/[pcre]/code/trunk/pcre_compile.c
ViewVC logotype

Contents of /code/trunk/pcre_compile.c

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1066 - (show annotations)
Tue Oct 16 15:54:19 2012 UTC (7 years ago) by chpe
File MIME type: text/plain
File size: 272204 byte(s)
pcre32: compile: Fix signed/unsigned mismatch in read_repeat_counts
1 /*************************************************
2 * Perl-Compatible Regular Expressions *
3 *************************************************/
4
5 /* PCRE 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 Copyright (c) 1997-2012 University of Cambridge
10
11 -----------------------------------------------------------------------------
12 Redistribution and use in source and binary forms, with or without
13 modification, are permitted provided that the following conditions are met:
14
15 * Redistributions of source code must retain the above copyright notice,
16 this list of conditions and the following disclaimer.
17
18 * Redistributions in binary form must reproduce the above copyright
19 notice, this list of conditions and the following disclaimer in the
20 documentation and/or other materials provided with the distribution.
21
22 * Neither the name of the University of Cambridge nor the names of its
23 contributors may be used to endorse or promote products derived from
24 this software without specific prior written permission.
25
26 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
27 AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
28 IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
29 ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
30 LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
31 CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
32 SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
33 INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
34 CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
35 ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
36 POSSIBILITY OF SUCH DAMAGE.
37 -----------------------------------------------------------------------------
38 */
39
40
41 /* This module contains the external function pcre_compile(), along with
42 supporting internal functions that are not used by other modules. */
43
44
45 #ifdef HAVE_CONFIG_H
46 #include "config.h"
47 #endif
48
49 #define NLBLOCK cd /* Block containing newline information */
50 #define PSSTART start_pattern /* Field containing processed string start */
51 #define PSEND end_pattern /* Field containing processed string end */
52
53 #include "pcre_internal.h"
54
55
56 /* When PCRE_DEBUG is defined, we need the pcre(16|32)_printint() function, which
57 is also used by pcretest. PCRE_DEBUG is not defined when building a production
58 library. We do not need to select pcre16_printint.c specially, because the
59 COMPILE_PCREx macro will already be appropriately set. */
60
61 #ifdef PCRE_DEBUG
62 /* pcre_printint.c should not include any headers */
63 #define PCRE_INCLUDED
64 #include "pcre_printint.c"
65 #undef PCRE_INCLUDED
66 #endif
67
68
69 /* Macro for setting individual bits in class bitmaps. */
70
71 #define SETBIT(a,b) a[(b)/8] |= (1 << ((b)&7))
72
73 /* Maximum length value to check against when making sure that the integer that
74 holds the compiled pattern length does not overflow. We make it a bit less than
75 INT_MAX to allow for adding in group terminating bytes, so that we don't have
76 to check them every time. */
77
78 #define OFLOW_MAX (INT_MAX - 20)
79
80 /* Definitions to allow mutual recursion */
81
82 static int
83 add_list_to_class(pcre_uint8 *, pcre_uchar **, int, compile_data *,
84 const pcre_uint32 *, unsigned int);
85
86 static BOOL
87 compile_regex(int, pcre_uchar **, const pcre_uchar **, int *, BOOL, BOOL,
88 int, int, int *, int *, branch_chain *, compile_data *, int *);
89
90
91
92 /*************************************************
93 * Code parameters and static tables *
94 *************************************************/
95
96 /* This value specifies the size of stack workspace that is used during the
97 first pre-compile phase that determines how much memory is required. The regex
98 is partly compiled into this space, but the compiled parts are discarded as
99 soon as they can be, so that hopefully there will never be an overrun. The code
100 does, however, check for an overrun. The largest amount I've seen used is 218,
101 so this number is very generous.
