1 |
/*************************************************
|
2 |
* Perl-Compatible Regular Expressions *
|
3 |
*************************************************/
|
4 |
|
5 |
/* PCRE2 is a library of functions to support regular expressions whose syntax
|
6 |
and semantics are as close as possible to those of the Perl 5 language.
|
7 |
|
8 |
Written by Philip Hazel
|
9 |
Original API code Copyright (c) 1997-2012 University of Cambridge
|
10 |
New API code Copyright (c) 2016 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 |
/* We do not support both EBCDIC and Unicode at the same time. The "configure"
|
42 |
script prevents both being selected, but not everybody uses "configure". EBCDIC
|
43 |
is only supported for the 8-bit library, but the check for this has to be later
|
44 |
in this file, because the first part is not width-dependent, and is included by
|
45 |
pcre2test.c with CODE_UNIT_WIDTH == 0. */
|
46 |
|
47 |
#if defined EBCDIC && defined SUPPORT_UNICODE
|
48 |
#error The use of both EBCDIC and SUPPORT_UNICODE is not supported.
|
49 |
#endif
|
50 |
|
51 |
/* Standard C headers */
|
52 |
|
53 |
#include <ctype.h>
|
54 |
#include <limits.h>
|
55 |
#include <stddef.h>
|
56 |
#include <stdio.h>
|
57 |
#include <stdlib.h>
|
58 |
#include <string.h>
|
59 |
|
60 |
/* Macros to make boolean values more obvious. The #ifndef is to pacify
|
61 |
compiler warnings in environments where these macros are defined elsewhere.
|
62 |
Unfortunately, there is no way to do the same for the typedef. */
|
63 |
|
64 |
typedef int BOOL;
|
65 |
#ifndef FALSE
|
66 |
#define FALSE 0
|
67 |
#define TRUE 1
|
68 |
#endif
|
69 |
|
70 |
/* Valgrind (memcheck) support */
|
71 |
|
72 |
#ifdef SUPPORT_VALGRIND
|
73 |
#include <valgrind/memcheck.h>
|
74 |
#endif
|
75 |
|
76 |
/* Older versions of MSVC lack snprintf(). This define allows for
|
77 |
warning/error-free compilation and testing with MSVC compilers back to at least
|
78 |
MSVC 10/2010. Except for VC6 (which is missing some fundamentals and fails). */
|
79 |
|
80 |
#if defined(_MSC_VER) && (_MSC_VER < 1900)
|
81 |
#define snprintf _snprintf
|
82 |
#endif
|
83 |
|
84 |
/* When compiling a DLL for Windows, the exported symbols have to be declared
|
85 |
using some MS magic. I found some useful information on this web page:
|
86 |
http://msdn2.microsoft.com/en-us/library/y4h7bcy6(VS.80).aspx. According to the
|
87 |
information there, using __declspec(dllexport) without "extern" we have a
|
88 |
definition; with "extern" we have a declaration. The settings here override the
|
89 |
setting in pcre2.h (which is included below); it defines only PCRE2_EXP_DECL,
|
90 |
which is all that is needed for applications (they just import the symbols). We
|
91 |
use:
|
92 |
|
93 |
PCRE2_EXP_DECL for declarations
|
94 |
PCRE2_EXP_DEFN for definitions
|
95 |
|
96 |
The reason for wrapping this in #ifndef PCRE2_EXP_DECL is so that pcre2test,
|
97 |
which is an application, but needs to import this file in order to "peek" at
|
98 |
internals, can #include pcre2.h first to get an application's-eye view.
|
99 |
|
100 |
In principle, people compiling for non-Windows, non-Unix-like (i.e. uncommon,
|
101 |
special-purpose environments) might want to stick other stuff in front of
|
102 |
exported symbols. That's why, in the non-Windows case, we set PCRE2_EXP_DEFN
|
103 |
only if it is not already set. */
|
104 |
|
105 |
#ifndef PCRE2_EXP_DECL
|
106 |
# ifdef _WIN32
|
107 |
# ifndef PCRE2_STATIC
|
108 |
# define PCRE2_EXP_DECL extern __declspec(dllexport)
|
109 |
# define PCRE2_EXP_DEFN __declspec(dllexport)
|
110 |
# else
|
111 |
# define PCRE2_EXP_DECL extern
|
112 |
# define PCRE2_EXP_DEFN
|
113 |
# endif
|
114 |
# else
|
115 |
# ifdef __cplusplus
|
116 |
# define PCRE2_EXP_DECL extern "C"
|
117 |
# else
|
118 |
# define PCRE2_EXP_DECL extern
|
119 |
# endif
|
120 |
# ifndef PCRE2_EXP_DEFN
|
121 |
# define PCRE2_EXP_DEFN PCRE2_EXP_DECL
|
122 |
# endif
|
123 |
# endif
|
124 |
#endif
|
125 |
|
126 |
/* Include the public PCRE2 header and the definitions of UCP character
|
127 |
property values. This must follow the setting of PCRE2_EXP_DECL above. */
|
128 |
|
129 |
#include "pcre2.h"
|
130 |
#include "pcre2_ucp.h"
|
131 |
|
132 |
/* When PCRE2 is compiled as a C++ library, the subject pointer can be replaced
|
133 |
with a custom type. This makes it possible, for example, to allow pcre2_match()
|
134 |
to process subject strings that are discontinuous by using a smart pointer
|
135 |
class. It must always be possible to inspect all of the subject string in
|
136 |
pcre2_match() because of the way it backtracks. */
|
137 |
|
138 |
/* WARNING: This is as yet untested for PCRE2. */
|
139 |
|
140 |
#ifdef CUSTOM_SUBJECT_PTR
|
141 |
#undef PCRE2_SPTR
|
142 |
#define PCRE2_SPTR CUSTOM_SUBJECT_PTR
|
143 |
#endif
|
144 |
|
145 |
/* When checking for integer overflow in pcre2_compile(), we need to handle
|
146 |
large integers. If a 64-bit integer type is available, we can use that.
|
147 |
Otherwise we have to cast to double, which of course requires floating point
|
148 |
arithmetic. Handle this by defining a macro for the appropriate type. If
|
149 |
stdint.h is available, include it; it may define INT64_MAX. Systems that do not
|
150 |
have stdint.h (e.g. Solaris) may have inttypes.h. The macro int64_t may be set
|
151 |
by "configure". */
|
152 |
|
153 |
#if defined HAVE_STDINT_H
|
154 |
#include <stdint.h>
|
155 |
#elif defined HAVE_INTTYPES_H
|
156 |
#include <inttypes.h>
|
157 |
#endif
|
158 |
|
159 |
#if defined INT64_MAX || defined int64_t
|
160 |
#define INT64_OR_DOUBLE int64_t
|
161 |
#else
|
162 |
#define INT64_OR_DOUBLE double
|
163 |
#endif
|
164 |
|
165 |
/* When compiling for use with the Virtual Pascal compiler, these functions
|
166 |
need to have their names changed. PCRE2 must be compiled with the -DVPCOMPAT
|
167 |
option on the command line. */
|
168 |
|
169 |
#ifdef VPCOMPAT
|
170 |
#define strlen(s) _strlen(s)
|
171 |
#define strncmp(s1,s2,m) _strncmp(s1,s2,m)
|
172 |
#define memcmp(s,c,n) _memcmp(s,c,n)
|
173 |
#define memcpy(d,s,n) _memcpy(d,s,n)
|
174 |
#define memmove(d,s,n) _memmove(d,s,n)
|
175 |
#define memset(s,c,n) _memset(s,c,n)
|
176 |
#else /* VPCOMPAT */
|
177 |
|
178 |
/* To cope with SunOS4 and other systems that lack memmove() but have bcopy(),
|
179 |
define a macro for memmove() if HAVE_MEMMOVE is false, provided that HAVE_BCOPY
|
180 |
is set. Otherwise, include an emulating function for those systems that have
|
181 |
neither (there some non-Unix environments where this is the case). */
|
182 |
|
183 |
#ifndef HAVE_MEMMOVE
|
184 |
#undef memmove /* some systems may have a macro */
|
185 |
#ifdef HAVE_BCOPY
|
186 |
#define memmove(a, b, c) bcopy(b, a, c)
|
187 |
#else /* HAVE_BCOPY */
|
188 |
static void *
|
189 |
pcre2_memmove(void *d, const void *s, size_t n)
|
190 |
{
|
191 |
size_t i;
|
192 |
unsigned char *dest = (unsigned char *)d;
|
193 |
const unsigned char *src = (const unsigned char *)s;
|
194 |
if (dest > src)
|
195 |
{
|
196 |
dest += n;
|
197 |
src += n;
|
198 |
for (i = 0; i < n; ++i) *(--dest) = *(--src);
|
199 |
return (void *)dest;
|
200 |
}
|
201 |
else
|
202 |
{
|
203 |
for (i = 0; i < n; ++i) *dest++ = *src++;
|
204 |
return (void *)(dest - n);
|
205 |
}
|
206 |
}
|
207 |
#define memmove(a, b, c) pcre2_memmove(a, b, c)
|
208 |
#endif /* not HAVE_BCOPY */
|
209 |
#endif /* not HAVE_MEMMOVE */
|
210 |
#endif /* not VPCOMPAT */
|
211 |
|
212 |
/* External (in the C sense) functions and tables that are private to the
|
213 |
libraries are always referenced using the PRIV macro. This makes it possible
|
214 |
for pcre2test.c to include some of the source files from the libraries using a
|
215 |
different PRIV definition to avoid name clashes. It also makes it clear in the
|
216 |
code that a non-static object is being referenced. */
|
217 |
|
218 |
#ifndef PRIV
|
219 |
#define PRIV(name) _pcre2_##name
|
220 |
#endif
|
221 |
|
222 |
/* This is an unsigned int value that no UTF character can ever have, as
|
223 |
Unicode doesn't go beyond 0x0010ffff. */
|
224 |
|
225 |
#define NOTACHAR 0xffffffff
|
226 |
|
227 |
/* This is the largest valid UTF/Unicode code point. */
|
228 |
|
229 |
#define MAX_UTF_CODE_POINT 0x10ffff
|
230 |
|
231 |
/* Compile-time positive error numbers (all except UTF errors, which are
|
232 |
negative) start at this value. It should probably never be changed, in case
|
233 |
some application is checking for specific numbers. There is a copy of this
|
234 |
#define in pcre2posix.c (which now no longer includes this file). Ideally, a
|
235 |
way of having a single definition should be found, but as the number is
|
236 |
unlikely to change, this is not a pressing issue. The original reason for
|
237 |
having a base other than 0 was to keep the absolute values of compile-time and
|
238 |
run-time error numbers numerically different, but in the event the code does
|
239 |
not rely on this. */
|
240 |
|
241 |
#define COMPILE_ERROR_BASE 100
|
242 |
|
243 |
/* Define the default BSR convention. */
|
244 |
|
245 |
#ifdef BSR_ANYCRLF
|
246 |
#define BSR_DEFAULT PCRE2_BSR_ANYCRLF
|
247 |
#else
|
248 |
#define BSR_DEFAULT PCRE2_BSR_UNICODE
|
249 |
#endif
|
250 |
|
251 |
|
252 |
/* ---------------- Basic UTF-8 macros ---------------- */
|
253 |
|
254 |
/* These UTF-8 macros are always defined because they are used in pcre2test for
|
255 |
handling wide characters in 16-bit and 32-bit modes, even if an 8-bit library
|
256 |
is not supported. */
|
257 |
|
258 |
/* Tests whether a UTF-8 code point needs extra bytes to decode. */
|
259 |
|
260 |
#define HASUTF8EXTRALEN(c) ((c) >= 0xc0)
|
261 |
|
262 |
/* The following macros were originally written in the form of loops that used
|
263 |
data from the tables whose names start with PRIV(utf8_table). They were
|
264 |
rewritten by a user so as not to use loops, because in some environments this
|
265 |
gives a significant performance advantage, and it seems never to do any harm.
