREF, /* 135 Used to hold a capture number as condition */
1713 OP_NCREF, /* 136 Same, but generated by a name reference*/
1714 OP_RREF, /* 137 Used to hold a recursion number as condition */
1715 OP_NRREF, /* 138 Same, but generated by a name reference*/
1716 OP_DEF, /* 139 The DEFINE condition */
1717
1718 OP_BRAZERO, /* 140 These two must remain together and in this */
1719 OP_BRAMINZERO, /* 141 order. */
1720 OP_BRAPOSZERO, /* 142 */
1721
1722 /* These are backtracking control verbs */
1723
1724 OP_MARK, /* 143 always has an argument */
1725 OP_PRUNE, /* 144 */
1726 OP_PRUNE_ARG, /* 145 same, but with argument */
1727 OP_SKIP, /* 146 */
1728 OP_SKIP_ARG, /* 147 same, but with argument */
1729 OP_THEN, /* 148 */
1730 OP_THEN_ARG, /* 149 same, but with argument */
1731 OP_COMMIT, /* 150 */
1732
1733 /* These are forced failure and success verbs */
1734
1735 OP_FAIL, /* 151 */
1736 OP_ACCEPT, /* 152 */
1737 OP_ASSERT_ACCEPT, /* 153 Used inside assertions */
1738 OP_CLOSE, /* 154 Used before OP_ACCEPT to close open captures */
1739
1740 /* This is used to skip a subpattern with a {0} quantifier */
1741
1742 OP_SKIPZERO, /* 155 */
1743
1744 /* This is not an opcode, but is used to check that tables indexed by opcode
1745 are the correct length, in order to catch updating errors - there have been
1746 some in the past. */
1747
1748 OP_TABLE_LENGTH
1749 };
1750
1751 /* *** NOTE NOTE NOTE *** Whenever the list above is updated, the two macro
1752 definitions that follow must also be updated to match. There are also tables
1753 called "coptable" and "poptable" in pcre_dfa_exec.c that must be updated. */
1754
1755
1756 /* This macro defines textual names for all the opcodes. These are used only
1757 for debugging, and some of them are only partial names. The macro is referenced
1758 only in pcre_printint.c, which fills out the full names in many cases (and in
1759 some cases doesn't actually use these names at all). */
1760
1761 #define OP_NAME_LIST \
1762 "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d", \
1763 "\\S", "\\s", "\\W", "\\w", "Any", "AllAny", "Anybyte", \
1764 "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v", \
1765 "extuni", "\\Z", "\\z", \
1766 "^", "^", "$", "$", "char", "chari", "not", "noti", \
1767 "*", "*?", "+", "+?", "?", "??", \
1768 "{", "{", "{", \
1769 "*+","++", "?+", "{", \
1770 "*", "*?", "+", "+?", "?", "??", \
1771 "{", "{", "{", \
1772 "*+","++", "?+", "{", \
1773 "*", "*?", "+", "+?", "?", "??", \
1774 "{", "{", "{", \
1775 "*+","++", "?+", "{", \
1776 "*", "*?", "+", "+?", "?", "??", \
1777 "{", "{", "{", \
1778 "*+","++", "?+", "{", \
1779 "*", "*?", "+", "+?", "?", "??", "{", "{", "{", \
1780 "*+","++", "?+", "{", \
1781 "*", "*?", "+", "+?", "?", "??", "{", "{", \
1782 "class", "nclass", "xclass", "Ref", "Refi", \
1783 "Recurse", "Callout", \
1784 "Alt", "Ket", "KetRmax", "KetRmin", "KetRpos", \
1785 "Reverse", "Assert", "Assert not", "AssertB", "AssertB not", \
1786 "Once", "Once_NC", \
1787 "Bra", "BraPos", "CBra", "CBraPos", \
1788 "Cond", \
1789 "SBra", "SBraPos", "SCBra", "SCBraPos", \
1790 "SCond", \
1791 "Cond ref", "Cond nref", "Cond rec", "Cond nrec", "Cond def", \
1792 "Brazero", "Braminzero", "Braposzero", \
1793 "*MARK", "*PRUNE", "*PRUNE", "*SKIP", "*SKIP", \
1794 "*THEN", "*THEN", "*COMMIT", "*FAIL", \
1795 "*ACCEPT", "*ASSERT_ACCEPT", \
1796 "Close", "Skip zero"
1797
1798
1799 /* This macro defines the length of fixed length operations in the compiled
1800 regex. The lengths are used when searching for specific things, and also in the
1801 debugging printing of a compiled regex. We use a macro so that it can be
1802 defined close to the definitions of the opcodes themselves.