102
103 The same workspace is used during the second, actual compile phase for
104 remembering forward references to groups so that they can be filled in at the
105 end. Each entry in this list occupies LINK_SIZE bytes, so even when LINK_SIZE
106 is 4 there is plenty of room for most patterns. However, the memory can get
107 filled up by repetitions of forward references, for example patterns like
108 /(?1){0,1999}(b)/, and one user did hit the limit. The code has been changed so
109 that the workspace is expanded using malloc() in this situation. The value
110 below is therefore a minimum, and we put a maximum on it for safety. The
111 minimum is now also defined in terms of LINK_SIZE so that the use of malloc()
112 kicks in at the same number of forward references in all cases. */
113
114 #define COMPILE_WORK_SIZE (2048*LINK_SIZE)
115 #define COMPILE_WORK_SIZE_MAX (100*COMPILE_WORK_SIZE)
116
117 /* The overrun tests check for a slightly smaller size so that they detect the
118 overrun before it actually does run off the end of the data block. */
119
120 #define WORK_SIZE_SAFETY_MARGIN (100)
121
122 /* Private flags added to firstchar and reqchar. */
123
124 #define REQ_CASELESS 0x10000000l /* Indicates caselessness */
125 #define REQ_VARY 0x20000000l /* Reqchar followed non-literal item */
126 #define REQ_MASK (REQ_CASELESS | REQ_VARY)
127
128 /* Repeated character flags. */
129
130 #define UTF_LENGTH 0x10000000l /* The char contains its length. */
131
132 /* Table for handling escaped characters in the range '0'-'z'. Positive returns
133 are simple data values; negative values are for special things like \d and so
134 on. Zero means further processing is needed (for things like \x), or the escape
135 is invalid. */
136
137 #ifndef EBCDIC
138
139 /* This is the "normal" table for ASCII systems or for EBCDIC systems running
140 in UTF-8 mode. */
141
142 static const short int escapes[] = {
143 0, 0,
144 0, 0,
145 0, 0,
146 0, 0,
147 0, 0,
148 CHAR_COLON, CHAR_SEMICOLON,
149 CHAR_LESS_THAN_SIGN, CHAR_EQUALS_SIGN,
150 CHAR_GREATER_THAN_SIGN, CHAR_QUESTION_MARK,
151 CHAR_COMMERCIAL_AT, -ESC_A,
152 -ESC_B, -ESC_C,
153 -ESC_D, -ESC_E,
154 0, -ESC_G,
155 -ESC_H, 0,
156 0, -ESC_K,
157 0, 0,
158 -ESC_N, 0,
159 -ESC_P, -ESC_Q,
160 -ESC_R, -ESC_S,
161 0, 0,
162 -ESC_V, -ESC_W,
163 -ESC_X, 0,
164 -ESC_Z, CHAR_LEFT_SQUARE_BRACKET,
165 CHAR_BACKSLASH, CHAR_RIGHT_SQUARE_BRACKET,
166 CHAR_CIRCUMFLEX_ACCENT, CHAR_UNDERSCORE,
167 CHAR_GRAVE_ACCENT, 7,
168 -ESC_b, 0,
169 -ESC_d, ESC_e,
170 ESC_f, 0,
171 -ESC_h, 0,
172 0, -ESC_k,
173 0, 0,
174 ESC_n, 0,
175 -ESC_p, 0,
176 ESC_r, -ESC_s,
177 ESC_tee, 0,
178 -ESC_v, -ESC_w,
179 0, 0,
180 -ESC_z
181 };
182
183 #else
184
185 /* This is the "abnormal" table for EBCDIC systems without UTF-8 support. */
186
187 static const short int escapes[] = {
188 /* 48 */ 0, 0, 0, '.', '<', '(', '+', '|',
189 /* 50 */ '&', 0, 0, 0, 0, 0, 0, 0,
190 /* 58 */ 0, 0, '!', '$', '*', ')', ';', '~',
191 /* 60 */ '-', '/', 0, 0, 0, 0, 0, 0,
192 /* 68 */ 0, 0, '|', ',', '%', '_', '>', '?',
193 /* 70 */ 0, 0, 0, 0, 0, 0, 0, 0,
194 /* 78 */ 0, '`', ':', '#', '@', '\'', '=', '"',
195 /* 80 */ 0, 7, -ESC_b, 0, -ESC_d, ESC_e, ESC_f, 0,
196 /* 88 */-ESC_h, 0, 0, '{', 0, 0, 0, 0,
197 /* 90 */ 0, 0, -ESC_k, 'l', 0, ESC_n, 0, -ESC_p,
198 /* 98 */ 0, ESC_r, 0, '}', 0, 0, 0, 0,
199 /* A0 */ 0, '~', -ESC_s, ESC_tee, 0,-ESC_v, -ESC_w, 0,
200 /* A8 */ 0,-ESC_z, 0, 0, 0, '[', 0, 0,
201 /* B0 */ 0, 0, 0, 0, 0, 0, 0, 0,
202 /* B8 */ 0, 0, 0, 0, 0, ']', '=', '-',
203 /* C0 */ '{',-ESC_A, -ESC_B, -ESC_C, -ESC_D,-ESC_E, 0, -ESC_G,
204 /* C8 */-ESC_H, 0, 0, 0, 0, 0, 0, 0,
205 /* D0 */ '}', 0, -ESC_K, 0, 0,-ESC_N, 0, -ESC_P,
206 /* D8 */-ESC_Q,-ESC_R, 0, 0, 0, 0, 0, 0,
207 /* E0 */ '\\', 0, -ESC_S, 0, 0,-ESC_V, -ESC_W, -ESC_X,
208 /* E8 */ 0,-ESC_Z, 0, 0, 0, 0, 0, 0,
209 /* F0 */ 0, 0, 0, 0, 0, 0, 0, 0,
210 /* F8 */ 0, 0, 0, 0, 0, 0, 0, 0
211 };
212 #endif
213
214
215 /* Table of special "verbs" like (*PRUNE). This is a short table, so it is
216 searched linearly. Put all the names into a single string, in order to reduce