|
266 |
*/
|
267 |
|
268 |
/* Base macro to pick up the remaining bytes of a UTF-8 character, not
|
269 |
advancing the pointer. */
|
270 |
|
271 |
#define GETUTF8(c, eptr) \
|
272 |
{ \
|
273 |
if ((c & 0x20u) == 0) \
|
274 |
c = ((c & 0x1fu) << 6) | (eptr[1] & 0x3fu); \
|
275 |
else if ((c & 0x10u) == 0) \
|
276 |
c = ((c & 0x0fu) << 12) | ((eptr[1] & 0x3fu) << 6) | (eptr[2] & 0x3fu); \
|
277 |
else if ((c & 0x08u) == 0) \
|
278 |
c = ((c & 0x07u) << 18) | ((eptr[1] & 0x3fu) << 12) | \
|
279 |
((eptr[2] & 0x3fu) << 6) | (eptr[3] & 0x3fu); \
|
280 |
else if ((c & 0x04u) == 0) \
|
281 |
c = ((c & 0x03u) << 24) | ((eptr[1] & 0x3fu) << 18) | \
|
282 |
((eptr[2] & 0x3fu) << 12) | ((eptr[3] & 0x3fu) << 6) | \
|
283 |
(eptr[4] & 0x3fu); \
|
284 |
else \
|
285 |
c = ((c & 0x01u) << 30) | ((eptr[1] & 0x3fu) << 24) | \
|
286 |
((eptr[2] & 0x3fu) << 18) | ((eptr[3] & 0x3fu) << 12) | \
|
287 |
((eptr[4] & 0x3fu) << 6) | (eptr[5] & 0x3fu); \
|
288 |
}
|
289 |
|
290 |
/* Base macro to pick up the remaining bytes of a UTF-8 character, advancing
|
291 |
the pointer. */
|
292 |
|
293 |
#define GETUTF8INC(c, eptr) \
|
294 |
{ \
|
295 |
if ((c & 0x20u) == 0) \
|
296 |
c = ((c & 0x1fu) << 6) | (*eptr++ & 0x3fu); \
|
297 |
else if ((c & 0x10u) == 0) \
|
298 |
{ \
|
299 |
c = ((c & 0x0fu) << 12) | ((*eptr & 0x3fu) << 6) | (eptr[1] & 0x3fu); \
|
300 |
eptr += 2; \
|
301 |
} \
|
302 |
else if ((c & 0x08u) == 0) \
|
303 |
{ \
|
304 |
c = ((c & 0x07u) << 18) | ((*eptr & 0x3fu) << 12) | \
|
305 |
((eptr[1] & 0x3fu) << 6) | (eptr[2] & 0x3fu); \
|
306 |
eptr += 3; \
|
307 |
} \
|
308 |
else if ((c & 0x04u) == 0) \
|
309 |
{ \
|
310 |
c = ((c & 0x03u) << 24) | ((*eptr & 0x3fu) << 18) | \
|
311 |
((eptr[1] & 0x3fu) << 12) | ((eptr[2] & 0x3fu) << 6) | \
|
312 |
(eptr[3] & 0x3fu); \
|
313 |
eptr += 4; \
|
314 |
} \
|
315 |
else \
|
316 |
{ \
|
317 |
c = ((c & 0x01u) << 30) | ((*eptr & 0x3fu) << 24) | \
|
318 |
((eptr[1] & 0x3fu) << 18) | ((eptr[2] & 0x3fu) << 12) | \
|
319 |
((eptr[3] & 0x3fu) << 6) | (eptr[4] & 0x3fu); \
|
320 |
eptr += 5; \
|
321 |
} \
|
322 |
}
|
323 |
|
324 |
/* Base macro to pick up the remaining bytes of a UTF-8 character, not
|
325 |
advancing the pointer, incrementing the length. */
|
326 |
|
327 |
#define GETUTF8LEN(c, eptr, len) \
|
328 |
{ \
|
329 |
if ((c & 0x20u) == 0) \
|
330 |
{ \
|
331 |
c = ((c & 0x1fu) << 6) | (eptr[1] & 0x3fu); \
|
332 |
len++; \
|
333 |
} \
|
334 |
else if ((c & 0x10u) == 0) \
|
335 |
{ \
|
336 |
c = ((c & 0x0fu) << 12) | ((eptr[1] & 0x3fu) << 6) | (eptr[2] & 0x3fu); \
|
337 |
len += 2; \
|
338 |
} \
|
339 |
else if ((c & 0x08u) == 0) \
|
340 |
{\
|
341 |
c = ((c & 0x07u) << 18) | ((eptr[1] & 0x3fu) << 12) | \
|
342 |
((eptr[2] & 0x3fu) << 6) | (eptr[3] & 0x3fu); \
|
343 |
len += 3; \
|
344 |
} \
|
345 |
else if ((c & 0x04u) == 0) \
|
346 |
{ \
|
347 |
c = ((c & 0x03u) << 24) | ((eptr[1] & 0x3fu) << 18) | \
|
348 |
((eptr[2] & 0x3fu) << 12) | ((eptr[3] & 0x3fu) << 6) | \
|
349 |
(eptr[4] & 0x3fu); \
|
350 |
len += 4; \
|
351 |
} \
|
352 |
else \
|
353 |
{\
|
354 |
c = ((c & 0x01u) << 30) | ((eptr[1] & 0x3fu) << 24) | \
|
355 |
((eptr[2] & 0x3fu) << 18) | ((eptr[3] & 0x3fu) << 12) | \
|
356 |
((eptr[4] & 0x3fu) << 6) | (eptr[5] & 0x3fu); \
|
357 |
len += 5; \
|
358 |
} \
|
359 |
}
|
360 |
|
361 |
/* --------------- Whitespace macros ---------------- */
|
362 |
|
363 |
/* Tests for Unicode horizontal and vertical whitespace characters must check a
|
364 |
number of different values. Using a switch statement for this generates the
|
365 |
fastest code (no loop, no memory access), and there are several places in the
|
366 |
interpreter code where this happens. In order to ensure that all the case lists
|
367 |
remain in step, we use macros so that there is only one place where the lists
|
368 |
are defined.
|
369 |
|
370 |
These values are also required as lists in pcre2_compile.c when processing \h,
|
371 |
\H, \v and \V in a character class. The lists are defined in pcre2_tables.c,
|
372 |
but macros that define the values are here so that all the definitions are
|
373 |
together. The lists must be in ascending character order, terminated by
|
374 |
NOTACHAR (which is 0xffffffff).
|
375 |
|
376 |
Any changes should ensure that the various macros are kept in step with each
|
377 |
other. NOTE: The values also appear in pcre2_jit_compile.c. */
|
378 |
|
379 |
/* -------------- ASCII/Unicode environments -------------- */
|
380 |
|
381 |
#ifndef EBCDIC
|
382 |
|
383 |
/* Character U+180E (Mongolian Vowel Separator) is not included in the list of
|
384 |
spaces in the Unicode file PropList.txt, and Perl does not recognize it as a
|
385 |
space. However, in many other sources it is listed as a space and has been in
|
386 |
PCRE (both APIs) for a long time. */
|
387 |
|
388 |
#define HSPACE_LIST \
|
389 |
CHAR_HT, CHAR_SPACE, CHAR_NBSP, \
|
390 |
0x1680, 0x180e, 0x2000, 0x2001, 0x2002, 0x2003, 0x2004, 0x2005, \
|
391 |
0x2006, 0x2007, 0x2008, 0x2009, 0x200A, 0x202f, 0x205f, 0x3000, \
|
392 |
NOTACHAR
|
393 |
|
394 |
#define HSPACE_MULTIBYTE_CASES \
|
395 |
case 0x1680: /* OGHAM SPACE MARK */ \
|
396 |
case 0x180e: /* MONGOLIAN VOWEL SEPARATOR */ \
|
397 |
case 0x2000: /* EN QUAD */ \
|
398 |
case 0x2001: /* EM QUAD */ \
|
399 |
case 0x2002: /* EN SPACE */ \
|
400 |
case 0x2003: /* EM SPACE */ \
|
401 |
case 0x2004: /* THREE-PER-EM SPACE */ \
|
402 |
case 0x2005: /* FOUR-PER-EM SPACE */ \
|
403 |
case 0x2006: /* SIX-PER-EM SPACE */ \
|
404 |
case 0x2007: /* FIGURE SPACE */ \
|
405 |
case 0x2008: /* PUNCTUATION SPACE */ \
|
406 |
case 0x2009: /* THIN SPACE */ \
|
407 |
case 0x200A: /* HAIR SPACE */ \
|
408 |
case 0x202f: /* NARROW NO-BREAK SPACE */ \
|
409 |
case 0x205f: /* MEDIUM MATHEMATICAL SPACE */ \
|
410 |
case 0x3000 /* IDEOGRAPHIC SPACE */
|
411 |
|
412 |
#define HSPACE_BYTE_CASES \
|
413 |
case CHAR_HT: \
|
414 |
case CHAR_SPACE: \
|
415 |
case CHAR_NBSP
|
416 |
|
417 |
#define HSPACE_CASES \
|
418 |
HSPACE_BYTE_CASES: \
|
419 |
HSPACE_MULTIBYTE_CASES
|
420 |
|
421 |
#define VSPACE_LIST \
|
422 |
CHAR_LF, CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, 0x2028, 0x2029, NOTACHAR
|
423 |
|
424 |
#define VSPACE_MULTIBYTE_CASES \
|
425 |
case 0x2028: /* LINE SEPARATOR */ \
|
426 |
case 0x2029 /* PARAGRAPH SEPARATOR */
|
427 |
|
428 |
#define VSPACE_BYTE_CASES \
|
429 |
case CHAR_LF: \
|
430 |
case CHAR_VT: \
|
431 |
case CHAR_FF: \
|
432 |
case CHAR_CR: \
|
433 |
case CHAR_NEL
|
434 |
|
435 |
#define VSPACE_CASES \
|
436 |
VSPACE_BYTE_CASES: \
|
437 |
VSPACE_MULTIBYTE_CASES
|
438 |
|
439 |
/* -------------- EBCDIC environments -------------- */
|
440 |
|
441 |
#else
|
442 |
#define HSPACE_LIST CHAR_HT, CHAR_SPACE, CHAR_NBSP, NOTACHAR
|
443 |
|
444 |
#define HSPACE_BYTE_CASES \
|
445 |
case CHAR_HT: \
|
446 |
case CHAR_SPACE: \
|
447 |
case CHAR_NBSP
|
448 |
|
449 |
#define HSPACE_CASES HSPACE_BYTE_CASES
|
450 |
|
451 |
#ifdef EBCDIC_NL25
|
452 |
#define VSPACE_LIST \
|
453 |
CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, CHAR_LF, NOTACHAR
|
454 |
#else
|
455 |
#define VSPACE_LIST \
|
456 |
CHAR_VT, CHAR_FF, CHAR_CR, CHAR_LF, CHAR_NEL, NOTACHAR
|
457 |
#endif
|
458 |
|
459 |
#define VSPACE_BYTE_CASES \
|
460 |
case CHAR_LF: \
|
461 |
case CHAR_VT: \
|
462 |
case CHAR_FF: \
|
463 |
case CHAR_CR: \
|
464 |
case CHAR_NEL
|
465 |
|
466 |
#define VSPACE_CASES VSPACE_BYTE_CASES
|
467 |
#endif /* EBCDIC */
|
468 |
|
469 |
/* -------------- End of whitespace macros -------------- */
|
470 |
|
471 |
|
472 |
/* PCRE2 is able to support several different kinds of newline (CR, LF, CRLF,
|
473 |
"any" and "anycrlf" at present). The following macros are used to package up
|
474 |
testing for newlines. NLBLOCK, PSSTART, and PSEND are defined in the various
|
475 |
modules to indicate in which datablock the parameters exist, and what the
|
476 |
start/end of string field names are. */
|
477 |
|
478 |
#define NLTYPE_FIXED 0 /* Newline is a fixed length string */
|
479 |
#define NLTYPE_ANY 1 /* Newline is any Unicode line ending */
|
480 |
#define NLTYPE_ANYCRLF 2 /* Newline is CR, LF, or CRLF */
|
481 |
|
482 |
/* This macro checks for a newline at the given position */
|
483 |
|
484 |
#define IS_NEWLINE(p) \
|
485 |
((NLBLOCK->nltype != NLTYPE_FIXED)? \
|
486 |
((p) < NLBLOCK->PSEND && \
|
487 |
PRIV(is_newline)((p), NLBLOCK->nltype, NLBLOCK->PSEND, \
|
488 |
&(NLBLOCK->nllen), utf)) \
|
489 |
: \
|
490 |
((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \
|
491 |
UCHAR21TEST(p) == NLBLOCK->nl[0] && \
|
492 |
(NLBLOCK->nllen == 1 || UCHAR21TEST(p+1) == NLBLOCK->nl[1]) \
|
493 |
) \
|
494 |
)
|
495 |
|
496 |
/* This macro checks for a newline immediately preceding the given position */
|
497 |
|
498 |
#define WAS_NEWLINE(p) \
|
499 |
((NLBLOCK->nltype != NLTYPE_FIXED)? \
|
500 |
((p) > NLBLOCK->PSSTART && \
|
501 |
PRIV(was_newline)((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \
|
502 |
&(NLBLOCK->nllen), utf)) \
|
503 |
: \
|
504 |
((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \
|
505 |
UCHAR21TEST(p - NLBLOCK->nllen) == NLBLOCK->nl[0] && \
|
506 |
(NLBLOCK->nllen == 1 || UCHAR21TEST(p - NLBLOCK->nllen + 1) == NLBLOCK->nl[1]) \
|
507 |
) \
|
508 |
)
|
509 |
|
510 |
/* Private flags containing information about the compiled pattern. The first
|
511 |
three must not be changed, because whichever is set is actually the number of
|
512 |
bytes in a code unit in that mode. */
|
513 |
|
514 |
#define PCRE2_MODE8 0x00000001 /* compiled in 8 bit mode */
|
515 |
#define PCRE2_MODE16 0x00000002 /* compiled in 16 bit mode */
|
516 |
#define PCRE2_MODE32 0x00000004 /* compiled in 32 bit mode */
|
517 |
#define PCRE2_FIRSTSET 0x00000010 /* first_code unit is set */
|
518 |
#define PCRE2_FIRSTCASELESS 0x00000020 /* caseless first code unit */
|
519 |
#define PCRE2_FIRSTMAPSET 0x00000040 /* bitmap of first code units is set */
|
520 |
#define PCRE2_LASTSET 0x00000080 /* last code unit is set */
|
521 |
#define PCRE2_LASTCASELESS 0x00000100 /* caseless last code unit */
|
522 |
#define PCRE2_STARTLINE 0x00000200 /* start after \n for multiline */
|
523 |
#define PCRE2_JCHANGED 0x00000400 /* j option used in pattern */
|
524 |
#define PCRE2_HASCRORLF 0x00000800 /* explicit \r or \n in pattern */
|
525 |
#define PCRE2_HASTHEN 0x00001000 /* pattern contains (*THEN) */
|
526 |
#define PCRE2_MATCH_EMPTY 0x00002000 /* pattern can match empty string */
|
527 |
#define PCRE2_BSR_SET 0x00004000 /* BSR was set in the pattern */
|
528 |
#define PCRE2_NL_SET 0x00008000 /* newline was set in the pattern */
|
529 |
#define PCRE2_NOTEMPTY_SET 0x00010000 /* (*NOTEMPTY) used ) keep */
|
530 |
#define PCRE2_NE_ATST_SET 0x00020000 /* (*NOTEMPTY_ATSTART) used) together */
|
531 |
#define PCRE2_DEREF_TABLES 0x00040000 /* release character tables */
|
532 |
#define PCRE2_NOJIT 0x00080000 /* (*NOJIT) used */
|
533 |
#define PCRE2_HASBKPORX 0x00100000 /* contains \P, \p, or \X */
|
534 |
#define PCRE2_DUPCAPUSED 0x00200000 /* contains (?| */
|
535 |
#define PCRE2_HASBKC 0x00400000 /* contains \C */
|
536 |
|
537 |
#define PCRE2_MODE_MASK (PCRE2_MODE8 | PCRE2_MODE16 | PCRE2_MODE32)
|
538 |
|
539 |
/* Values for the matchedby field in a match data block. */
|
540 |
|
541 |
enum { PCRE2_MATCHEDBY_INTERPRETER, /* pcre2_match() */
|
542 |
PCRE2_MATCHEDBY_DFA_INTERPRETER, /* pcre2_dfa_match() */
|
543 |
PCRE2_MATCHEDBY_JIT }; /* pcre2_jit_match() */
|
544 |
|
545 |
/* Magic number to provide a small check against being handed junk. */
|
546 |
|
547 |
#define MAGIC_NUMBER 0x50435245UL /* 'PCRE' */
|
548 |
|
549 |
/* The maximum remaining length of subject we are prepared to search for a
|
550 |
req_unit match. */
|
551 |
|
552 |
#define REQ_CU_MAX 1000
|
553 |
|
554 |
/* Offsets for the bitmap tables in the cbits set of tables. Each table
|
555 |
contains a set of bits for a class map. Some classes are built by combining
|
556 |
these tables. */
|
557 |
|
558 |
#define cbit_space 0 /* [:space:] or \s */
|
559 |
#define cbit_xdigit 32 /* [:xdigit:] */
|
560 |
#define cbit_digit 64 /* [:digit:] or \d */
|
561 |
#define cbit_upper 96 /* [:upper:] */
|
562 |
#define cbit_lower 128 /* [:lower:] */
|
563 |
#define cbit_word 160 /* [:word:] or \w */
|
564 |
#define cbit_graph 192 /* [:graph:] */
|
565 |
#define cbit_print 224 /* [:print:] */
|
566 |
#define cbit_punct 256 /* [:punct:] */
|
567 |
#define cbit_cntrl 288 /* [:cntrl:] */
|
568 |
#define cbit_length 320 /* Length of the cbits table */
|
569 |
|
570 |
/* Bit definitions for entries in the ctypes table. */
|
571 |
|
572 |
#define ctype_space 0x01
|
573 |
#define ctype_letter 0x02
|
574 |
#define ctype_digit 0x04
|
575 |
#define ctype_xdigit 0x08
|
576 |
#define ctype_word 0x10 /* alphanumeric or '_' */
|
577 |
#define ctype_meta 0x80 /* regexp meta char or zero (end pattern) */
|
578 |
|
579 |
/* Offsets of the various tables from the base tables pointer, and
|
580 |
total length of the tables. */
|
581 |
|
582 |
#define lcc_offset 0 /* Lower case */
|
583 |
#define fcc_offset 256 /* Flip case */
|
584 |
#define cbits_offset 512 /* Character classes */
|
585 |
#define ctypes_offset (cbits_offset + cbit_length) /* Character types */
|
586 |
#define tables_length (ctypes_offset + 256)
|
587 |
|
588 |
|
589 |
/* -------------------- Character and string names ------------------------ */
|
590 |
|
591 |
/* If PCRE2 is to support UTF-8 on EBCDIC platforms, we cannot use normal
|
592 |
character constants like '*' because the compiler would emit their EBCDIC code,
|
593 |
which is different from their ASCII/UTF-8 code. Instead we define macros for
|
594 |
the characters so that they always use the ASCII/UTF-8 code when UTF-8 support
|
595 |
is enabled. When UTF-8 support is not enabled, the definitions use character
|
596 |
literals. Both character and string versions of each character are needed, and
|
597 |
there are some longer strings as well.