1803
1804 As things have been extended, some of these are no longer fixed lenths, but are
1805 minima instead. For example, the length of a single-character repeat may vary
1806 in UTF-8 mode. The code that uses this table must know about such things. */
1807
1808 #define OP_LENGTHS \
1809 1, /* End */ \
1810 1, 1, 1, 1, 1, /* \A, \G, \K, \B, \b */ \
1811 1, 1, 1, 1, 1, 1, /* \D, \d, \S, \s, \W, \w */ \
1812 1, 1, 1, /* Any, AllAny, Anybyte */ \
1813 3, 3, /* \P, \p */ \
1814 1, 1, 1, 1, 1, /* \R, \H, \h, \V, \v */ \
1815 1, /* \X */ \
1816 1, 1, 1, 1, 1, 1, /* \Z, \z, ^, ^M, $, $M */ \
1817 2, /* Char - the minimum length */ \
1818 2, /* Chari - the minimum length */ \
1819 2, /* not */ \
1820 2, /* noti */ \
1821 /* Positive single-char repeats ** These are */ \
1822 2, 2, 2, 2, 2, 2, /* *, *?, +, +?, ?, ?? ** minima in */ \
1823 2+IMM2_SIZE, 2+IMM2_SIZE, /* upto, minupto ** mode */ \
1824 2+IMM2_SIZE, /* exact */ \
1825 2, 2, 2, 2+IMM2_SIZE, /* *+, ++, ?+, upto+ */ \
1826 2, 2, 2, 2, 2, 2, /* *I, *?I, +I, +?I, ?I, ??I ** UTF-8 */ \
1827 2+IMM2_SIZE, 2+IMM2_SIZE, /* upto I, minupto I */ \
1828 2+IMM2_SIZE, /* exact I */ \
1829 2, 2, 2, 2+IMM2_SIZE, /* *+I, ++I, ?+I, upto+I */ \
1830 /* Negative single-char repeats - only for chars < 256 */ \
1831 2, 2, 2, 2, 2, 2, /* NOT *, *?, +, +?, ?, ?? */ \
1832 2+IMM2_SIZE, 2+IMM2_SIZE, /* NOT upto, minupto */ \
1833 2+IMM2_SIZE, /* NOT exact */ \
1834 2, 2, 2, 2+IMM2_SIZE, /* Possessive NOT *, +, ?, upto */ \
1835 2, 2, 2, 2, 2, 2, /* NOT *I, *?I, +I, +?I, ?I, ??I */ \
1836 2+IMM2_SIZE, 2+IMM2_SIZE, /* NOT upto I, minupto I */ \
1837 2+IMM2_SIZE, /* NOT exact I */ \
1838 2, 2, 2, 2+IMM2_SIZE, /* Possessive NOT *I, +I, ?I, upto I */ \
1839 /* Positive type repeats */ \
1840 2, 2, 2, 2, 2, 2, /* Type *, *?, +, +?, ?, ?? */ \
1841 2+IMM2_SIZE, 2+IMM2_SIZE, /* Type upto, minupto */ \
1842 2+IMM2_SIZE, /* Type exact */ \
1843 2, 2, 2, 2+IMM2_SIZE, /* Possessive *+, ++, ?+, upto+ */ \
1844 /* Character class & ref repeats */ \
1845 1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ \
1846 1+2*IMM2_SIZE, 1+2*IMM2_SIZE, /* CRRANGE, CRMINRANGE */ \
1847 1+(32/sizeof(pcre_uchar)), /* CLASS */ \
1848 1+(32/sizeof(pcre_uchar)), /* NCLASS */ \
1849 0, /* XCLASS - variable length */ \
1850 1+IMM2_SIZE, /* REF */ \
1851 1+IMM2_SIZE, /* REFI */ \
1852 1+LINK_SIZE, /* RECURSE */ \
1853 2+2*LINK_SIZE, /* CALLOUT */ \
1854 1+LINK_SIZE, /* Alt */ \
1855 1+LINK_SIZE, /* Ket */ \
1856 1+LINK_SIZE, /* KetRmax */ \
1857 1+LINK_SIZE, /* KetRmin */ \
1858 1+LINK_SIZE, /* KetRpos */ \
1859 1+LINK_SIZE, /* Reverse */ \
1860 1+LINK_SIZE, /* Assert */ \
1861 1+LINK_SIZE, /* Assert not */ \
1862 1+LINK_SIZE, /* Assert behind */ \