217 the number of relocations when a shared library is dynamically linked. The
218 string is built from string macros so that it works in UTF-8 mode on EBCDIC
219 platforms. */
220
221 typedef struct verbitem {
222 int len; /* Length of verb name */
223 int op; /* Op when no arg, or -1 if arg mandatory */
224 int op_arg; /* Op when arg present, or -1 if not allowed */
225 } verbitem;
226
227 static const char verbnames[] =
228 "\0" /* Empty name is a shorthand for MARK */
229 STRING_MARK0
230 STRING_ACCEPT0
231 STRING_COMMIT0
232 STRING_F0
233 STRING_FAIL0
234 STRING_PRUNE0
235 STRING_SKIP0
236 STRING_THEN;
237
238 static const verbitem verbs[] = {
239 { 0, -1, OP_MARK },
240 { 4, -1, OP_MARK },
241 { 6, OP_ACCEPT, -1 },
242 { 6, OP_COMMIT, -1 },
243 { 1, OP_FAIL, -1 },
244 { 4, OP_FAIL, -1 },
245 { 5, OP_PRUNE, OP_PRUNE_ARG },
246 { 4, OP_SKIP, OP_SKIP_ARG },
247 { 4, OP_THEN, OP_THEN_ARG }
248 };
249
250 static const int verbcount = sizeof(verbs)/sizeof(verbitem);
251
252
253 /* Tables of names of POSIX character classes and their lengths. The names are
254 now all in a single string, to reduce the number of relocations when a shared
255 library is dynamically loaded. The list of lengths is terminated by a zero
256 length entry. The first three must be alpha, lower, upper, as this is assumed
257 for handling case independence. */
258
259 static const char posix_names[] =
260 STRING_alpha0 STRING_lower0 STRING_upper0 STRING_alnum0
261 STRING_ascii0 STRING_blank0 STRING_cntrl0 STRING_digit0
262 STRING_graph0 STRING_print0 STRING_punct0 STRING_space0
263 STRING_word0 STRING_xdigit;
264
265 static const pcre_uint8 posix_name_lengths[] = {
266 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 4, 6, 0 };
267
268 /* Table of class bit maps for each POSIX class. Each class is formed from a
269 base map, with an optional addition or removal of another map. Then, for some
270 classes, there is some additional tweaking: for [:blank:] the vertical space
271 characters are removed, and for [:alpha:] and [:alnum:] the underscore
272 character is removed. The triples in the table consist of the base map offset,
273 second map offset or -1 if no second map, and a non-negative value for map
274 addition or a negative value for map subtraction (if there are two maps). The
275 absolute value of the third field has these meanings: 0 => no tweaking, 1 =>
276 remove vertical space characters, 2 => remove underscore. */
277
278 static const int posix_class_maps[] = {
279 cbit_word, cbit_digit, -2, /* alpha */
280 cbit_lower, -1, 0, /* lower */
281 cbit_upper, -1, 0, /* upper */
282 cbit_word, -1, 2, /* alnum - word without underscore */
283 cbit_print, cbit_cntrl, 0, /* ascii */
284 cbit_space, -1, 1, /* blank - a GNU extension */
285 cbit_cntrl, -1, 0, /* cntrl */
286 cbit_digit, -1, 0, /* digit */
287 cbit_graph, -1, 0, /* graph */
288 cbit_print, -1, 0, /* print */
289 cbit_punct, -1, 0, /* punct */
290 cbit_space, -1, 0, /* space */
291 cbit_word, -1, 0, /* word - a Perl extension */
292 cbit_xdigit,-1, 0 /* xdigit */
293 };
294
295 /* Table of substitutes for \d etc when PCRE_UCP is set. The POSIX class
296 substitutes must be in the order of the names, defined above, and there are
297 both positive and negative cases. NULL means no substitute. */
298
299 #ifdef SUPPORT_UCP
300 static const pcre_uchar string_PNd[] = {
301 CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET,
302 CHAR_N, CHAR_d, CHAR_RIGHT_CURLY_BRACKET, '\0' };
303 static const pcre_uchar string_pNd[] = {
304 CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET,
305 CHAR_N, CHAR_d, CHAR_RIGHT_CURLY_BRACKET, '\0' };
306 static const pcre_uchar string_PXsp[] = {
307 CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET,
308 CHAR_X, CHAR_s, CHAR_p, CHAR_RIGHT_CURLY_BRACKET, '\0' };
309 static const pcre_uchar string_pXsp[] = {
310 CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET,
311 CHAR_X, CHAR_s, CHAR_p, CHAR_RIGHT_CURLY_BRACKET, '\0' };
312 static const pcre_uchar string_PXwd[] = {