|
598 |
|
599 |
This means that, on EBCDIC platforms, the PCRE2 library can handle either
|
600 |
EBCDIC, or UTF-8, but not both. To support both in the same compiled library
|
601 |
would need different lookups depending on whether PCRE2_UTF was set or not.
|
602 |
This would make it impossible to use characters in switch/case statements,
|
603 |
which would reduce performance. For a theoretical use (which nobody has asked
|
604 |
for) in a minority area (EBCDIC platforms), this is not sensible. Any
|
605 |
application that did need both could compile two versions of the library, using
|
606 |
macros to give the functions distinct names. */
|
607 |
|
608 |
#ifndef SUPPORT_UNICODE
|
609 |
|
610 |
/* UTF-8 support is not enabled; use the platform-dependent character literals
|
611 |
so that PCRE2 works in both ASCII and EBCDIC environments, but only in non-UTF
|
612 |
mode. Newline characters are problematic in EBCDIC. Though it has CR and LF
|
613 |
characters, a common practice has been to use its NL (0x15) character as the
|
614 |
line terminator in C-like processing environments. However, sometimes the LF
|
615 |
(0x25) character is used instead, according to this Unicode document:
|
616 |
|
617 |
http://unicode.org/standard/reports/tr13/tr13-5.html
|
618 |
|
619 |
PCRE2 defaults EBCDIC NL to 0x15, but has a build-time option to select 0x25
|
620 |
instead. Whichever is *not* chosen is defined as NEL.
|
621 |
|
622 |
In both ASCII and EBCDIC environments, CHAR_NL and CHAR_LF are synonyms for the
|
623 |
same code point. */
|
624 |
|
625 |
#ifdef EBCDIC
|
626 |
|
627 |
#ifndef EBCDIC_NL25
|
628 |
#define CHAR_NL '\x15'
|
629 |
#define CHAR_NEL '\x25'
|
630 |
#define STR_NL "\x15"
|
631 |
#define STR_NEL "\x25"
|
632 |
#else
|
633 |
#define CHAR_NL '\x25'
|
634 |
#define CHAR_NEL '\x15'
|
635 |
#define STR_NL "\x25"
|
636 |
#define STR_NEL "\x15"
|
637 |
#endif
|
638 |
|
639 |
#define CHAR_LF CHAR_NL
|
640 |
#define STR_LF STR_NL
|
641 |
|
642 |
#define CHAR_ESC '\047'
|
643 |
#define CHAR_DEL '\007'
|
644 |
#define CHAR_NBSP ((unsigned char)'\x41')
|
645 |
#define STR_ESC "\047"
|
646 |
#define STR_DEL "\007"
|
647 |
|
648 |
#else /* Not EBCDIC */
|
649 |
|
650 |
/* In ASCII/Unicode, linefeed is '\n' and we equate this to NL for
|
651 |
compatibility. NEL is the Unicode newline character; make sure it is
|
652 |
a positive value. */
|
653 |
|
654 |
#define CHAR_LF '\n'
|
655 |
#define CHAR_NL CHAR_LF
|
656 |
#define CHAR_NEL ((unsigned char)'\x85')
|
657 |
#define CHAR_ESC '\033'
|
658 |
#define CHAR_DEL '\177'
|
659 |
#define CHAR_NBSP ((unsigned char)'\xa0')
|
660 |
|
661 |
#define STR_LF "\n"
|
662 |
#define STR_NL STR_LF
|
663 |
#define STR_NEL "\x85"
|
664 |
#define STR_ESC "\033"
|
665 |
#define STR_DEL "\177"
|
666 |
|
667 |
#endif /* EBCDIC */
|
668 |
|
669 |
/* The remaining definitions work in both environments. */
|
670 |
|
671 |
#define CHAR_NULL '\0'
|
672 |
#define CHAR_HT '\t'
|
673 |
#define CHAR_VT '\v'
|
674 |
#define CHAR_FF '\f'
|
675 |
#define CHAR_CR '\r'
|
676 |
#define CHAR_BS '\b'
|
677 |
#define CHAR_BEL '\a'
|
678 |
|
679 |
#define CHAR_SPACE ' '
|
680 |
#define CHAR_EXCLAMATION_MARK '!'
|
681 |
#define CHAR_QUOTATION_MARK '"'
|
682 |
#define CHAR_NUMBER_SIGN '#'
|
683 |
#define CHAR_DOLLAR_SIGN '$'
|
684 |
#define CHAR_PERCENT_SIGN '%'
|
685 |
#define CHAR_AMPERSAND '&'
|
686 |
#define CHAR_APOSTROPHE '\''
|
687 |
#define CHAR_LEFT_PARENTHESIS '('
|
688 |
#define CHAR_RIGHT_PARENTHESIS ')'
|
689 |
#define CHAR_ASTERISK '*'
|
690 |
#define CHAR_PLUS '+'
|
691 |
#define CHAR_COMMA ','
|
692 |
#define CHAR_MINUS '-'
|
693 |
#define CHAR_DOT '.'
|
694 |
#define CHAR_SLASH '/'
|
695 |
#define CHAR_0 '0'
|
696 |
#define CHAR_1 '1'
|
697 |
#define CHAR_2 '2'
|
698 |
#define CHAR_3 '3'
|
699 |
#define CHAR_4 '4'
|
700 |
#define CHAR_5 '5'
|
701 |
#define CHAR_6 '6'
|
702 |
#define CHAR_7 '7'
|
703 |
#define CHAR_8 '8'
|
704 |
#define CHAR_9 '9'
|
705 |
#define CHAR_COLON ':'
|
706 |
#define CHAR_SEMICOLON ';'
|
707 |
#define CHAR_LESS_THAN_SIGN '<'
|
708 |
#define CHAR_EQUALS_SIGN '='
|
709 |
#define CHAR_GREATER_THAN_SIGN '>'
|
710 |
#define CHAR_QUESTION_MARK '?'
|
711 |
#define CHAR_COMMERCIAL_AT '@'
|
712 |
#define CHAR_A 'A'
|
713 |
#define CHAR_B 'B'
|
714 |
#define CHAR_C 'C'
|
715 |
#define CHAR_D 'D'
|
716 |
#define CHAR_E 'E'
|
717 |
#define CHAR_F 'F'
|
718 |
#define CHAR_G 'G'
|
719 |
#define CHAR_H 'H'
|
720 |
#define CHAR_I 'I'
|
721 |
#define CHAR_J 'J'
|
722 |
#define CHAR_K 'K'
|
723 |
#define CHAR_L 'L'
|
724 |
#define CHAR_M 'M'
|
725 |
#define CHAR_N 'N'
|
726 |
#define CHAR_O 'O'
|
727 |
#define CHAR_P 'P'
|
728 |
#define CHAR_Q 'Q'
|
729 |
#define CHAR_R 'R'
|
730 |
#define CHAR_S 'S'
|
731 |
#define CHAR_T 'T'
|
732 |
#define CHAR_U 'U'
|
733 |
#define CHAR_V 'V'
|
734 |
#define CHAR_W 'W'
|
735 |
#define CHAR_X 'X'
|
736 |
#define CHAR_Y 'Y'
|
737 |
#define CHAR_Z 'Z'
|
738 |
#define CHAR_LEFT_SQUARE_BRACKET '['
|
739 |
#define CHAR_BACKSLASH '\\'
|
740 |
#define CHAR_RIGHT_SQUARE_BRACKET ']'
|
741 |
#define CHAR_CIRCUMFLEX_ACCENT '^'
|
742 |
#define CHAR_UNDERSCORE '_'
|
743 |
#define CHAR_GRAVE_ACCENT '`'
|
744 |
#define CHAR_a 'a'
|
745 |
#define CHAR_b 'b'
|
746 |
#define CHAR_c 'c'
|
747 |
#define CHAR_d 'd'
|
748 |
#define CHAR_e 'e'
|
749 |
#define CHAR_f 'f'
|
750 |
#define CHAR_g 'g'
|
751 |
#define CHAR_h 'h'
|
752 |
#define CHAR_i 'i'
|
753 |
#define CHAR_j 'j'
|
754 |
#define CHAR_k 'k'
|
755 |
#define CHAR_l 'l'
|
756 |
#define CHAR_m 'm'
|
757 |
#define CHAR_n 'n'
|
758 |
#define CHAR_o 'o'
|
759 |
#define CHAR_p 'p'
|
760 |
#define CHAR_q 'q'
|
761 |
#define CHAR_r 'r'
|
762 |
#define CHAR_s 's'
|
763 |
#define CHAR_t 't'
|
764 |
#define CHAR_u 'u'
|
765 |
#define CHAR_v 'v'
|
766 |
#define CHAR_w 'w'
|
767 |
#define CHAR_x 'x'
|
768 |
#define CHAR_y 'y'
|
769 |
#define CHAR_z 'z'
|
770 |
#define CHAR_LEFT_CURLY_BRACKET '{'
|
771 |
#define CHAR_VERTICAL_LINE '|'
|
772 |
#define CHAR_RIGHT_CURLY_BRACKET '}'
|
773 |
#define CHAR_TILDE '~'
|
774 |
|
775 |
#define STR_HT "\t"
|
776 |
#define STR_VT "\v"
|
777 |
#define STR_FF "\f"
|
778 |
#define STR_CR "\r"
|
779 |
#define STR_BS "\b"
|
780 |
#define STR_BEL "\a"
|
781 |
|
782 |
#define STR_SPACE " "
|
783 |
#define STR_EXCLAMATION_MARK "!"
|
784 |
#define STR_QUOTATION_MARK "\""
|
785 |
#define STR_NUMBER_SIGN "#"
|
786 |
#define STR_DOLLAR_SIGN "$"
|
787 |
#define STR_PERCENT_SIGN "%"
|
788 |
#define STR_AMPERSAND "&"
|
789 |
#define STR_APOSTROPHE "'"
|
790 |
#define STR_LEFT_PARENTHESIS "("
|
791 |
#define STR_RIGHT_PARENTHESIS ")"
|
792 |
#define STR_ASTERISK "*"
|
793 |
#define STR_PLUS "+"
|
794 |
#define STR_COMMA ","
|
795 |
#define STR_MINUS "-"
|
796 |
#define STR_DOT "."
|
797 |
#define STR_SLASH "/"
|
798 |
#define STR_0 "0"
|
799 |
#define STR_1 "1"
|
800 |
#define STR_2 "2"
|
801 |
#define STR_3 "3"
|
802 |
#define STR_4 "4"
|
803 |
#define STR_5 "5"
|
804 |
#define STR_6 "6"
|
805 |
#define STR_7 "7"
|
806 |
#define STR_8 "8"
|
807 |
#define STR_9 "9"
|
808 |
#define STR_COLON ":"
|
809 |
#define STR_SEMICOLON ";"
|
810 |
#define STR_LESS_THAN_SIGN "<"
|
811 |
#define STR_EQUALS_SIGN "="
|
812 |
#define STR_GREATER_THAN_SIGN ">"
|
813 |
#define STR_QUESTION_MARK "?"