1863 1+LINK_SIZE, /* Assert behind not */ \
1864 1+LINK_SIZE, /* ONCE */ \
1865 1+LINK_SIZE, /* ONCE_NC */ \
1866 1+LINK_SIZE, /* BRA */ \
1867 1+LINK_SIZE, /* BRAPOS */ \
1868 1+LINK_SIZE+IMM2_SIZE, /* CBRA */ \
1869 1+LINK_SIZE+IMM2_SIZE, /* CBRAPOS */ \
1870 1+LINK_SIZE, /* COND */ \
1871 1+LINK_SIZE, /* SBRA */ \
1872 1+LINK_SIZE, /* SBRAPOS */ \
1873 1+LINK_SIZE+IMM2_SIZE, /* SCBRA */ \
1874 1+LINK_SIZE+IMM2_SIZE, /* SCBRAPOS */ \
1875 1+LINK_SIZE, /* SCOND */ \
1876 1+IMM2_SIZE, 1+IMM2_SIZE, /* CREF, NCREF */ \
1877 1+IMM2_SIZE, 1+IMM2_SIZE, /* RREF, NRREF */ \
1878 1, /* DEF */ \
1879 1, 1, 1, /* BRAZERO, BRAMINZERO, BRAPOSZERO */ \
1880 3, 1, 3, /* MARK, PRUNE, PRUNE_ARG */ \
1881 1, 3, /* SKIP, SKIP_ARG */ \
1882 1, 3, /* THEN, THEN_ARG */ \
1883 1, 1, 1, 1, /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT */ \
1884 1+IMM2_SIZE, 1 /* CLOSE, SKIPZERO */
1885
1886 /* A magic value for OP_RREF and OP_NRREF to indicate the "any recursion"
1887 condition. */
1888
1889 #define RREF_ANY 0xffff
1890
1891 /* Compile time error code numbers. They are given names so that they can more
1892 easily be tracked. When a new number is added, the table called eint in
1893 pcreposix.c must be updated. */
1894
1895 enum { ERR0, ERR1, ERR2, ERR3, ERR4, ERR5, ERR6, ERR7, ERR8, ERR9,
1896 ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,
1897 ERR20, ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR28, ERR29,
1898 ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,
1899 ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,
1900 ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,
1901 ERR60, ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERR68, ERR69,
1902 ERR70, ERR71, ERRCOUNT };
1903
1904 /* The real format of the start of the pcre block; the index of names and the
1905 code vector run on as long as necessary after the end. We store an explicit
1906 offset to the name table so that if a regex is compiled on one host, saved, and
1907 then run on another where the size of pointers is different, all might still
1908 be well. For the case of compiled-on-4 and run-on-8, we include an extra
1909 pointer that is always NULL. For future-proofing, a few dummy fields were
1910 originally included - even though you can never get this planning right - but
1911 there is only one left now.
1912
1913 NOTE NOTE NOTE:
1914 Because people can now save and re-use compiled patterns, any additions to this
1915 structure should be made at the end, and something earlier (e.g. a new
1916 flag in the options or one of the dummy fields) should indicate that the new
1917 fields are present. Currently PCRE always sets the dummy fields to zero.