313 CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET,
314 CHAR_X, CHAR_w, CHAR_d, CHAR_RIGHT_CURLY_BRACKET, '\0' };
315 static const pcre_uchar string_pXwd[] = {
316 CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET,
317 CHAR_X, CHAR_w, CHAR_d, CHAR_RIGHT_CURLY_BRACKET, '\0' };
318
319 static const pcre_uchar *substitutes[] = {
320 string_PNd, /* \D */
321 string_pNd, /* \d */
322 string_PXsp, /* \S */ /* NOTE: Xsp is Perl space */
323 string_pXsp, /* \s */
324 string_PXwd, /* \W */
325 string_pXwd /* \w */
326 };
327
328 static const pcre_uchar string_pL[] = {
329 CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET,
330 CHAR_L, CHAR_RIGHT_CURLY_BRACKET, '\0' };
331 static const pcre_uchar string_pLl[] = {
332 CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET,
333 CHAR_L, CHAR_l, CHAR_RIGHT_CURLY_BRACKET, '\0' };
334 static const pcre_uchar string_pLu[] = {
335 CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET,
336 CHAR_L, CHAR_u, CHAR_RIGHT_CURLY_BRACKET, '\0' };
337 static const pcre_uchar string_pXan[] = {
338 CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET,
339 CHAR_X, CHAR_a, CHAR_n, CHAR_RIGHT_CURLY_BRACKET, '\0' };
340 static const pcre_uchar string_h[] = {
341 CHAR_BACKSLASH, CHAR_h, '\0' };
342 static const pcre_uchar string_pXps[] = {
343 CHAR_BACKSLASH, CHAR_p, CHAR_LEFT_CURLY_BRACKET,
344 CHAR_X, CHAR_p, CHAR_s, CHAR_RIGHT_CURLY_BRACKET, '\0' };
345 static const pcre_uchar string_PL[] = {
346 CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET,
347 CHAR_L, CHAR_RIGHT_CURLY_BRACKET, '\0' };
348 static const pcre_uchar string_PLl[] = {
349 CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET,
350 CHAR_L, CHAR_l, CHAR_RIGHT_CURLY_BRACKET, '\0' };
351 static const pcre_uchar string_PLu[] = {
352 CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET,
353 CHAR_L, CHAR_u, CHAR_RIGHT_CURLY_BRACKET, '\0' };
354 static const pcre_uchar string_PXan[] = {
355 CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET,
356 CHAR_X, CHAR_a, CHAR_n, CHAR_RIGHT_CURLY_BRACKET, '\0' };
357 static const pcre_uchar string_H[] = {
358 CHAR_BACKSLASH, CHAR_H, '\0' };
359 static const pcre_uchar string_PXps[] = {
360 CHAR_BACKSLASH, CHAR_P, CHAR_LEFT_CURLY_BRACKET,
361 CHAR_X, CHAR_p, CHAR_s, CHAR_RIGHT_CURLY_BRACKET, '\0' };
362
363 static const pcre_uchar *posix_substitutes[] = {
364 string_pL, /* alpha */
365 string_pLl, /* lower */
366 string_pLu, /* upper */
367 string_pXan, /* alnum */
368 NULL, /* ascii */
369 string_h, /* blank */
370 NULL, /* cntrl */
371 string_pNd, /* digit */
372 NULL, /* graph */
373 NULL, /* print */
374 NULL, /* punct */
375 string_pXps, /* space */ /* NOTE: Xps is POSIX space */
376 string_pXwd, /* word */
377 NULL, /* xdigit */
378 /* Negated cases */
379 string_PL, /* ^alpha */
380 string_PLl, /* ^lower */
381 string_PLu, /* ^upper */
382 string_PXan, /* ^alnum */
383 NULL, /* ^ascii */
384 string_H, /* ^blank */
385 NULL, /* ^cntrl */
386 string_PNd, /* ^digit */
387 NULL, /* ^graph */
388 NULL, /* ^print */
389 NULL, /* ^punct */
390 string_PXps, /* ^space */ /* NOTE: Xps is POSIX space */
391 string_PXwd, /* ^word */
392 NULL /* ^xdigit */
393 };
394 #define POSIX_SUBSIZE (sizeof(posix_substitutes) / sizeof(pcre_uchar *))
395 #endif
396
397 #define STRING(a) # a
398 #define XSTRING(s) STRING(s)
399
400 /* The texts of compile-time error messages. These are "char *" because they
401 are passed to the outside world. Do not ever re-use any error number, because
402 they are documented. Always add a new error instead. Messages marked DEAD below
403 are no longer used. This used to be a table of strings, but in order to reduce
404 the number of relocations needed when a shared library is loaded dynamically,
405 it is now one long string. We cannot use a table of offsets, because the
406 lengths of inserts such as XSTRING(MAX_NAME_SIZE) are not known. Instead, we
407 simply count through to the one we want - this isn't a performance issue
408 because these strings are used only when there is a compilation error.