|
814 |
#define STR_COMMERCIAL_AT "@"
|
815 |
#define STR_A "A"
|
816 |
#define STR_B "B"
|
817 |
#define STR_C "C"
|
818 |
#define STR_D "D"
|
819 |
#define STR_E "E"
|
820 |
#define STR_F "F"
|
821 |
#define STR_G "G"
|
822 |
#define STR_H "H"
|
823 |
#define STR_I "I"
|
824 |
#define STR_J "J"
|
825 |
#define STR_K "K"
|
826 |
#define STR_L "L"
|
827 |
#define STR_M "M"
|
828 |
#define STR_N "N"
|
829 |
#define STR_O "O"
|
830 |
#define STR_P "P"
|
831 |
#define STR_Q "Q"
|
832 |
#define STR_R "R"
|
833 |
#define STR_S "S"
|
834 |
#define STR_T "T"
|
835 |
#define STR_U "U"
|
836 |
#define STR_V "V"
|
837 |
#define STR_W "W"
|
838 |
#define STR_X "X"
|
839 |
#define STR_Y "Y"
|
840 |
#define STR_Z "Z"
|
841 |
#define STR_LEFT_SQUARE_BRACKET "["
|
842 |
#define STR_BACKSLASH "\\"
|
843 |
#define STR_RIGHT_SQUARE_BRACKET "]"
|
844 |
#define STR_CIRCUMFLEX_ACCENT "^"
|
845 |
#define STR_UNDERSCORE "_"
|
846 |
#define STR_GRAVE_ACCENT "`"
|
847 |
#define STR_a "a"
|
848 |
#define STR_b "b"
|
849 |
#define STR_c "c"
|
850 |
#define STR_d "d"
|
851 |
#define STR_e "e"
|
852 |
#define STR_f "f"
|
853 |
#define STR_g "g"
|
854 |
#define STR_h "h"
|
855 |
#define STR_i "i"
|
856 |
#define STR_j "j"
|
857 |
#define STR_k "k"
|
858 |
#define STR_l "l"
|
859 |
#define STR_m "m"
|
860 |
#define STR_n "n"
|
861 |
#define STR_o "o"
|
862 |
#define STR_p "p"
|
863 |
#define STR_q "q"
|
864 |
#define STR_r "r"
|
865 |
#define STR_s "s"
|
866 |
#define STR_t "t"
|
867 |
#define STR_u "u"
|
868 |
#define STR_v "v"
|
869 |
#define STR_w "w"
|
870 |
#define STR_x "x"
|
871 |
#define STR_y "y"
|
872 |
#define STR_z "z"
|
873 |
#define STR_LEFT_CURLY_BRACKET "{"
|
874 |
#define STR_VERTICAL_LINE "|"
|
875 |
#define STR_RIGHT_CURLY_BRACKET "}"
|
876 |
#define STR_TILDE "~"
|
877 |
|
878 |
#define STRING_ACCEPT0 "ACCEPT\0"
|
879 |
#define STRING_COMMIT0 "COMMIT\0"
|
880 |
#define STRING_F0 "F\0"
|
881 |
#define STRING_FAIL0 "FAIL\0"
|
882 |
#define STRING_MARK0 "MARK\0"
|
883 |
#define STRING_PRUNE0 "PRUNE\0"
|
884 |
#define STRING_SKIP0 "SKIP\0"
|
885 |
#define STRING_THEN "THEN"
|
886 |
|
887 |
#define STRING_alpha0 "alpha\0"
|
888 |
#define STRING_lower0 "lower\0"
|
889 |
#define STRING_upper0 "upper\0"
|
890 |
#define STRING_alnum0 "alnum\0"
|
891 |
#define STRING_ascii0 "ascii\0"
|
892 |
#define STRING_blank0 "blank\0"
|
893 |
#define STRING_cntrl0 "cntrl\0"
|
894 |
#define STRING_digit0 "digit\0"
|
895 |
#define STRING_graph0 "graph\0"
|
896 |
#define STRING_print0 "print\0"
|
897 |
#define STRING_punct0 "punct\0"
|
898 |
#define STRING_space0 "space\0"
|
899 |
#define STRING_word0 "word\0"
|
900 |
#define STRING_xdigit "xdigit"
|
901 |
|
902 |
#define STRING_DEFINE "DEFINE"
|
903 |
#define STRING_VERSION "VERSION"
|
904 |
#define STRING_WEIRD_STARTWORD "[:<:]]"
|
905 |
#define STRING_WEIRD_ENDWORD "[:>:]]"
|
906 |
|
907 |
#define STRING_CR_RIGHTPAR "CR)"
|
908 |
#define STRING_LF_RIGHTPAR "LF)"
|
909 |
#define STRING_CRLF_RIGHTPAR "CRLF)"
|
910 |
#define STRING_ANY_RIGHTPAR "ANY)"
|
911 |
#define STRING_ANYCRLF_RIGHTPAR "ANYCRLF)"
|
912 |
#define STRING_BSR_ANYCRLF_RIGHTPAR "BSR_ANYCRLF)"
|
913 |
#define STRING_BSR_UNICODE_RIGHTPAR "BSR_UNICODE)"
|
914 |
#define STRING_UTF8_RIGHTPAR "UTF8)"
|
915 |
#define STRING_UTF16_RIGHTPAR "UTF16)"
|
916 |
#define STRING_UTF32_RIGHTPAR "UTF32)"
|
917 |
#define STRING_UTF_RIGHTPAR "UTF)"
|
918 |
#define STRING_UCP_RIGHTPAR "UCP)"
|
919 |
#define STRING_NO_AUTO_POSSESS_RIGHTPAR "NO_AUTO_POSSESS)"
|
920 |
#define STRING_NO_DOTSTAR_ANCHOR_RIGHTPAR "NO_DOTSTAR_ANCHOR)"
|
921 |
#define STRING_NO_JIT_RIGHTPAR "NO_JIT)"
|
922 |
#define STRING_NO_START_OPT_RIGHTPAR "NO_START_OPT)"
|
923 |
#define STRING_NOTEMPTY_RIGHTPAR "NOTEMPTY)"
|
924 |
#define STRING_NOTEMPTY_ATSTART_RIGHTPAR "NOTEMPTY_ATSTART)"
|
925 |
#define STRING_LIMIT_MATCH_EQ "LIMIT_MATCH="
|
926 |
#define STRING_LIMIT_RECURSION_EQ "LIMIT_RECURSION="
|
927 |
#define STRING_MARK "MARK"
|
928 |
|
929 |
#else /* SUPPORT_UNICODE */
|
930 |
|
931 |
/* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This
|
932 |
works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode
|
933 |
only. */
|
934 |
|
935 |
#define CHAR_HT '\011'
|
936 |
#define CHAR_VT '\013'
|
937 |
#define CHAR_FF '\014'
|
938 |
#define CHAR_CR '\015'
|
939 |
#define CHAR_LF '\012'
|
940 |
#define CHAR_NL CHAR_LF
|
941 |
#define CHAR_NEL ((unsigned char)'\x85')
|
942 |
#define CHAR_BS '\010'
|
943 |
#define CHAR_BEL '\007'
|
944 |
#define CHAR_ESC '\033'
|
945 |
#define CHAR_DEL '\177'
|
946 |
|
947 |
#define CHAR_NULL '\0'
|
948 |
#define CHAR_SPACE '\040'
|
949 |
#define CHAR_EXCLAMATION_MARK '\041'
|
950 |
#define CHAR_QUOTATION_MARK '\042'
|
951 |
#define CHAR_NUMBER_SIGN '\043'
|
952 |
#define CHAR_DOLLAR_SIGN '\044'
|
953 |
#define CHAR_PERCENT_SIGN '\045'
|
954 |
#define CHAR_AMPERSAND '\046'
|
955 |
#define CHAR_APOSTROPHE '\047'
|
956 |
#define CHAR_LEFT_PARENTHESIS '\050'
|
957 |
#define CHAR_RIGHT_PARENTHESIS '\051'
|
958 |
#define CHAR_ASTERISK '\052'
|
959 |
#define CHAR_PLUS '\053'
|
960 |
#define CHAR_COMMA '\054'
|
961 |
#define CHAR_MINUS '\055'
|
962 |
#define CHAR_DOT '\056'
|
963 |
#define CHAR_SLASH '\057'
|
964 |
#define CHAR_0 '\060'
|
965 |
#define CHAR_1 '\061'
|
966 |
#define CHAR_2 '\062'
|
967 |
#define CHAR_3 '\063'
|
968 |
#define CHAR_4 '\064'
|
969 |
#define CHAR_5 '\065'
|
970 |
#define CHAR_6 '\066'
|
971 |
#define CHAR_7 '\067'
|
972 |
#define CHAR_8 '\070'
|
973 |
#define CHAR_9 '\071'
|
974 |
#define CHAR_COLON '\072'
|
975 |
#define CHAR_SEMICOLON '\073'
|
976 |
#define CHAR_LESS_THAN_SIGN '\074'
|
977 |
#define CHAR_EQUALS_SIGN '\075'
|
978 |
#define CHAR_GREATER_THAN_SIGN '\076'
|
979 |
#define CHAR_QUESTION_MARK '\077'
|
980 |
#define CHAR_COMMERCIAL_AT '\100'
|
981 |
#define CHAR_A '\101'
|
982 |
#define CHAR_B '\102'
|
983 |
#define CHAR_C '\103'
|
984 |
#define CHAR_D '\104'
|
985 |
#define CHAR_E '\105'
|
986 |
#define CHAR_F '\106'
|
987 |
#define CHAR_G '\107'
|
988 |
#define CHAR_H '\110'
|
989 |
#define CHAR_I '\111'
|
990 |
#define CHAR_J '\112'
|
991 |
#define CHAR_K '\113'
|
992 |
#define CHAR_L '\114'
|
993 |
#define CHAR_M '\115'
|
994 |
#define CHAR_N '\116'
|
995 |
#define CHAR_O '\117'
|
996 |
#define CHAR_P '\120'
|
997 |
#define CHAR_Q '\121'
|
998 |
#define CHAR_R '\122'
|
999 |
#define CHAR_S '\123'
|
1000 |
#define CHAR_T '\124'
|
1001 |
#define CHAR_U '\125'
|
1002 |
#define CHAR_V '\126'
|
1003 |
#define CHAR_W '\127'
|
1004 |
#define CHAR_X '\130'
|
1005 |
#define CHAR_Y '\131'
|
1006 |
#define CHAR_Z '\132'
|
1007 |
#define CHAR_LEFT_SQUARE_BRACKET '\133'
|
1008 |
#define CHAR_BACKSLASH '\134'
|
1009 |
#define CHAR_RIGHT_SQUARE_BRACKET '\135'
|
1010 |
#define CHAR_CIRCUMFLEX_ACCENT '\136'
|
1011 |
#define CHAR_UNDERSCORE '\137'
|
1012 |
#define CHAR_GRAVE_ACCENT '\140'
|
1013 |
#define CHAR_a '\141'
|
1014 |
#define CHAR_b '\142'
|
1015 |
#define CHAR_c '\143'
|
1016 |
#define CHAR_d '\144'
|
1017 |
#define CHAR_e '\145'
|
1018 |
#define CHAR_f '\146'
|
1019 |
#define CHAR_g '\147'
|
1020 |
#define CHAR_h '\150'
|
1021 |
#define CHAR_i '\151'
|
1022 |
#define CHAR_j '\152'
|
1023 |
#define CHAR_k '\153'
|
1024 |
#define CHAR_l '\154'
|
1025 |
#define CHAR_m '\155'
|
1026 |
#define CHAR_n '\156'
|
1027 |
#define CHAR_o '\157'
|
1028 |
#define CHAR_p '\160'
|
1029 |
#define CHAR_q '\161'
|
1030 |
#define CHAR_r '\162'
|
1031 |
#define CHAR_s '\163'
|
1032 |
#define CHAR_t '\164'
|
1033 |
#define CHAR_u '\165'
|
1034 |
#define CHAR_v '\166'
|
1035 |
#define CHAR_w '\167'
|
1036 |
#define CHAR_x '\170'
|
1037 |
#define CHAR_y '\171'
|
1038 |
#define CHAR_z '\172'
|
1039 |
#define CHAR_LEFT_CURLY_BRACKET '\173'
|
1040 |
#define CHAR_VERTICAL_LINE '\174'
|
1041 |
#define CHAR_RIGHT_CURLY_BRACKET '\175'
|
1042 |
#define CHAR_TILDE '\176'
|
1043 |
#define CHAR_NBSP ((unsigned char)'\xa0')
|
1044 |
|
1045 |
#define STR_HT "\011"
|
1046 |
#define STR_VT "\013"
|
1047 |
#define STR_FF "\014"
|
1048 |
#define STR_CR "\015"
|
1049 |
#define STR_NL "\012"
|
1050 |
#define STR_BS "\010"
|
1051 |
#define STR_BEL "\007"
|
1052 |
#define STR_ESC "\033"
|
1053 |
#define STR_DEL "\177"
|
1054 |
|
1055 |
#define STR_SPACE "\040"
|
1056 |
#define STR_EXCLAMATION_MARK "\041"
|
1057 |
#define STR_QUOTATION_MARK "\042"
|
1058 |
#define STR_NUMBER_SIGN "\043"
|
1059 |
#define STR_DOLLAR_SIGN "\044"
|
1060 |
#define STR_PERCENT_SIGN "\045"
|
1061 |
#define STR_AMPERSAND "\046"
|
1062 |
#define STR_APOSTROPHE "\047"
|
1063 |
#define STR_LEFT_PARENTHESIS "\050"
|
1064 |
#define STR_RIGHT_PARENTHESIS "\051"
|
1065 |
#define STR_ASTERISK "\052"
|
1066 |
#define STR_PLUS "\053"
|
1067 |
#define STR_COMMA "\054"
|
1068 |
#define STR_MINUS "\055"
|
1069 |
#define STR_DOT "\056"
|
1070 |
#define STR_SLASH "\057"
|
1071 |
#define STR_0 "\060"
|
1072 |
#define STR_1 "\061"
|
1073 |
#define STR_2 "\062"
|
1074 |
#define STR_3 "\063"
|
1075 |
#define STR_4 "\064"
|
1076 |
#define STR_5 "\065"
|
1077 |
#define STR_6 "\066"
|
1078 |
#define STR_7 "\067"
|
1079 |
#define STR_8 "\070"
|
1080 |
#define STR_9 "\071"
|
1081 |
#define STR_COLON "\072"
|
1082 |
#define STR_SEMICOLON "\073"
|
1083 |
#define STR_LESS_THAN_SIGN "\074"
|
1084 |
#define STR_EQUALS_SIGN "\075"
|
1085 |
#define STR_GREATER_THAN_SIGN "\076"
|
1086 |
#define STR_QUESTION_MARK "\077"
|
1087 |
#define STR_COMMERCIAL_AT "\100"
|
1088 |
#define STR_A "\101"
|
1089 |
#define STR_B "\102"
|
1090 |
#define STR_C "\103"
|
1091 |
#define STR_D "\104"
|
1092 |
#define STR_E "\105"
|
1093 |
#define STR_F "\106"
|
1094 |
#define STR_G "\107"
|
1095 |
#define STR_H "\110"
|
1096 |
#define STR_I "\111"
|
1097 |
#define STR_J "\112"
|
1098 |
#define STR_K "\113"
|
1099 |
#define STR_L "\114"
|
1100 |
#define STR_M "\115"
|
1101 |
#define STR_N "\116"
|
1102 |
#define STR_O "\117"
|
1103 |
#define STR_P "\120"
|
1104 |
#define STR_Q "\121"
|
1105 |
#define STR_R "\122"
|
1106 |
#define STR_S "\123"
|
1107 |
#define STR_T "\124"
|
1108 |
#define STR_U "\125"
|
1109 |
#define STR_V "\126"
|
1110 |
#define STR_W "\127"
|
1111 |
#define STR_X "\130"
|
1112 |
#define STR_Y "\131"
|
1113 |
#define STR_Z "\132"
|
1114 |
#define STR_LEFT_SQUARE_BRACKET "\133"
|
1115 |
#define STR_BACKSLASH "\134"
|
1116 |
#define STR_RIGHT_SQUARE_BRACKET "\135"
|
1117 |
#define STR_CIRCUMFLEX_ACCENT "\136"
|
1118 |
#define STR_UNDERSCORE "\137"
|
1119 |
#define STR_GRAVE_ACCENT "\140"
|
1120 |
#define STR_a "\141"
|
1121 |