1918 NOTE NOTE NOTE
1919 */
1920
1921 typedef struct real_pcre {
1922 pcre_uint32 magic_number;
1923 pcre_uint32 size; /* Total that was malloced */
1924 pcre_uint32 options; /* Public options */
1925 pcre_uint16 flags; /* Private flags */
1926 pcre_uint16 dummy1; /* For future use */
1927 pcre_uint16 top_bracket;
1928 pcre_uint16 top_backref;
1929 pcre_uint16 first_char; /* Starting character */
1930 pcre_uint16 req_char; /* This character must be seen */
1931 pcre_uint16 name_table_offset; /* Offset to name table that follows */
1932 pcre_uint16 name_entry_size; /* Size of any name items */
1933 pcre_uint16 name_count; /* Number of name items */
1934 pcre_uint16 ref_count; /* Reference count */
1935
1936 const pcre_uint8 *tables; /* Pointer to tables or NULL for std */
1937 const pcre_uint8 *nullpad; /* NULL padding */
1938 } real_pcre;
1939
1940 /* The format of the block used to store data from pcre_study(). The same
1941 remark (see NOTE above) about extending this structure applies. */
1942
1943 typedef struct pcre_study_data {
1944 pcre_uint32 size; /* Total that was malloced */
1945 pcre_uint32 flags; /* Private flags */
1946 pcre_uint8 start_bits[32]; /* Starting char bits */
1947 pcre_uint32 minlength; /* Minimum subject length */
1948 } pcre_study_data;
1949
1950 /* Structure for building a chain of open capturing subpatterns during
1951 compiling, so that instructions to close them can be compiled when (*ACCEPT) is
1952 encountered. This is also used to identify subpatterns that contain recursive
1953 back references to themselves, so that they can be made atomic. */
1954
1955 typedef struct open_capitem {
1956 struct open_capitem *next; /* Chain link */
1957 pcre_uint16 number; /* Capture number */
1958 pcre_uint16 flag; /* Set TRUE if recursive back ref */
1959 } open_capitem;
1960
1961 /* Structure for passing "static" information around between the functions
1962 doing the compiling, so that they are thread-safe. */
1963
1964 typedef struct compile_data {
1965 const pcre_uint8 *lcc; /* Points to lower casing table */
1966 const pcre_uint8 *fcc; /* Points to case-flipping table */
1967 const pcre_uint8 *cbits; /* Points to character type table */
1968 const pcre_uint8 *ctypes; /* Points to table of type maps */
1969 const pcre_uchar *start_workspace;/* The start of working space */
1970 const pcre_uchar *start_code; /* The start of the compiled code */
1971 const pcre_uchar *start_pattern; /* The start of the pattern */
1972 const pcre_uchar *end_pattern; /* The end of the pattern */
1973 open_capitem *open_caps; /* Chain of open capture items */
1974 pcre_uchar *hwm; /* High watermark of workspace */
1975 pcre_uchar *name_table; /* The name/number table */
1976 int names_found; /* Number of entries so far */
1977 int name_entry_size; /* Size of each entry */
1978 int bracount; /* Count of capturing parens as we compile */
1979 int final_bracount; /* Saved value after first pass */
1980 int top_backref; /* Maximum back reference */
1981 unsigned int backref_map; /* Bitmap of low back refs */
1982 int assert_depth; /* Depth of nested assertions */
1983 int external_options; /* External (initial) options */
1984 int external_flags; /* External flag bits to be set */
1985 int req_varyopt; /* "After variable item" flag for reqbyte */
1986 BOOL had_accept; /* (*ACCEPT) encountered */
1987 BOOL check_lookbehind; /* Lookbehinds need later checking */
1988 int nltype; /* Newline type */
1989 int nllen; /* Newline string length */
1990 pcre_uchar nl[4]; /* Newline string when fixed length */
1991 } compile_data;
1992
1993 /* Structure for maintaining a chain of pointers to the currently incomplete
1994 branches, for testing for left recursion while compiling. */