409
410 Each substring ends with \0 to insert a null character. This includes the final
411 substring, so that the whole string ends with \0\0, which can be detected when
412 counting through. */
413
414 static const char error_texts[] =
415 "no error\0"
416 "\\ at end of pattern\0"
417 "\\c at end of pattern\0"
418 "unrecognized character follows \\\0"
419 "numbers out of order in {} quantifier\0"
420 /* 5 */
421 "number too big in {} quantifier\0"
422 "missing terminating ] for character class\0"
423 "invalid escape sequence in character class\0"
424 "range out of order in character class\0"
425 "nothing to repeat\0"
426 /* 10 */
427 "operand of unlimited repeat could match the empty string\0" /** DEAD **/
428 "internal error: unexpected repeat\0"
429 "unrecognized character after (? or (?-\0"
430 "POSIX named classes are supported only within a class\0"
431 "missing )\0"
432 /* 15 */
433 "reference to non-existent subpattern\0"
434 "erroffset passed as NULL\0"
435 "unknown option bit(s) set\0"
436 "missing ) after comment\0"
437 "parentheses nested too deeply\0" /** DEAD **/
438 /* 20 */
439 "regular expression is too large\0"
440 "failed to get memory\0"
441 "unmatched parentheses\0"
442 "internal error: code overflow\0"
443 "unrecognized character after (?<\0"
444 /* 25 */
445 "lookbehind assertion is not fixed length\0"
446 "malformed number or name after (?(\0"
447 "conditional group contains more than two branches\0"
448 "assertion expected after (?(\0"
449 "(?R or (?[+-]digits must be followed by )\0"
450 /* 30 */
451 "unknown POSIX class name\0"
452 "POSIX collating elements are not supported\0"
453 "this version of PCRE is compiled without UTF support\0"
454 "spare error\0" /** DEAD **/
455 "character value in \\x{...} sequence is too large\0"
456 /* 35 */
457 "invalid condition (?(0)\0"
458 "\\C not allowed in lookbehind assertion\0"
459 "PCRE does not support \\L, \\l, \\N{name}, \\U, or \\u\0"
460 "number after (?C is > 255\0"
461 "closing ) for (?C expected\0"
462 /* 40 */
463 "recursive call could loop indefinitely\0"
464 "unrecognized character after (?P\0"
465 "syntax error in subpattern name (missing terminator)\0"
466 "two named subpatterns have the same name\0"
467 "invalid UTF-8 string\0"
468 /* 45 */
469 "support for \\P, \\p, and \\X has not been compiled\0"
470 "malformed \\P or \\p sequence\0"
471 "unknown property name after \\P or \\p\0"
472 "subpattern name is too long (maximum " XSTRING(MAX_NAME_SIZE) " characters)\0"
473 "too many named subpatterns (maximum " XSTRING(MAX_NAME_COUNT) ")\0"
474 /* 50 */
475 "repeated subpattern is too long\0" /** DEAD **/
476 "octal value is greater than \\377 in 8-bit non-UTF-8 mode\0"
477 "internal error: overran compiling workspace\0"
478 "internal error: previously-checked referenced subpattern not found\0"
479 "DEFINE group contains more than one branch\0"
480 /* 55 */
481 "repeating a DEFINE group is not allowed\0" /** DEAD **/
482 "inconsistent NEWLINE options\0"
483 "\\g is not followed by a braced, angle-bracketed, or quoted name/number or by a plain number\0"
484 "a numbered reference must not be zero\0"
485 "an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)\0"
486 /* 60 */
487 "(*VERB) not recognized\0"
488 "number is too big\0"
489 "subpattern name expected\0"
490 "digit expected after (?+\0"
491 "] is an invalid data character in JavaScript compatibility mode\0"
492 /* 65 */
493 "different names for subpatterns of the same number are not allowed\0"
494 "(*MARK) must have an argument\0"
495 "this version of PCRE is not compiled with Unicode property support\0"
496 "\\c must be followed by an ASCII character\0"
497 "\\k is not followed by a braced, angle-bracketed, or quoted name\0"
498 /* 70 */
499 "internal error: unknown opcode in find_fixedlength()\0"
500 "\\N is not supported in a class\0"
501 "too many forward references\0"
502 "disallowed Unicode code point (>= 0xd800 && <= 0xdfff)\0"
503 "invalid UTF-16 string\0"
504 /* 75 */
505 "name is too long in (*MARK), (*PRUNE), (*SKIP), or (*THEN)\0"
506 "character value in \\u.... sequence is too large\0"
507 "invalid UTF-32 string\0"
508 ;
509
510 /* Table to identify digits and hex digits. This is used when compiling
511 patterns. Note that the tables in chartables are dependent on the locale, and
512 may mark arbitrary characters as digits - but the PCRE compiling code expects
513 to handle only 0-9, a-z, and A-Z as digits when compiling. That is why we have
514 a private table here. It costs 256 bytes, but it is a lot faster than doing
515 character value tests (at least in some simple cases I timed), and in some
516 applications one wants PCRE to compile efficiently as well as match
517 efficiently.
518
519 For convenience, we use the same bit definitions as in chartables:
520
521 0x04 decimal digit
522 0x08 hexadecimal digit
523
524 Then we can use ctype_digit and ctype_xdigit in the code. */