#define STR_b "\142"
|
1122 |
#define STR_c "\143"
|
1123 |
#define STR_d "\144"
|
1124 |
#define STR_e "\145"
|
1125 |
#define STR_f "\146"
|
1126 |
#define STR_g "\147"
|
1127 |
#define STR_h "\150"
|
1128 |
#define STR_i "\151"
|
1129 |
#define STR_j "\152"
|
1130 |
#define STR_k "\153"
|
1131 |
#define STR_l "\154"
|
1132 |
#define STR_m "\155"
|
1133 |
#define STR_n "\156"
|
1134 |
#define STR_o "\157"
|
1135 |
#define STR_p "\160"
|
1136 |
#define STR_q "\161"
|
1137 |
#define STR_r "\162"
|
1138 |
#define STR_s "\163"
|
1139 |
#define STR_t "\164"
|
1140 |
#define STR_u "\165"
|
1141 |
#define STR_v "\166"
|
1142 |
#define STR_w "\167"
|
1143 |
#define STR_x "\170"
|
1144 |
#define STR_y "\171"
|
1145 |
#define STR_z "\172"
|
1146 |
#define STR_LEFT_CURLY_BRACKET "\173"
|
1147 |
#define STR_VERTICAL_LINE "\174"
|
1148 |
#define STR_RIGHT_CURLY_BRACKET "\175"
|
1149 |
#define STR_TILDE "\176"
|
1150 |
|
1151 |
#define STRING_ACCEPT0 STR_A STR_C STR_C STR_E STR_P STR_T "\0"
|
1152 |
#define STRING_COMMIT0 STR_C STR_O STR_M STR_M STR_I STR_T "\0"
|
1153 |
#define STRING_F0 STR_F "\0"
|
1154 |
#define STRING_FAIL0 STR_F STR_A STR_I STR_L "\0"
|
1155 |
#define STRING_MARK0 STR_M STR_A STR_R STR_K "\0"
|
1156 |
#define STRING_PRUNE0 STR_P STR_R STR_U STR_N STR_E "\0"
|
1157 |
#define STRING_SKIP0 STR_S STR_K STR_I STR_P "\0"
|
1158 |
#define STRING_THEN STR_T STR_H STR_E STR_N
|
1159 |
|
1160 |
#define STRING_alpha0 STR_a STR_l STR_p STR_h STR_a "\0"
|
1161 |
#define STRING_lower0 STR_l STR_o STR_w STR_e STR_r "\0"
|
1162 |
#define STRING_upper0 STR_u STR_p STR_p STR_e STR_r "\0"
|
1163 |
#define STRING_alnum0 STR_a STR_l STR_n STR_u STR_m "\0"
|
1164 |
#define STRING_ascii0 STR_a STR_s STR_c STR_i STR_i "\0"
|
1165 |
#define STRING_blank0 STR_b STR_l STR_a STR_n STR_k "\0"
|
1166 |
#define STRING_cntrl0 STR_c STR_n STR_t STR_r STR_l "\0"
|
1167 |
#define STRING_digit0 STR_d STR_i STR_g STR_i STR_t "\0"
|
1168 |
#define STRING_graph0 STR_g STR_r STR_a STR_p STR_h "\0"
|
1169 |
#define STRING_print0 STR_p STR_r STR_i STR_n STR_t "\0"
|
1170 |
#define STRING_punct0 STR_p STR_u STR_n STR_c STR_t "\0"
|
1171 |
#define STRING_space0 STR_s STR_p STR_a STR_c STR_e "\0"
|
1172 |
#define STRING_word0 STR_w STR_o STR_r STR_d "\0"
|
1173 |
#define STRING_xdigit STR_x STR_d STR_i STR_g STR_i STR_t
|
1174 |
|
1175 |
#define STRING_DEFINE STR_D STR_E STR_F STR_I STR_N STR_E
|
1176 |
#define STRING_VERSION STR_V STR_E STR_R STR_S STR_I STR_O STR_N
|
1177 |
#define STRING_WEIRD_STARTWORD STR_LEFT_SQUARE_BRACKET STR_COLON STR_LESS_THAN_SIGN STR_COLON STR_RIGHT_SQUARE_BRACKET STR_RIGHT_SQUARE_BRACKET
|
1178 |
#define STRING_WEIRD_ENDWORD STR_LEFT_SQUARE_BRACKET STR_COLON STR_GREATER_THAN_SIGN STR_COLON STR_RIGHT_SQUARE_BRACKET STR_RIGHT_SQUARE_BRACKET
|
1179 |
|
1180 |
#define STRING_CR_RIGHTPAR STR_C STR_R STR_RIGHT_PARENTHESIS
|
1181 |
#define STRING_LF_RIGHTPAR STR_L STR_F STR_RIGHT_PARENTHESIS
|
1182 |
#define STRING_CRLF_RIGHTPAR STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
|
1183 |
#define STRING_ANY_RIGHTPAR STR_A STR_N STR_Y STR_RIGHT_PARENTHESIS
|
1184 |
#define STRING_ANYCRLF_RIGHTPAR STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
|
1185 |
#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
|
1186 |
#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
|
1187 |
#define STRING_UTF8_RIGHTPAR STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS
|
1188 |
#define STRING_UTF16_RIGHTPAR STR_U STR_T STR_F STR_1 STR_6 STR_RIGHT_PARENTHESIS
|
1189 |
#define STRING_UTF32_RIGHTPAR STR_U STR_T STR_F STR_3 STR_2 STR_RIGHT_PARENTHESIS
|
1190 |
#define STRING_UTF_RIGHTPAR STR_U STR_T STR_F STR_RIGHT_PARENTHESIS
|
1191 |
#define STRING_UCP_RIGHTPAR STR_U STR_C STR_P STR_RIGHT_PARENTHESIS
|
1192 |
#define STRING_NO_AUTO_POSSESS_RIGHTPAR STR_N STR_O STR_UNDERSCORE STR_A STR_U STR_T STR_O STR_UNDERSCORE STR_P STR_O STR_S STR_S STR_E STR_S STR_S STR_RIGHT_PARENTHESIS
|
1193 |
#define STRING_NO_DOTSTAR_ANCHOR_RIGHTPAR STR_N STR_O STR_UNDERSCORE STR_D STR_O STR_T STR_S STR_T STR_A STR_R STR_UNDERSCORE STR_A STR_N STR_C STR_H STR_O STR_R STR_RIGHT_PARENTHESIS
|
1194 |
#define STRING_NO_JIT_RIGHTPAR STR_N STR_O STR_UNDERSCORE STR_J STR_I STR_T STR_RIGHT_PARENTHESIS
|
1195 |
#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
|
1196 |
#define STRING_NOTEMPTY_RIGHTPAR STR_N STR_O STR_T STR_E STR_M STR_P STR_T STR_Y STR_RIGHT_PARENTHESIS
|
1197 |
#define STRING_NOTEMPTY_ATSTART_RIGHTPAR STR_N STR_O STR_T STR_E STR_M STR_P STR_T STR_Y STR_UNDERSCORE STR_A STR_T STR_S STR_T STR_A STR_R STR_T STR_RIGHT_PARENTHESIS
|
1198 |
#define STRING_LIMIT_MATCH_EQ STR_L STR_I STR_M STR_I STR_T STR_UNDERSCORE STR_M STR_A STR_T STR_C STR_H STR_EQUALS_SIGN
|
1199 |
#define STRING_LIMIT_RECURSION_EQ STR_L STR_I STR_M STR_I STR_T STR_UNDERSCORE STR_R STR_E STR_C STR_U STR_R STR_S STR_I STR_O STR_N STR_EQUALS_SIGN
|
1200 |
#define STRING_MARK STR_M STR_A STR_R STR_K
|
1201 |
|
1202 |
#endif /* SUPPORT_UNICODE */
|
1203 |
|
1204 |
/* -------------------- End of character and string names -------------------*/
|
1205 |
|
1206 |
/* -------------------- Definitions for compiled patterns -------------------*/
|
1207 |
|
1208 |
/* Codes for different types of Unicode property */
|
1209 |
|
1210 |
#define PT_ANY 0 /* Any property - matches all chars */
|
1211 |
#define PT_LAMP 1 /* L& - the union of Lu, Ll, Lt */
|
1212 |
#define PT_GC 2 /* Specified general characteristic (e.g. L) */
|
1213 |
#define PT_PC 3 /* Specified particular characteristic (e.g. Lu) */
|
1214 |
#define PT_SC 4 /* Script (e.g. Han) */
|
1215 |
#define PT_ALNUM 5 /* Alphanumeric - the union of L and N */
|
1216 |
#define PT_SPACE 6 /* Perl space - Z plus 9,10,12,13 */
|
1217 |
#define PT_PXSPACE 7 /* POSIX space - Z plus 9,10,11,12,13 */
|
1218 |
#define PT_WORD 8 /* Word - L plus N plus underscore */
|
1219 |
#define PT_CLIST 9 /* Pseudo-property: match character list */
|
1220 |
#define PT_UCNC 10 /* Universal Character nameable character */
|
1221 |
#define PT_TABSIZE 11 /* Size of square table for autopossessify tests */
|
1222 |
|
1223 |
/* The following special properties are used only in XCLASS items, when POSIX
|
1224 |
classes are specified and PCRE2_UCP is set - in other words, for Unicode
|
1225 |
handling of these classes. They are not available via the \p or \P escapes like
|
1226 |
those in the above list, and so they do not take part in the autopossessifying
|
1227 |
table. */
|
1228 |
|
1229 |
#define PT_PXGRAPH 11 /* [:graph:] - characters that mark the paper */
|
1230 |
#define PT_PXPRINT 12 /* [:print:] - [:graph:] plus non-control spaces */
|
1231 |
#define PT_PXPUNCT 13 /* [:punct:] - punctuation characters */
|
1232 |
|
1233 |
/* Flag bits and data types for the extended class (OP_XCLASS) for classes that
|
1234 |
contain characters with values greater than 255. */
|
1235 |
|
1236 |
#define XCL_NOT 0x01 /* Flag: this is a negative class */
|
1237 |
#define XCL_MAP 0x02 /* Flag: a 32-byte map is present */
|
1238 |
#define XCL_HASPROP 0x04 /* Flag: property checks are present. */
|
1239 |
|
1240 |
#define XCL_END 0 /* Marks end of individual items */
|
1241 |
#define XCL_SINGLE 1 /* Single item (one multibyte char) follows */
|
1242 |
#define XCL_RANGE 2 /* A range (two multibyte chars) follows */
|
1243 |
#define XCL_PROP 3 /* Unicode property (2-byte property code follows) */
|
1244 |
#define XCL_NOTPROP 4 /* Unicode inverted property (ditto) */
|
1245 |
|
1246 |
/* Escape items that are just an encoding of a particular data value. These
|
1247 |
appear in the escapes[] table in pcre2_compile.c as positive numbers. */
|
1248 |
|
1249 |
#ifndef ESC_a
|
1250 |
#define ESC_a CHAR_BEL
|
1251 |
#endif
|
1252 |
|
1253 |
#ifndef ESC_e
|
1254 |
#define ESC_e CHAR_ESC
|
1255 |
#endif
|
1256 |
|
1257 |
#ifndef ESC_f
|
1258 |
#define ESC_f CHAR_FF
|
1259 |
#endif
|
1260 |
|
1261 |
#ifndef ESC_n
|
1262 |
#define ESC_n CHAR_LF
|
1263 |
#endif
|
1264 |
|
1265 |
#ifndef ESC_r
|
1266 |
#define ESC_r CHAR_CR
|
1267 |
#endif
|
1268 |
|
1269 |
/* We can't officially use ESC_t because it is a POSIX reserved identifier
|
1270 |
(presumably because of all the others like size_t). */
|
1271 |
|
1272 |
#ifndef ESC_tee
|
1273 |
#define ESC_tee CHAR_HT
|
1274 |
#endif
|
1275 |
|
1276 |
/* These are escaped items that aren't just an encoding of a particular data
|
1277 |
value such as \n. They must have non-zero values, as check_escape() returns 0
|
1278 |
for a data character. In the escapes[] table in pcre2_compile.c their values
|
1279 |
are negated in order to distinguish them from data values.
|
1280 |
|
1281 |
They must appear here in the same order as in the opcode definitions below, up
|
1282 |
to ESC_z. There's a dummy for OP_ALLANY because it corresponds to "." in DOTALL
|
1283 |
mode rather than an escape sequence. It is also used for [^] in JavaScript
|
1284 |
compatibility mode, and for \C in non-utf mode. In non-DOTALL mode, "." behaves
|
1285 |
like \N.
|
1286 |
|
1287 |
Negative numbers are used to encode a backreference (\1, \2, \3, etc.) in
|
1288 |
check_escape(). There are tests in the code for an escape greater than ESC_b
|
1289 |
and less than ESC_Z to detect the types that may be repeated. These are the
|
1290 |
types that consume characters. If any new escapes are put in between that don't
|
1291 |
consume a character, that code will have to change. */
|
1292 |
|
1293 |
enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s,
|
1294 |
ESC_W, ESC_w, ESC_N, ESC_dum, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H,
|
1295 |
ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z,
|
1296 |
ESC_E, ESC_Q, ESC_g, ESC_k };
|
1297 |
|
1298 |
|
1299 |
/********************** Opcode definitions ******************/
|
1300 |
|
1301 |
/****** NOTE NOTE NOTE ******
|
1302 |
|
1303 |
Starting from 1 (i.e. after OP_END), the values up to OP_EOD must correspond in
|
1304 |
order to the list of escapes immediately above. Furthermore, values up to
|
1305 |
OP_DOLLM must not be changed without adjusting the table called autoposstab in
|
1306 |
pcre2_auto_possess.c
|
1307 |
|
1308 |
Whenever this list is updated, the two macro definitions that follow must be
|
1309 |
updated to match. The possessification table called "opcode_possessify" in
|
1310 |
pcre2_compile.c must also be updated, and also the tables called "coptable"
|
1311 |
and "poptable" in pcre2_dfa_match.c.