1995
1996 typedef struct branch_chain {
1997 struct branch_chain *outer;
1998 pcre_uchar *current_branch;
1999 } branch_chain;
2000
2001 /* Structure for items in a linked list that represents an explicit recursive
2002 call within the pattern; used by pcre_exec(). */
2003
2004 typedef struct recursion_info {
2005 struct recursion_info *prevrec; /* Previous recursion record (or NULL) */
2006 int group_num; /* Number of group that was called */
2007 int *offset_save; /* Pointer to start of saved offsets */
2008 int saved_max; /* Number of saved offsets */
2009 PCRE_PUCHAR subject_position; /* Position at start of recursion */
2010 } recursion_info;
2011
2012 /* A similar structure for pcre_dfa_exec(). */
2013
2014 typedef struct dfa_recursion_info {
2015 struct dfa_recursion_info *prevrec;
2016 int group_num;
2017 PCRE_PUCHAR subject_position;
2018 } dfa_recursion_info;
2019
2020 /* Structure for building a chain of data for holding the values of the subject
2021 pointer at the start of each subpattern, so as to detect when an empty string
2022 has been matched by a subpattern - to break infinite loops; used by
2023 pcre_exec(). */
2024
2025 typedef struct eptrblock {
2026 struct eptrblock *epb_prev;
2027 PCRE_PUCHAR epb_saved_eptr;
2028 } eptrblock;
2029
2030
2031 /* Structure for passing "static" information around between the functions
2032 doing traditional NFA matching, so that they are thread-safe. */
2033
2034 typedef struct match_data {
2035 unsigned long int match_call_count; /* As it says */
2036 unsigned long int match_limit; /* As it says */
2037 unsigned long int match_limit_recursion; /* As it says */
2038 int *offset_vector; /* Offset vector */
2039 int offset_end; /* One past the end */
2040 int offset_max; /* The maximum usable for return data */
2041 int nltype; /* Newline type */
2042 int nllen; /* Newline string length */
2043 int name_count; /* Number of names in name table */
2044 int name_entry_size; /* Size of entry in names table */
2045 pcre_uchar *name_table; /* Table of names */
2046 pcre_uchar nl[4]; /* Newline string when fixed */
2047 const pcre_uint8 *lcc; /* Points to lower casing table */
2048 const pcre_uint8 *ctypes; /* Points to table of type maps */
2049 BOOL offset_overflow; /* Set if too many extractions */
2050 BOOL notbol; /* NOTBOL flag */
2051 BOOL noteol; /* NOTEOL flag */
2052 BOOL utf; /* UTF-8 / UTF-16 flag */
2053 BOOL jscript_compat; /* JAVASCRIPT_COMPAT flag */
2054 BOOL use_ucp; /* PCRE_UCP flag */
2055 BOOL endonly; /* Dollar not before final \n */
2056 BOOL notempty; /* Empty string match not wanted */
2057 BOOL notempty_atstart; /* Empty string match at start not wanted */
2058 BOOL hitend; /* Hit the end of the subject at some point */
2059 BOOL bsr_anycrlf; /* \R is just any CRLF, not full Unicode */
2060 BOOL hasthen; /* Pattern contains (*THEN) */
2061 const pcre_uchar *start_code; /* For use when recursing */
2062 PCRE_PUCHAR start_subject; /* Start of the subject string */
2063 PCRE_PUCHAR end_subject; /* End of the subject string */
2064 PCRE_PUCHAR start_match_ptr; /* Start of matched string */
2065 PCRE_PUCHAR end_match_ptr; /* Subject position at end match */
2066 PCRE_PUCHAR start_used_ptr; /* Earliest consulted character */
2067 int partial; /* PARTIAL options */
2068 int end_offset_top; /* Highwater mark at end of match */
2069 int capture_last; /* Most recent capture number */
2070 int start_offset; /* The start offset value */
2071 int match_function_type; /* Set for certain special calls of MATCH() */
2072 eptrblock *eptrchain; /* Chain of eptrblocks for tail recursions */
2073 int eptrn; /* Next free eptrblock */