525
526 /* Using a simple comparison for decimal numbers rather than a memory read
527 is much faster, and the resulting code is simpler (the compiler turns it
528 into a subtraction and unsigned comparison). */
529
530 #define IS_DIGIT(x) ((x) >= CHAR_0 && (x) <= CHAR_9)
531
532 #ifndef EBCDIC
533
534 /* This is the "normal" case, for ASCII systems, and EBCDIC systems running in
535 UTF-8 mode. */
536
537 static const pcre_uint8 digitab[] =
538 {
539 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 0- 7 */
540 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 8- 15 */
541 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 16- 23 */
542 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 24- 31 */
543 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - ' */
544 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* ( - / */
545 0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c, /* 0 - 7 */
546 0x0c,0x0c,0x00,0x00,0x00,0x00,0x00,0x00, /* 8 - ? */
547 0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /* @ - G */
548 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* H - O */
549 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* P - W */
550 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* X - _ */
551 0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /* ` - g */
552 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* h - o */
553 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* p - w */
554 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* x -127 */
555 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 128-135 */
556 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 136-143 */
557 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 144-151 */
558 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 152-159 */
559 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 160-167 */
560 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 168-175 */
561 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 176-183 */
562 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 184-191 */
563 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 192-199 */
564 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 200-207 */
565 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 208-215 */
566 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 216-223 */
567 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 224-231 */
568 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 232-239 */
569 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 240-247 */
570 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00};/* 248-255 */
571
572 #else
573
574 /* This is the "abnormal" case, for EBCDIC systems not running in UTF-8 mode. */
575
576 static const pcre_uint8 digitab[] =
577 {
578 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 0- 7 0 */
579 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 8- 15 */
580 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 16- 23 10 */
581 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 24- 31 */
582 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 32- 39 20 */
583 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 40- 47 */
584 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 48- 55 30 */
585 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 56- 63 */
586 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - 71 40 */
587 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 72- | */
588 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* & - 87 50 */
589 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 88- 95 */
590 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - -103 60 */
591 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 104- ? */
592 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 112-119 70 */
593 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 120- " */
594 0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /* 128- g 80 */
595 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* h -143 */
596 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 144- p 90 */
597 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* q -159 */
598 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 160- x A0 */
599 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* y -175 */
600 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* ^ -183 B0 */
601 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 184-191 */
602 0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /* { - G C0 */
603 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* H -207 */
604 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* } - P D0 */
605 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* Q -223 */
606 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* \ - X E0 */
607 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* Y -239 */