|
1312 |
|
1313 |
****** NOTE NOTE NOTE ******/
|
1314 |
|
1315 |
|
1316 |
/* The values between FIRST_AUTOTAB_OP and LAST_AUTOTAB_RIGHT_OP, inclusive,
|
1317 |
are used in a table for deciding whether a repeated character type can be
|
1318 |
auto-possessified. */
|
1319 |
|
1320 |
#define FIRST_AUTOTAB_OP OP_NOT_DIGIT
|
1321 |
#define LAST_AUTOTAB_LEFT_OP OP_EXTUNI
|
1322 |
#define LAST_AUTOTAB_RIGHT_OP OP_DOLLM
|
1323 |
|
1324 |
enum {
|
1325 |
OP_END, /* 0 End of pattern */
|
1326 |
|
1327 |
/* Values corresponding to backslashed metacharacters */
|
1328 |
|
1329 |
OP_SOD, /* 1 Start of data: \A */
|
1330 |
OP_SOM, /* 2 Start of match (subject + offset): \G */
|
1331 |
OP_SET_SOM, /* 3 Set start of match (\K) */
|
1332 |
OP_NOT_WORD_BOUNDARY, /* 4 \B */
|
1333 |
OP_WORD_BOUNDARY, /* 5 \b */
|
1334 |
OP_NOT_DIGIT, /* 6 \D */
|
1335 |
OP_DIGIT, /* 7 \d */
|
1336 |
OP_NOT_WHITESPACE, /* 8 \S */
|
1337 |
OP_WHITESPACE, /* 9 \s */
|
1338 |
OP_NOT_WORDCHAR, /* 10 \W */
|
1339 |
OP_WORDCHAR, /* 11 \w */
|
1340 |
|
1341 |
OP_ANY, /* 12 Match any character except newline (\N) */
|
1342 |
OP_ALLANY, /* 13 Match any character */
|
1343 |
OP_ANYBYTE, /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */
|
1344 |
OP_NOTPROP, /* 15 \P (not Unicode property) */
|
1345 |
OP_PROP, /* 16 \p (Unicode property) */
|
1346 |
OP_ANYNL, /* 17 \R (any newline sequence) */
|
1347 |
OP_NOT_HSPACE, /* 18 \H (not horizontal whitespace) */
|
1348 |
OP_HSPACE, /* 19 \h (horizontal whitespace) */
|
1349 |
OP_NOT_VSPACE, /* 20 \V (not vertical whitespace) */
|
1350 |
OP_VSPACE, /* 21 \v (vertical whitespace) */
|
1351 |
OP_EXTUNI, /* 22 \X (extended Unicode sequence */
|
1352 |
OP_EODN, /* 23 End of data or \n at end of data (\Z) */
|
1353 |
OP_EOD, /* 24 End of data (\z) */
|
1354 |
|
1355 |
/* Line end assertions */
|
1356 |
|
1357 |
OP_DOLL, /* 25 End of line - not multiline */
|
1358 |
OP_DOLLM, /* 26 End of line - multiline */
|
1359 |
OP_CIRC, /* 27 Start of line - not multiline */
|
1360 |
OP_CIRCM, /* 28 Start of line - multiline */
|
1361 |
|
1362 |
/* Single characters; caseful must precede the caseless ones, and these
|
1363 |
must remain in this order, and adjacent. */
|
1364 |
|
1365 |
OP_CHAR, /* 29 Match one character, casefully */
|
1366 |
OP_CHARI, /* 30 Match one character, caselessly */
|
1367 |
OP_NOT, /* 31 Match one character, not the given one, casefully */
|
1368 |
OP_NOTI, /* 32 Match one character, not the given one, caselessly */
|
1369 |
|
1370 |
/* The following sets of 13 opcodes must always be kept in step because
|
1371 |
the offset from the first one is used to generate the others. */
|
1372 |
|
1373 |
/* Repeated characters; caseful must precede the caseless ones */
|
1374 |
|
1375 |
OP_STAR, /* 33 The maximizing and minimizing versions of */
|
1376 |
OP_MINSTAR, /* 34 these six opcodes must come in pairs, with */
|
1377 |
OP_PLUS, /* 35 the minimizing one second. */
|
1378 |
OP_MINPLUS, /* 36 */
|
1379 |
OP_QUERY, /* 37 */
|
1380 |
OP_MINQUERY, /* 38 */
|
1381 |
|
1382 |
OP_UPTO, /* 39 From 0 to n matches of one character, caseful*/
|
1383 |
OP_MINUPTO, /* 40 */
|
1384 |
OP_EXACT, /* 41 Exactly n matches */
|
1385 |
|
1386 |
OP_POSSTAR, /* 42 Possessified star, caseful */
|
1387 |
OP_POSPLUS, /* 43 Possessified plus, caseful */
|
1388 |
OP_POSQUERY, /* 44 Posesssified query, caseful */
|
1389 |
OP_POSUPTO, /* 45 Possessified upto, caseful */
|
1390 |
|
1391 |
/* Repeated characters; caseless must follow the caseful ones */
|
1392 |
|
1393 |
OP_STARI, /* 46 */
|
1394 |
OP_MINSTARI, /* 47 */
|
1395 |
OP_PLUSI, /* 48 */
|
1396 |
OP_MINPLUSI, /* 49 */
|
1397 |
OP_QUERYI, /* 50 */
|
1398 |
OP_MINQUERYI, /* 51 */
|
1399 |
|
1400 |
OP_UPTOI, /* 52 From 0 to n matches of one character, caseless */
|
1401 |
OP_MINUPTOI, /* 53 */
|
1402 |
OP_EXACTI, /* 54 */
|
1403 |
|
1404 |
OP_POSSTARI, /* 55 Possessified star, caseless */
|
1405 |
OP_POSPLUSI, /* 56 Possessified plus, caseless */
|
1406 |
OP_POSQUERYI, /* 57 Posesssified query, caseless */
|
1407 |
OP_POSUPTOI, /* 58 Possessified upto, caseless */
|
1408 |
|
1409 |
/* The negated ones must follow the non-negated ones, and match them */
|
1410 |
/* Negated repeated character, caseful; must precede the caseless ones */
|
1411 |
|
1412 |
OP_NOTSTAR, /* 59 The maximizing and minimizing versions of */
|
1413 |
OP_NOTMINSTAR, /* 60 these six opcodes must come in pairs, with */
|
1414 |
OP_NOTPLUS, /* 61 the minimizing one second. They must be in */
|
1415 |
OP_NOTMINPLUS, /* 62 exactly the same order as those above. */
|
1416 |
OP_NOTQUERY, /* 63 */
|
1417 |
OP_NOTMINQUERY, /* 64 */
|
1418 |
|
1419 |
OP_NOTUPTO, /* 65 From 0 to n matches, caseful */
|
1420 |
OP_NOTMINUPTO, /* 66 */
|
1421 |
OP_NOTEXACT, /* 67 Exactly n matches */
|
1422 |
|
1423 |
OP_NOTPOSSTAR, /* 68 Possessified versions, caseful */
|
1424 |
OP_NOTPOSPLUS, /* 69 */
|
1425 |
OP_NOTPOSQUERY, /* 70 */
|
1426 |
OP_NOTPOSUPTO, /* 71 */
|
1427 |
|
1428 |
/* Negated repeated character, caseless; must follow the caseful ones */
|
1429 |
|
1430 |
OP_NOTSTARI, /* 72 */
|
1431 |
OP_NOTMINSTARI, /* 73 */
|
1432 |
OP_NOTPLUSI, /* 74 */
|
1433 |
OP_NOTMINPLUSI, /* 75 */
|
1434 |
OP_NOTQUERYI, /* 76 */
|
1435 |
OP_NOTMINQUERYI, /* 77 */
|
1436 |
|
1437 |
OP_NOTUPTOI, /* 78 From 0 to n matches, caseless */
|
1438 |
OP_NOTMINUPTOI, /* 79 */
|
1439 |
OP_NOTEXACTI, /* 80 Exactly n matches */
|
1440 |
|
1441 |
OP_NOTPOSSTARI, /* 81 Possessified versions, caseless */
|
1442 |
OP_NOTPOSPLUSI, /* 82 */
|
1443 |
OP_NOTPOSQUERYI, /* 83 */
|
1444 |
OP_NOTPOSUPTOI, /* 84 */
|
1445 |
|
1446 |
/* Character types */
|
1447 |
|
1448 |
OP_TYPESTAR, /* 85 The maximizing and minimizing versions of */
|
1449 |
OP_TYPEMINSTAR, /* 86 these six opcodes must come in pairs, with */
|
1450 |
OP_TYPEPLUS, /* 87 the minimizing one second. These codes must */
|
1451 |
OP_TYPEMINPLUS, /* 88 be in exactly the same order as those above. */
|
1452 |
OP_TYPEQUERY, /* 89 */
|
1453 |
OP_TYPEMINQUERY, /* 90 */
|
1454 |
|
1455 |
OP_TYPEUPTO, /* 91 From 0 to n matches */
|
1456 |
OP_TYPEMINUPTO, /* 92 */
|
1457 |
OP_TYPEEXACT, /* 93 Exactly n matches */
|
1458 |
|
1459 |
OP_TYPEPOSSTAR, /* 94 Possessified versions */
|
1460 |
OP_TYPEPOSPLUS, /* 95 */
|
1461 |
OP_TYPEPOSQUERY, /* 96 */
|
1462 |
OP_TYPEPOSUPTO, /* 97 */
|
1463 |
|
1464 |
/* These are used for character classes and back references; only the
|
1465 |
first six are the same as the sets above. */
|
1466 |
|
1467 |
OP_CRSTAR, /* 98 The maximizing and minimizing versions of */
|
1468 |
OP_CRMINSTAR, /* 99 all these opcodes must come in pairs, with */
|
1469 |
OP_CRPLUS, /* 100 the minimizing one second. These codes must */
|
1470 |
OP_CRMINPLUS, /* 101 be in exactly the same order as those above. */
|
1471 |
OP_CRQUERY, /* 102 */
|
1472 |
OP_CRMINQUERY, /* 103 */
|
1473 |
|
1474 |
OP_CRRANGE, /* 104 These are different to the three sets above. */
|
1475 |
OP_CRMINRANGE, /* 105 */
|
1476 |
|
1477 |
OP_CRPOSSTAR, /* 106 Possessified versions */
|
1478 |
OP_CRPOSPLUS, /* 107 */
|
1479 |
OP_CRPOSQUERY, /* 108 */
|
1480 |
OP_CRPOSRANGE, /* 109 */
|
1481 |
|
1482 |
/* End of quantifier opcodes */
|
1483 |
|
1484 |
OP_CLASS, /* 110 Match a character class, chars < 256 only */
|
1485 |
OP_NCLASS, /* 111 Same, but the bitmap was created from a negative
|
1486 |
class - the difference is relevant only when a
|
1487 |
character > 255 is encountered. */
|
1488 |
OP_XCLASS, /* 112 Extended class for handling > 255 chars within the
|
1489 |
class. This does both positive and negative. */
|
1490 |
OP_REF, /* 113 Match a back reference, casefully */
|
1491 |
OP_REFI, /* 114 Match a back reference, caselessly */
|
1492 |
OP_DNREF, /* 115 Match a duplicate name backref, casefully */
|
1493 |
OP_DNREFI, /* 116 Match a duplicate name backref, caselessly */
|
1494 |
OP_RECURSE, /* 117 Match a numbered subpattern (possibly recursive) */
|
1495 |
OP_CALLOUT, /* 118 Call out to external function if provided */
|
1496 |
OP_CALLOUT_STR, /* 119 Call out with string argument */
|
1497 |
|
1498 |
OP_ALT, /* 120 Start of alternation */
|
1499 |
OP_KET, /* 121 End of group that doesn't have an unbounded repeat */
|
1500 |
OP_KETRMAX, /* 122 These two must remain together and in this */
|
1501 |
OP_KETRMIN, /* 123 order. They are for groups the repeat for ever. */
|
1502 |
OP_KETRPOS, /* 124 Possessive unlimited repeat. */
|
1503 |
|
1504 |
/* The assertions must come before BRA, CBRA, ONCE, and COND, and the four
|
1505 |
asserts must remain in order. */
|
1506 |
|
1507 |
OP_REVERSE, /* 125 Move pointer back - used in lookbehind assertions */
|
1508 |
OP_ASSERT, /* 126 Positive lookahead */
|
1509 |
OP_ASSERT_NOT, /* 127 Negative lookahead */
|
1510 |
OP_ASSERTBACK, /* 128 Positive lookbehind */
|
1511 |
OP_ASSERTBACK_NOT, /* 129 Negative lookbehind */
|
1512 |
|
1513 |
/* ONCE, ONCE_NC, BRA, BRAPOS, CBRA, CBRAPOS, and COND must come immediately
|
1514 |
after the assertions, with ONCE first, as there's a test for >= ONCE for a
|
1515 |
subpattern that isn't an assertion. The POS versions must immediately follow
|
1516 |
the non-POS versions in each case. */
|
1517 |
|
1518 |
OP_ONCE, /* 130 Atomic group, contains captures */
|
1519 |
OP_ONCE_NC, /* 131 Atomic group containing no captures */
|
1520 |
OP_BRA, /* 132 Start of non-capturing bracket */
|
1521 |
OP_BRAPOS, /* 133 Ditto, with unlimited, possessive repeat */
|
1522 |
OP_CBRA, /* 134 Start of capturing bracket */
|
1523 |
OP_CBRAPOS, /* 135 Ditto, with unlimited, possessive repeat */
|
1524 |
OP_COND, /* 136 Conditional group */
|
1525 |
|
1526 |
/* These five must follow the previous five, in the same order. There's a
|
1527 |
check for >= SBRA to distinguish the two sets. */
|
1528 |
|
1529 |
OP_SBRA, /* 137 Start of non-capturing bracket, check empty */
|
1530 |
OP_SBRAPOS, /* 138 Ditto, with unlimited, possessive repeat */
|
1531 |
OP_SCBRA, /* 139 Start of capturing bracket, check empty */
|
1532 |
OP_SCBRAPOS, /* 140 Ditto, with unlimited, possessive repeat */
|
1533 |
OP_SCOND, /* 141 Conditional group, check empty */
|
1534 |
|
1535 |
/* The next two pairs must (respectively) be kept together. */