2074 recursion_info *recursive; /* Linked list of recursion data */
2075 void *callout_data; /* To pass back to callouts */
2076 const pcre_uchar *mark; /* Mark pointer to pass back */
2077 const pcre_uchar *once_target; /* Where to back up to for atomic groups */
2078 } match_data;
2079
2080 /* A similar structure is used for the same purpose by the DFA matching
2081 functions. */
2082
2083 typedef struct dfa_match_data {
2084 const pcre_uchar *start_code; /* Start of the compiled pattern */
2085 const pcre_uchar *start_subject ; /* Start of the subject string */
2086 const pcre_uchar *end_subject; /* End of subject string */
2087 const pcre_uchar *start_used_ptr; /* Earliest consulted character */
2088 const pcre_uint8 *tables; /* Character tables */
2089 int start_offset; /* The start offset value */
2090 int moptions; /* Match options */
2091 int poptions; /* Pattern options */
2092 int nltype; /* Newline type */
2093 int nllen; /* Newline string length */
2094 pcre_uchar nl[4]; /* Newline string when fixed */
2095 void *callout_data; /* To pass back to callouts */
2096 dfa_recursion_info *recursive; /* Linked list of recursion data */
2097 } dfa_match_data;
2098
2099 /* Bit definitions for entries in the pcre_ctypes table. */
2100
2101 #define ctype_space 0x01
2102 #define ctype_letter 0x02
2103 #define ctype_digit 0x04
2104 #define ctype_xdigit 0x08
2105 #define ctype_word 0x10 /* alphanumeric or '_' */
2106 #define ctype_meta 0x80 /* regexp meta char or zero (end pattern) */
2107
2108 /* Offsets for the bitmap tables in pcre_cbits. Each table contains a set
2109 of bits for a class map. Some classes are built by combining these tables. */
2110
2111 #define cbit_space 0 /* [:space:] or \s */
2112 #define cbit_xdigit 32 /* [:xdigit:] */
2113 #define cbit_digit 64 /* [:digit:] or \d */
2114 #define cbit_upper 96 /* [:upper:] */
2115 #define cbit_lower 128 /* [:lower:] */
2116 #define cbit_word 160 /* [:word:] or \w */
2117 #define cbit_graph 192 /* [:graph:] */
2118 #define cbit_print 224 /* [:print:] */
2119 #define cbit_punct 256 /* [:punct:] */
2120 #define cbit_cntrl 288 /* [:cntrl:] */
2121 #define cbit_length 320 /* Length of the cbits table */
2122
2123 /* Offsets of the various tables from the base tables pointer, and
2124 total length. */
2125
2126 #define lcc_offset 0
2127 #define fcc_offset 256
2128 #define cbits_offset 512
2129 #define ctypes_offset (cbits_offset + cbit_length)
2130 #define tables_length (ctypes_offset + 256)
2131
2132 /* Internal function prefix */
2133
2134 #ifdef COMPILE_PCRE8
2135 #define PRIV(name) _pcre_##name
2136 #else
2137 #ifdef COMPILE_PCRE16
2138 #define PRIV(name) _pcre16_##name
2139 #else
2140 #error Unsupported compiling mode
2141 #endif /* COMPILE_PCRE16 */
2142 #endif /* COMPILE_PCRE8 */
2143
2144 /* Layout of the UCP type table that translates property names into types and
2145 codes. Each entry used to point directly to a name, but to reduce the number of
2146 relocations in shared libraries, it now has an offset into a single string
2147 instead. */
2148
2149 typedef struct {
2150 pcre_uint16 name_offset;
2151 pcre_uint16 type;
2152 pcre_uint16 value;
2153 } ucp_type_table;
2154
2155
2156 /* Internal shared data tables. These are tables that are used by more than one
2157 of the exported public functions. They have to be "external" in the C sense,
2158 but are not part of the PCRE public API. The data for these tables is in the
2159 pcre_tables.c module. */
2160
2161 #ifdef COMPILE_PCRE8
2162
2163 extern const int PRIV(utf8_table1)[];
2164 extern const int PRIV(utf8_table1_size);
2165 extern const int PRIV(utf8_table2)[];
2166 extern const int PRIV(utf8_table3)[];
2167 extern const pcre_uint8 PRIV(utf8_table4)[];