608 0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c, /* 0 - 7 F0 */
609 0x0c,0x0c,0x00,0x00,0x00,0x00,0x00,0x00};/* 8 -255 */
610
611 static const pcre_uint8 ebcdic_chartab[] = { /* chartable partial dup */
612 0x80,0x00,0x00,0x00,0x00,0x01,0x00,0x00, /* 0- 7 */
613 0x00,0x00,0x00,0x00,0x01,0x01,0x00,0x00, /* 8- 15 */
614 0x00,0x00,0x00,0x00,0x00,0x01,0x00,0x00, /* 16- 23 */
615 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 24- 31 */
616 0x00,0x00,0x00,0x00,0x00,0x01,0x00,0x00, /* 32- 39 */
617 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 40- 47 */
618 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 48- 55 */
619 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 56- 63 */
620 0x01,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - 71 */
621 0x00,0x00,0x00,0x80,0x00,0x80,0x80,0x80, /* 72- | */
622 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* & - 87 */
623 0x00,0x00,0x00,0x80,0x80,0x80,0x00,0x00, /* 88- 95 */
624 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - -103 */
625 0x00,0x00,0x00,0x00,0x00,0x10,0x00,0x80, /* 104- ? */
626 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 112-119 */
627 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 120- " */
628 0x00,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /* 128- g */
629 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* h -143 */
630 0x00,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* 144- p */
631 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* q -159 */
632 0x00,0x00,0x12,0x12,0x12,0x12,0x12,0x12, /* 160- x */
633 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* y -175 */
634 0x80,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* ^ -183 */
635 0x00,0x00,0x80,0x00,0x00,0x00,0x00,0x00, /* 184-191 */
636 0x80,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /* { - G */
637 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* H -207 */
638 0x00,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* } - P */
639 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* Q -223 */
640 0x00,0x00,0x12,0x12,0x12,0x12,0x12,0x12, /* \ - X */
641 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* Y -239 */
642 0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c, /* 0 - 7 */
643 0x1c,0x1c,0x00,0x00,0x00,0x00,0x00,0x00};/* 8 -255 */
644 #endif
645
646
647
648
649 /*************************************************
650 * Find an error text *
651 *************************************************/
652
653 /* The error texts are now all in one long string, to save on relocations. As
654 some of the text is of unknown length, we can't use a table of offsets.
655 Instead, just count through the strings. This is not a performance issue
656 because it happens only when there has been a compilation error.
657
658 Argument: the error number
659 Returns: pointer to the error string
660 */
661
662 static const char *
663 find_error_text(int n)
664 {
665 const char *s = error_texts;
666 for (; n > 0; n--)
667 {
668 while (*s++ != 0) {};
669 if (*s == 0) return "Error text not found (please report)";
670 }
671 return s;
672 }
673
674
675 /*************************************************
676 * Expand the workspace *
677 *************************************************/
678
679 /* This function is called during the second compiling phase, if the number of
680 forward references fills the existing workspace, which is originally a block on
681 the stack. A larger block is obtained from malloc() unless the ultimate limit
682 has been reached or the increase will be rather small.
683
684 Argument: pointer to the compile data block
685 Returns: 0 if all went well, else an error number
686 */
687
688 static int
689 expand_workspace(compile_data *cd)
690 {
691 pcre_uchar *newspace;
692 int newsize = cd->workspace_size * 2;
693
694 if (newsize > COMPILE_WORK_SIZE_MAX) newsize = COMPILE_WORK_SIZE_MAX;
695 if (cd->workspace_size >= COMPILE_WORK_SIZE_MAX ||
696 newsize - cd->workspace_size < WORK_SIZE_SAFETY_MARGIN)
697 return ERR72;
698
699 newspace = (PUBL(malloc))(IN_UCHARS(newsize));
700 if (newspace == NULL) return ERR21;
701 memcpy(newspace, cd->start_workspace, cd->workspace_size * sizeof(pcre_uchar));
702 cd->hwm = (pcre_uchar *)newspace + (cd->hwm - cd->start_workspace);
703 if (cd->workspace_size > COMPILE_WORK_SIZE)
704 (PUBL(free))((void *)cd->start_workspace);
705 cd->start_workspace = newspace;
706 cd->workspace_size = newsize;
707 return 0;
708 }
709
710
711
712 /*************************************************
713 * Check for counted repeat *
714 *************************************************/
715
716 /* This function is called when a '{' is encountered in a place where it might
717 start a quantifier. It looks ahead to see if it really is a quantifier or not.
718 It is only a quantifier if it is one of the forms {ddd} {ddd,} or {ddd,ddd}
719 where the ddds are digits.