|
1536 |
|
1537 |
OP_CREF, /* 142 Used to hold a capture number as condition */
|
1538 |
OP_DNCREF, /* 143 Used to point to duplicate names as a condition */
|
1539 |
OP_RREF, /* 144 Used to hold a recursion number as condition */
|
1540 |
OP_DNRREF, /* 145 Used to point to duplicate names as a condition */
|
1541 |
OP_FALSE, /* 146 Always false (used by DEFINE and VERSION) */
|
1542 |
OP_TRUE, /* 147 Always true (used by VERSION) */
|
1543 |
|
1544 |
OP_BRAZERO, /* 148 These two must remain together and in this */
|
1545 |
OP_BRAMINZERO, /* 149 order. */
|
1546 |
OP_BRAPOSZERO, /* 150 */
|
1547 |
|
1548 |
/* These are backtracking control verbs */
|
1549 |
|
1550 |
OP_MARK, /* 151 always has an argument */
|
1551 |
OP_PRUNE, /* 152 */
|
1552 |
OP_PRUNE_ARG, /* 153 same, but with argument */
|
1553 |
OP_SKIP, /* 154 */
|
1554 |
OP_SKIP_ARG, /* 155 same, but with argument */
|
1555 |
OP_THEN, /* 156 */
|
1556 |
OP_THEN_ARG, /* 157 same, but with argument */
|
1557 |
OP_COMMIT, /* 158 */
|
1558 |
|
1559 |
/* These are forced failure and success verbs */
|
1560 |
|
1561 |
OP_FAIL, /* 159 */
|
1562 |
OP_ACCEPT, /* 160 */
|
1563 |
OP_ASSERT_ACCEPT, /* 161 Used inside assertions */
|
1564 |
OP_CLOSE, /* 162 Used before OP_ACCEPT to close open captures */
|
1565 |
|
1566 |
/* This is used to skip a subpattern with a {0} quantifier */
|
1567 |
|
1568 |
OP_SKIPZERO, /* 163 */
|
1569 |
|
1570 |
/* This is used to identify a DEFINE group during compilation so that it can
|
1571 |
be checked for having only one branch. It is changed to OP_FALSE before
|
1572 |
compilation finishes. */
|
1573 |
|
1574 |
OP_DEFINE, /* 164 */
|
1575 |
|
1576 |
/* This is not an opcode, but is used to check that tables indexed by opcode
|
1577 |
are the correct length, in order to catch updating errors - there have been
|
1578 |
some in the past. */
|
1579 |
|
1580 |
OP_TABLE_LENGTH
|
1581 |
|
1582 |
};
|
1583 |
|
1584 |
/* *** NOTE NOTE NOTE *** Whenever the list above is updated, the two macro
|
1585 |
definitions that follow must also be updated to match. There are also tables
|
1586 |
called "opcode_possessify" in pcre2_compile.c and "coptable" and "poptable" in
|
1587 |
pcre2_dfa_exec.c that must be updated. */
|
1588 |
|
1589 |
|
1590 |
/* This macro defines textual names for all the opcodes. These are used only
|
1591 |
for debugging, and some of them are only partial names. The macro is referenced
|
1592 |
only in pcre2_printint.c, which fills out the full names in many cases (and in
|
1593 |
some cases doesn't actually use these names at all). */
|
1594 |
|
1595 |
#define OP_NAME_LIST \
|
1596 |
"End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d", \
|
1597 |
"\\S", "\\s", "\\W", "\\w", "Any", "AllAny", "Anybyte", \
|
1598 |
"notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v", \
|
1599 |
"extuni", "\\Z", "\\z", \
|
1600 |
"$", "$", "^", "^", "char", "chari", "not", "noti", \
|
1601 |
"*", "*?", "+", "+?", "?", "??", \
|
1602 |
"{", "{", "{", \
|
1603 |
"*+","++", "?+", "{", \
|
1604 |
"*", "*?", "+", "+?", "?", "??", \
|
1605 |
"{", "{", "{", \
|
1606 |
"*+","++", "?+", "{", \
|
1607 |
"*", "*?", "+", "+?", "?", "??", \
|
1608 |
"{", "{", "{", \
|
1609 |
"*+","++", "?+", "{", \
|
1610 |
"*", "*?", "+", "+?", "?", "??", \
|
1611 |
"{", "{", "{", \
|
1612 |
"*+","++", "?+", "{", \
|
1613 |
"*", "*?", "+", "+?", "?", "??", "{", "{", "{", \
|
1614 |
"*+","++", "?+", "{", \
|
1615 |
"*", "*?", "+", "+?", "?", "??", "{", "{", \
|
1616 |
"*+","++", "?+", "{", \
|
1617 |
"class", "nclass", "xclass", "Ref", "Refi", "DnRef", "DnRefi", \
|
1618 |
"Recurse", "Callout", "CalloutStr", \
|
1619 |
"Alt", "Ket", "KetRmax", "KetRmin", "KetRpos", \
|
1620 |
"Reverse", "Assert", "Assert not", "AssertB", "AssertB not", \
|
1621 |
"Once", "Once_NC", \
|
1622 |
"Bra", "BraPos", "CBra", "CBraPos", \
|
1623 |
"Cond", \
|
1624 |
"SBra", "SBraPos", "SCBra", "SCBraPos", \
|
1625 |
"SCond", \
|
1626 |
"Cond ref", "Cond dnref", "Cond rec", "Cond dnrec", \
|
1627 |
"Cond false", "Cond true", \
|
1628 |
"Brazero", "Braminzero", "Braposzero", \
|
1629 |
"*MARK", "*PRUNE", "*PRUNE", "*SKIP", "*SKIP", \
|
1630 |
"*THEN", "*THEN", "*COMMIT", "*FAIL", \
|
1631 |
"*ACCEPT", "*ASSERT_ACCEPT", \
|
1632 |
"Close", "Skip zero", "Define"
|
1633 |
|
1634 |
|
1635 |
/* This macro defines the length of fixed length operations in the compiled
|
1636 |
regex. The lengths are used when searching for specific things, and also in the
|
1637 |
debugging printing of a compiled regex. We use a macro so that it can be
|
1638 |
defined close to the definitions of the opcodes themselves.
|
1639 |
|
1640 |
As things have been extended, some of these are no longer fixed lenths, but are
|
1641 |
minima instead. For example, the length of a single-character repeat may vary
|
1642 |
in UTF-8 mode. The code that uses this table must know about such things. */
|
1643 |
|
1644 |
#define OP_LENGTHS \
|
1645 |
1, /* End */ \
|
1646 |
1, 1, 1, 1, 1, /* \A, \G, \K, \B, \b */ \
|
1647 |
1, 1, 1, 1, 1, 1, /* \D, \d, \S, \s, \W, \w */ \
|
1648 |
1, 1, 1, /* Any, AllAny, Anybyte */ \
|
1649 |
3, 3, /* \P, \p */ \
|
1650 |
1, 1, 1, 1, 1, /* \R, \H, \h, \V, \v */ \
|
1651 |
1, /* \X */ \
|
1652 |
1, 1, 1, 1, 1, 1, /* \Z, \z, $, $M ^, ^M */ \
|
1653 |
2, /* Char - the minimum length */ \
|
1654 |
2, /* Chari - the minimum length */ \
|
1655 |
2, /* not */ \
|
1656 |
2, /* noti */ \
|
1657 |
/* Positive single-char repeats ** These are */ \
|
1658 |
2, 2, 2, 2, 2, 2, /* *, *?, +, +?, ?, ?? ** minima in */ \
|
1659 |
2+IMM2_SIZE, 2+IMM2_SIZE, /* upto, minupto ** mode */ \
|
1660 |
2+IMM2_SIZE, /* exact */ \
|
1661 |
2, 2, 2, 2+IMM2_SIZE, /* *+, ++, ?+, upto+ */ \
|
1662 |
2, 2, 2, 2, 2, 2, /* *I, *?I, +I, +?I, ?I, ??I ** UTF-8 */ \
|
1663 |
2+IMM2_SIZE, 2+IMM2_SIZE, /* upto I, minupto I */ \
|
1664 |
2+IMM2_SIZE, /* exact I */ \
|
1665 |
2, 2, 2, 2+IMM2_SIZE, /* *+I, ++I, ?+I, upto+I */ \
|
1666 |
/* Negative single-char repeats - only for chars < 256 */ \
|
1667 |
2, 2, 2, 2, 2, 2, /* NOT *, *?, +, +?, ?, ?? */ \
|
1668 |
2+IMM2_SIZE, 2+IMM2_SIZE, /* NOT upto, minupto */ \
|
1669 |
2+IMM2_SIZE, /* NOT exact */ \
|
1670 |
2, 2, 2, 2+IMM2_SIZE, /* Possessive NOT *, +, ?, upto */ \
|
1671 |
2, 2, 2, 2, 2, 2, /* NOT *I, *?I, +I, +?I, ?I, ??I */ \
|
1672 |
2+IMM2_SIZE, 2+IMM2_SIZE, /* NOT upto I, minupto I */ \
|
1673 |
2+IMM2_SIZE, /* NOT exact I */ \
|
1674 |
2, 2, 2, 2+IMM2_SIZE, /* Possessive NOT *I, +I, ?I, upto I */ \
|
1675 |
/* Positive type repeats */ \
|
1676 |
2, 2, 2, 2, 2, 2, /* Type *, *?, +, +?, ?, ?? */ \
|
1677 |
2+IMM2_SIZE, 2+IMM2_SIZE, /* Type upto, minupto */ \
|
1678 |
2+IMM2_SIZE, /* Type exact */ \
|
1679 |
2, 2, 2, 2+IMM2_SIZE, /* Possessive *+, ++, ?+, upto+ */ \
|
1680 |
/* Character class & ref repeats */ \
|
1681 |
1, 1, 1, 1, 1, 1, /* *, *?, +, +?, ?, ?? */ \
|
1682 |
1+2*IMM2_SIZE, 1+2*IMM2_SIZE, /* CRRANGE, CRMINRANGE */ \
|
1683 |
1, 1, 1, 1+2*IMM2_SIZE, /* Possessive *+, ++, ?+, CRPOSRANGE */ \
|
1684 |
1+(32/sizeof(PCRE2_UCHAR)), /* CLASS */ \
|
1685 |
1+(32/sizeof(PCRE2_UCHAR)), /* NCLASS */ \
|
1686 |
0, /* XCLASS - variable length */ \
|
1687 |
1+IMM2_SIZE, /* REF */ \
|
1688 |
1+IMM2_SIZE, /* REFI */ \
|
1689 |
1+2*IMM2_SIZE, /* DNREF */ \
|
1690 |
1+2*IMM2_SIZE, /* DNREFI */ \
|
1691 |
1+LINK_SIZE, /* RECURSE */ \
|
1692 |
1+2*LINK_SIZE+1, /* CALLOUT */ \
|
1693 |
0, /* CALLOUT_STR - variable length */ \
|
1694 |
1+LINK_SIZE, /* Alt */ \
|
1695 |
1+LINK_SIZE, /* Ket */ \
|
1696 |
1+LINK_SIZE, /* KetRmax */ \
|
1697 |
1+LINK_SIZE, /* KetRmin */ \
|
1698 |
1+LINK_SIZE, /* KetRpos */ \
|
1699 |
1+LINK_SIZE, /* Reverse */ \
|
1700 |
1+LINK_SIZE, /* Assert */ \
|
1701 |
1+LINK_SIZE, /* Assert not */ \
|
1702 |
1+LINK_SIZE, /* Assert behind */ \
|
1703 |
1+LINK_SIZE, /* Assert behind not */ \
|
1704 |
1+LINK_SIZE, /* ONCE */ \
|
1705 |
1+LINK_SIZE, /* ONCE_NC */ \
|
1706 |
1+LINK_SIZE, /* BRA */ \
|
1707 |
1+LINK_SIZE, /* BRAPOS */ \
|
1708 |
1+LINK_SIZE+IMM2_SIZE, /* CBRA */ \
|
1709 |
1+LINK_SIZE+IMM2_SIZE, /* CBRAPOS */ \
|
1710 |
1+LINK_SIZE, /* COND */ \
|
1711 |
1+LINK_SIZE, /* SBRA */ \
|
1712 |
1+LINK_SIZE, /* SBRAPOS */ \
|
1713 |
1+LINK_SIZE+IMM2_SIZE, /* SCBRA */ \
|
1714 |
1+LINK_SIZE+IMM2_SIZE, /* SCBRAPOS */ \
|
1715 |
1+LINK_SIZE, /* SCOND */ \
|
1716 |
1+IMM2_SIZE, 1+2*IMM2_SIZE, /* CREF, DNCREF */ \
|
1717 |
1+IMM2_SIZE, 1+2*IMM2_SIZE, /* RREF, DNRREF */ \
|
1718 |
1, 1, /* FALSE, TRUE */ \
|
1719 |
1, 1, 1, /* BRAZERO, BRAMINZERO, BRAPOSZERO */ \
|
1720 |
3, 1, 3, /* MARK, PRUNE, PRUNE_ARG */ \
|
1721 |
1, 3, /* SKIP, SKIP_ARG */ \
|
1722 |
1, 3, /* THEN, THEN_ARG */ \
|
1723 |
1, 1, 1, 1, /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT */ \
|
1724 |
1+IMM2_SIZE, 1, /* CLOSE, SKIPZERO */ \
|
1725 |
1 /* DEFINE */
|
1726 |
|
1727 |
/* A magic value for OP_RREF to indicate the "any recursion" condition. */
|
1728 |
|
1729 |
#define RREF_ANY 0xffff
|
1730 |
|
1731 |
|
1732 |
/* ---------- Private structures that are mode-independent. ---------- */
|
1733 |
|
1734 |
/* Structure to hold data for custom memory management. */
|
1735 |
|
1736 |
typedef struct pcre2_memctl {
|
1737 |
void * (*malloc)(size_t, void *);
|
1738 |
void (*free)(void *, void *);
|
1739 |
void *memory_data;
|
1740 |
} pcre2_memctl;
|
1741 |
|
1742 |
/* Structure for building a chain of open capturing subpatterns during
|
1743 |
compiling, so that instructions to close them can be compiled when (*ACCEPT) is
|
1744 |
encountered. This is also used to identify subpatterns that contain recursive
|
1745 |
back references to themselves, so that they can be made atomic. */
|
1746 |
|
1747 |
typedef struct open_capitem {
|
1748 |
struct open_capitem *next; /* Chain link */
|
1749 |
uint16_t number; /* Capture number */
|
1750 |
uint16_t flag; /* Set TRUE if recursive back ref */
|
1751 |
} open_capitem;
|
1752 |
|
1753 |
/* Layout of the UCP type table that translates property names into types and
|
1754 |
codes. Each entry used to point directly to a name, but to reduce the number of
|
1755 |
relocations in shared libraries, it now has an offset into a single string