2168
2169 #endif /* COMPILE_PCRE8 */
2170
2171 extern const char PRIV(utt_names)[];
2172 extern const ucp_type_table PRIV(utt)[];
2173 extern const int PRIV(utt_size);
2174
2175 extern const pcre_uint8 PRIV(default_tables)[];
2176
2177 extern const pcre_uint8 PRIV(OP_lengths)[];
2178
2179
2180 /* Internal shared functions. These are functions that are used by more than
2181 one of the exported public functions. They have to be "external" in the C
2182 sense, but are not part of the PCRE public API. */
2183
2184 /* String comparison functions. */
2185 #ifdef COMPILE_PCRE8
2186
2187 #define STRCMP_UC_UC(str1, str2) \
2188 strcmp((char *)(str1), (char *)(str2))
2189 #define STRCMP_UC_C8(str1, str2) \
2190 strcmp((char *)(str1), (str2))
2191 #define STRNCMP_UC_UC(str1, str2, num) \
2192 strncmp((char *)(str1), (char *)(str2), (num))
2193 #define STRNCMP_UC_C8(str1, str2, num) \
2194 strncmp((char *)(str1), (str2), (num))
2195 #define STRLEN_UC(str) strlen((const char *)str)
2196
2197 #else
2198
2199 extern int PRIV(strcmp_uc_uc)(const pcre_uchar *,
2200 const pcre_uchar *);
2201 extern int PRIV(strcmp_uc_c8)(const pcre_uchar *,
2202 const char *);
2203 extern int PRIV(strncmp_uc_uc)(const pcre_uchar *,
2204 const pcre_uchar *, unsigned int num);
2205 extern int PRIV(strncmp_uc_c8)(const pcre_uchar *,
2206 const char *, unsigned int num);
2207 extern unsigned int PRIV(strlen_uc)(const pcre_uchar *str);
2208
2209 #define STRCMP_UC_UC(str1, str2) \
2210 PRIV(strcmp_uc_uc)((str1), (str2))
2211 #define STRCMP_UC_C8(str1, str2) \
2212 PRIV(strcmp_uc_c8)((str1), (str2))
2213 #define STRNCMP_UC_UC(str1, str2, num) \
2214 PRIV(strncmp_uc_uc)((str1), (str2), (num))
2215 #define STRNCMP_UC_C8(str1, str2, num) \
2216 PRIV(strncmp_uc_c8)((str1), (str2), (num))
2217 #define STRLEN_UC(str) PRIV(strlen_uc)(str)
2218
2219 #endif /* COMPILE_PCRE8 */
2220
2221 extern const pcre_uchar *PRIV(find_bracket)(const pcre_uchar *, BOOL, int);
2222 extern BOOL PRIV(is_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2223 int *, BOOL);
2224 extern int PRIV(ord2utf)(pcre_uint32, pcre_uchar *);
2225 extern real_pcre *PRIV(try_flipped)(const real_pcre *, real_pcre *,
2226 const pcre_study_data *, pcre_study_data *);
2227 extern int PRIV(valid_utf)(PCRE_PUCHAR, int, int *);
2228 extern BOOL PRIV(was_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2229 int *, BOOL);
2230 extern BOOL PRIV(xclass)(int, const pcre_uchar *);
2231
2232 #ifdef SUPPORT_JIT
2233 extern void PRIV(jit_compile)(const real_pcre *, pcre_extra *);
2234 extern int PRIV(jit_exec)(const real_pcre *, void *,
2235 const pcre_uchar *, int, int, int, int, int *, int);
2236 extern void PRIV(jit_free)(void *);
2237 #endif
2238
2239 /* Unicode character database (UCD) */
2240
2241 typedef struct {
2242 pcre_uint8 script;
2243 pcre_uint8 chartype;
2244 pcre_int32 other_case;
2245 } ucd_record;
2246
2247 extern const ucd_record PRIV(ucd_records)[];
2248 extern const pcre_uint8 PRIV(ucd_stage1)[];
2249 extern const pcre_uint16 PRIV(ucd_stage2)[];
2250 extern const int PRIV(ucp_gentype)[];
2251 #ifdef SUPPORT_JIT
2252 extern const int PRIV(ucp_typerange)[];
2253 #endif
2254
2255 /* UCD access macros */
2256
2257 #define UCD_BLOCK_SIZE 128
2258 #define GET_UCD(ch) (PRIV(ucd_records) + \
2259 PRIV(ucd_stage2)[PRIV(ucd_stage1)[(ch) / UCD_BLOCK_SIZE] * \
2260 UCD_BLOCK_SIZE + (ch) % UCD_BLOCK_SIZE])
2261
2262 #define UCD_CHARTYPE(ch) GET_UCD(ch)->chartype
2263 #define UCD_SCRIPT(ch) GET_UCD(ch)->script
2264 #define UCD_CATEGORY(ch) PRIV(ucp_gentype)[UCD_CHARTYPE(ch)]
2265 #define UCD_OTHERCASE(ch) (ch + GET_UCD(ch)->other_case)
2266
2267 #endif
2268
2269 /* 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