|
1756 |
instead. */
|
1757 |
|
1758 |
typedef struct {
|
1759 |
uint16_t name_offset;
|
1760 |
uint16_t type;
|
1761 |
uint16_t value;
|
1762 |
} ucp_type_table;
|
1763 |
|
1764 |
/* Unicode character database (UCD) record format */
|
1765 |
|
1766 |
typedef struct {
|
1767 |
uint8_t script; /* ucp_Arabic, etc. */
|
1768 |
uint8_t chartype; /* ucp_Cc, etc. (general categories) */
|
1769 |
uint8_t gbprop; /* ucp_gbControl, etc. (grapheme break property) */
|
1770 |
uint8_t caseset; /* offset to multichar other cases or zero */
|
1771 |
int32_t other_case; /* offset to other case, or zero if none */
|
1772 |
} ucd_record;
|
1773 |
|
1774 |
/* UCD access macros */
|
1775 |
|
1776 |
#define UCD_BLOCK_SIZE 128
|
1777 |
#define REAL_GET_UCD(ch) (PRIV(ucd_records) + \
|
1778 |
PRIV(ucd_stage2)[PRIV(ucd_stage1)[(int)(ch) / UCD_BLOCK_SIZE] * \
|
1779 |
UCD_BLOCK_SIZE + (int)(ch) % UCD_BLOCK_SIZE])
|
1780 |
|
1781 |
#if PCRE2_CODE_UNIT_WIDTH == 32
|
1782 |
#define GET_UCD(ch) ((ch > MAX_UTF_CODE_POINT)? \
|
1783 |
PRIV(dummy_ucd_record) : REAL_GET_UCD(ch))
|
1784 |
#else
|
1785 |
#define GET_UCD(ch) REAL_GET_UCD(ch)
|
1786 |
#endif
|
1787 |
|
1788 |
#define UCD_CHARTYPE(ch) GET_UCD(ch)->chartype
|
1789 |
#define UCD_SCRIPT(ch) GET_UCD(ch)->script
|
1790 |
#define UCD_CATEGORY(ch) PRIV(ucp_gentype)[UCD_CHARTYPE(ch)]
|
1791 |
#define UCD_GRAPHBREAK(ch) GET_UCD(ch)->gbprop
|
1792 |
#define UCD_CASESET(ch) GET_UCD(ch)->caseset
|
1793 |
#define UCD_OTHERCASE(ch) ((uint32_t)((int)ch + (int)(GET_UCD(ch)->other_case)))
|
1794 |
|
1795 |
/* Header for serialized pcre2 codes. */
|
1796 |
|
1797 |
typedef struct pcre2_serialized_data {
|
1798 |
uint32_t magic;
|
1799 |
uint32_t version;
|
1800 |
uint32_t config;
|
1801 |
int32_t number_of_codes;
|
1802 |
} pcre2_serialized_data;
|
1803 |
|
1804 |
|
1805 |
|
1806 |
/* ----------------- Items that need PCRE2_CODE_UNIT_WIDTH ----------------- */
|
1807 |
|
1808 |
/* When this file is included by pcre2test, PCRE2_CODE_UNIT_WIDTH is defined as
|
1809 |
0, so the following items are omitted. */
|
1810 |
|
1811 |
#if defined PCRE2_CODE_UNIT_WIDTH && PCRE2_CODE_UNIT_WIDTH != 0
|
1812 |
|
1813 |
/* EBCDIC is supported only for the 8-bit library. */
|
1814 |
|
1815 |
#if defined EBCDIC && PCRE2_CODE_UNIT_WIDTH != 8
|
1816 |
#error EBCDIC is not supported for the 16-bit or 32-bit libraries
|
1817 |
#endif
|
1818 |
|
1819 |
/* This is the largest non-UTF code point. */
|
1820 |
|
1821 |
#define MAX_NON_UTF_CHAR (0xffffffffU >> (32 - PCRE2_CODE_UNIT_WIDTH))
|
1822 |
|
1823 |
/* Internal shared data tables and variables. These are used by more than one
|
1824 |
of the exported public functions. They have to be "external" in the C sense,
|
1825 |
but are not part of the PCRE2 public API. Although the data for some of them is
|
1826 |
identical in all libraries, they must have different names so that multiple
|
1827 |
libraries can be simultaneously linked to a single application. However, UTF-8
|
1828 |
tables are needed only when compiling the 8-bit library. */
|
1829 |
|
1830 |
#if PCRE2_CODE_UNIT_WIDTH == 8
|
1831 |
extern const int PRIV(utf8_table1)[];
|
1832 |
extern const int PRIV(utf8_table1_size);
|
1833 |
extern const int PRIV(utf8_table2)[];
|
1834 |
extern const int PRIV(utf8_table3)[];
|
1835 |
extern const uint8_t PRIV(utf8_table4)[];
|
1836 |
#endif
|
1837 |
|
1838 |
#define _pcre2_OP_lengths PCRE2_SUFFIX(_pcre2_OP_lengths_)
|
1839 |
#define _pcre2_callout_end_delims PCRE2_SUFFIX(_pcre2_callout_end_delims_)
|
1840 |
#define _pcre2_callout_start_delims PCRE2_SUFFIX(_pcre2_callout_start_delims_)
|
1841 |
#define _pcre2_default_compile_context PCRE2_SUFFIX(_pcre2_default_compile_context_)
|
1842 |
#define _pcre2_default_match_context PCRE2_SUFFIX(_pcre2_default_match_context_)
|
1843 |
#define _pcre2_default_tables PCRE2_SUFFIX(_pcre2_default_tables_)
|
1844 |
#if PCRE2_CODE_UNIT_WIDTH == 32
|
1845 |
#define _pcre2_dummy_ucd_record PCRE2_SUFFIX(_pcre2_dummy_ucd_record_)
|
1846 |
#endif
|
1847 |
#define _pcre2_hspace_list PCRE2_SUFFIX(_pcre2_hspace_list_)
|
1848 |
#define _pcre2_vspace_list PCRE2_SUFFIX(_pcre2_vspace_list_)
|
1849 |
#define _pcre2_ucd_caseless_sets PCRE2_SUFFIX(_pcre2_ucd_caseless_sets_)
|
1850 |
#define _pcre2_ucd_records PCRE2_SUFFIX(_pcre2_ucd_records_)
|
1851 |
#define _pcre2_ucd_stage1 PCRE2_SUFFIX(_pcre2_ucd_stage1_)
|
1852 |
#define _pcre2_ucd_stage2 PCRE2_SUFFIX(_pcre2_ucd_stage2_)
|
1853 |
#define _pcre2_ucp_gbtable PCRE2_SUFFIX(_pcre2_ucp_gbtable_)
|
1854 |
#define _pcre2_ucp_gentype PCRE2_SUFFIX(_pcre2_ucp_gentype_)
|
1855 |
#define _pcre2_ucp_typerange PCRE2_SUFFIX(_pcre2_ucp_typerange_)
|
1856 |
#define _pcre2_unicode_version PCRE2_SUFFIX(_pcre2_unicode_version_)
|
1857 |
#define _pcre2_utt PCRE2_SUFFIX(_pcre2_utt_)
|
1858 |
#define _pcre2_utt_names PCRE2_SUFFIX(_pcre2_utt_names_)
|
1859 |
#define _pcre2_utt_size PCRE2_SUFFIX(_pcre2_utt_size_)
|
1860 |
|
1861 |
extern const uint8_t PRIV(OP_lengths)[];
|
1862 |
extern const uint32_t PRIV(callout_end_delims)[];
|
1863 |
extern const uint32_t PRIV(callout_start_delims)[];
|
1864 |
extern const pcre2_compile_context PRIV(default_compile_context);
|
1865 |
extern const pcre2_match_context PRIV(default_match_context);
|
1866 |
extern const uint8_t PRIV(default_tables)[];
|
1867 |
extern const uint32_t PRIV(hspace_list)[];
|
1868 |
extern const uint32_t PRIV(vspace_list)[];
|
1869 |
extern const uint32_t PRIV(ucd_caseless_sets)[];
|
1870 |
extern const ucd_record PRIV(ucd_records)[];
|
1871 |
#if PCRE2_CODE_UNIT_WIDTH == 32
|
1872 |
extern const ucd_record PRIV(dummy_ucd_record)[];
|
1873 |
#endif
|
1874 |
extern const uint8_t PRIV(ucd_stage1)[];
|
1875 |
extern const uint16_t PRIV(ucd_stage2)[];
|
1876 |
extern const uint32_t PRIV(ucp_gbtable)[];
|
1877 |
extern const uint32_t PRIV(ucp_gentype)[];
|
1878 |
#ifdef SUPPORT_JIT
|
1879 |
extern const int PRIV(ucp_typerange)[];
|
1880 |
#endif
|
1881 |
extern const char *PRIV(unicode_version);
|
1882 |
extern const ucp_type_table PRIV(utt)[];
|
1883 |
extern const char PRIV(utt_names)[];
|
1884 |
extern const size_t PRIV(utt_size);
|
1885 |
|
1886 |
/* Mode-dependent macros and hidden and private structures are defined in a
|
1887 |
separate file so that pcre2test can include them at all supported widths. When
|
1888 |
compiling the library, PCRE2_CODE_UNIT_WIDTH will be defined, and we can
|
1889 |
include them at the appropriate width, after setting up suffix macros for the
|
1890 |
private structures. */
|
1891 |
|
1892 |
#define branch_chain PCRE2_SUFFIX(branch_chain_)
|
1893 |
#define compile_block PCRE2_SUFFIX(compile_block_)
|
1894 |
#define dfa_match_block PCRE2_SUFFIX(dfa_match_block_)
|
1895 |
#define match_block PCRE2_SUFFIX(match_block_)
|
1896 |
#define named_group PCRE2_SUFFIX(named_group_)
|
1897 |
|
1898 |
#include "pcre2_intmodedep.h"
|
1899 |
|
1900 |
/* Private "external" functions. These are internal functions that are called
|
1901 |
from modules other than the one in which they are defined. They have to be
|
1902 |
"external" in the C sense, but are not part of the PCRE2 public API. They are
|
1903 |
not referenced from pcre2test, and must not be defined when no code unit width
|
1904 |
is available. */
|
1905 |
|
1906 |
#define _pcre2_auto_possessify PCRE2_SUFFIX(_pcre2_auto_possessify_)
|
1907 |
#define _pcre2_check_escape PCRE2_SUFFIX(_pcre2_check_escape_)
|
1908 |
#define _pcre2_find_bracket PCRE2_SUFFIX(_pcre2_find_bracket_)
|
1909 |
#define _pcre2_is_newline PCRE2_SUFFIX(_pcre2_is_newline_)
|
1910 |
#define _pcre2_jit_free_rodata PCRE2_SUFFIX(_pcre2_jit_free_rodata_)
|
1911 |
#define _pcre2_jit_free PCRE2_SUFFIX(_pcre2_jit_free_)
|
1912 |
#define _pcre2_jit_get_size PCRE2_SUFFIX(_pcre2_jit_get_size_)
|
1913 |
#define _pcre2_jit_get_target PCRE2_SUFFIX(_pcre2_jit_get_target_)
|
1914 |
#define _pcre2_memctl_malloc PCRE2_SUFFIX(_pcre2_memctl_malloc_)
|
1915 |
#define _pcre2_ord2utf PCRE2_SUFFIX(_pcre2_ord2utf_)
|
1916 |
#define _pcre2_strcmp PCRE2_SUFFIX(_pcre2_strcmp_)
|
1917 |
#define _pcre2_strcmp_c8 PCRE2_SUFFIX(_pcre2_strcmp_c8_)
|
1918 |
#define _pcre2_strcpy_c8 PCRE2_SUFFIX(_pcre2_strcpy_c8_)
|
1919 |
#define _pcre2_strlen PCRE2_SUFFIX(_pcre2_strlen_)
|
1920 |
#define _pcre2_strncmp PCRE2_SUFFIX(_pcre2_strncmp_)
|
1921 |
#define _pcre2_strncmp_c8 PCRE2_SUFFIX(_pcre2_strncmp_c8_)
|
1922 |
#define _pcre2_study PCRE2_SUFFIX(_pcre2_study_)
|
1923 |
#define _pcre2_valid_utf PCRE2_SUFFIX(_pcre2_valid_utf_)
|
1924 |
#define _pcre2_was_newline PCRE2_SUFFIX(_pcre2_was_newline_)
|
1925 |
#define _pcre2_xclass PCRE2_SUFFIX(_pcre2_xclass_)
|
1926 |
|
1927 |
extern int _pcre2_auto_possessify(PCRE2_UCHAR *, BOOL,
|
1928 |
const compile_block *);
|
1929 |
extern int _pcre2_check_escape(PCRE2_SPTR *, PCRE2_SPTR, uint32_t *,
|
1930 |
int *, uint32_t, BOOL, compile_block *);
|
1931 |
extern PCRE2_SPTR _pcre2_find_bracket(PCRE2_SPTR, BOOL, int);
|
1932 |
extern BOOL _pcre2_is_newline(PCRE2_SPTR, uint32_t, PCRE2_SPTR,
|
1933 |
uint32_t *, BOOL);
|
1934 |
extern void _pcre2_jit_free_rodata(void *, void *);
|
1935 |
extern void _pcre2_jit_free(void *, pcre2_memctl *);
|
1936 |
extern size_t _pcre2_jit_get_size(void *);
|
1937 |
const char * _pcre2_jit_get_target(void);
|
1938 |
extern void * _pcre2_memctl_malloc(size_t, pcre2_memctl *);
|
1939 |
extern unsigned int _pcre2_ord2utf(uint32_t, PCRE2_UCHAR *);
|
1940 |
extern int _pcre2_strcmp(PCRE2_SPTR, PCRE2_SPTR);
|
1941 |
extern int _pcre2_strcmp_c8(PCRE2_SPTR, const char *);
|
1942 |
extern PCRE2_SIZE _pcre2_strcpy_c8(PCRE2_UCHAR *, const char *);
|
1943 |
extern PCRE2_SIZE _pcre2_strlen(PCRE2_SPTR);
|
1944 |
extern int _pcre2_strncmp(PCRE2_SPTR, PCRE2_SPTR, size_t);
|
1945 |
extern int _pcre2_strncmp_c8(PCRE2_SPTR, const char *, size_t);
|
1946 |
extern int _pcre2_study(pcre2_real_code *);
|
1947 |
extern int _pcre2_valid_utf(PCRE2_SPTR, PCRE2_SIZE, PCRE2_SIZE *);
|
1948 |
extern BOOL _pcre2_was_newline(PCRE2_SPTR, uint32_t, PCRE2_SPTR,
|
1949 |
uint32_t *, BOOL);
|
1950 |
extern BOOL _pcre2_xclass(uint32_t, PCRE2_SPTR, BOOL);
|
1951 |
#endif /* PCRE2_CODE_UNIT_WIDTH */
|
1952 |
|
1953 |
/* End of pcre2_internal.h */
|