720
721 Arguments:
722 p pointer to the first char after '{'
723
724 Returns: TRUE or FALSE
725 */
726
727 static BOOL
728 is_counted_repeat(const pcre_uchar *p)
729 {
730 if (!IS_DIGIT(*p)) return FALSE;
731 p++;
732 while (IS_DIGIT(*p)) p++;
733 if (*p == CHAR_RIGHT_CURLY_BRACKET) return TRUE;
734
735 if (*p++ != CHAR_COMMA) return FALSE;
736 if (*p == CHAR_RIGHT_CURLY_BRACKET) return TRUE;
737
738 if (!IS_DIGIT(*p)) return FALSE;
739 p++;
740 while (IS_DIGIT(*p)) p++;
741
742 return (*p == CHAR_RIGHT_CURLY_BRACKET);
743 }
744
745
746
747 /*************************************************
748 * Handle escapes *
749 *************************************************/
750
751 /* This function is called when a \ has been encountered. It either returns a
752 positive value for a simple escape such as \n, or 0 for a data character
753 which will be placed in chptr. A backreference to group n is returned as
754 negative n. When UTF-8 is enabled, a positive value greater than 255 may
755 be returned in chptr.
756 On entry,ptr is pointing at the \. On exit, it is on the final character of the
757 escape sequence.
758
759 Arguments:
760 ptrptr points to the pattern position pointer
761 chptr points to the data character
762 errorcodeptr points to the errorcode variable
763 bracount number of previous extracting brackets
764 options the options bits
765 isclass TRUE if inside a character class
766
767 Returns: zero => a data character
768 positive => a special escape sequence
769 negative => a back reference
770 on error, errorcodeptr is set
771 */
772
773 static int
774 check_escape(const pcre_uchar **ptrptr, pcre_uint32 *chptr, int *errorcodeptr,
775 int bracount, int options, BOOL isclass)
776 {
777 /* PCRE_UTF16 has the same value as PCRE_UTF8. */
778 BOOL utf = (options & PCRE_UTF8) != 0;
779 const pcre_uchar *ptr = *ptrptr + 1;
780 pcre_uint32 c;
781 int escape = 0;
782 int i;
783
784 GETCHARINCTEST(c, ptr); /* Get character value, increment pointer */
785 ptr--; /* Set pointer back to the last byte */
786
787 /* If backslash is at the end of the pattern, it's an error. */
788
789 if (c == 0) *errorcodeptr = ERR1;
790
791 /* Non-alphanumerics are literals. For digits or letters, do an initial lookup
792 in a table. A non-zero result is something that can be returned immediately.
793 Otherwise further processing may be required. */
794
795 #ifndef EBCDIC /* ASCII/UTF-8 coding */
796 /* Not alphanumeric */
797 else if (c < CHAR_0 || c > CHAR_z) {}
798 else if ((i = escapes[c - CHAR_0]) != 0) { if (i > 0) c = (pcre_uint32)i; else escape = -i; }
799
800 #else /* EBCDIC coding */
801 /* Not alphanumeric */
802 else if (c < CHAR_a || (!MAX_255(c) || (ebcdic_chartab[c] & 0x0E) == 0)) {}
803 else if ((i = escapes[c - 0x48]) != 0) { if (i > 0) c = (pcre_uint32)i; else escape = -i; }
804 #endif
805
806 /* Escapes that need further processing, or are illegal. */
807
808 else
809 {
810 const pcre_uchar *oldptr;
811 BOOL braced, negated, overflow;
812 int s;
813
814 switch (c)
815 {
816 /* A number of Perl escapes are not handled by PCRE. We give an explicit
817 error. */
818
819 case CHAR_l:
820 case CHAR_L:
821 *errorcodeptr = ERR37;
822 break;
823
824 case CHAR_u:
825 if ((options & PCRE_JAVASCRIPT_COMPAT) != 0)
826 {
827 /* In JavaScript, \u must be followed by four hexadecimal numbers.
828 Otherwise it is a lowercase u letter. */
829 if (MAX_255(ptr[1]) && (digitab[ptr[1]] & ctype_xdigit) != 0
830 && MAX_255(ptr[2]) && (digitab[ptr[2]] & ctype_xdigit) != 0
831 && MAX_255(ptr[3]) && (digitab[ptr[3]] & ctype_xdigit) != 0
832 && MAX_255(ptr[4]) && (digitab[ptr[4]] & ctype_xdigit) != 0)
833 {
834 c = 0;
835 for (i = 0; i < 4; ++i)
836 {
837 register pcre_uint32 cc = *(++ptr);
838 #ifndef EBCDIC /* ASCII/UTF-8 coding */
839 if (cc >= CHAR_a) cc -= 32; /* Convert to upper case */
840 c = (c << 4) + cc - ((cc < CHAR_A)? CHAR_0 : (CHAR_A - 10));
841 #else /* EBCDIC coding */
842 if (cc >= CHAR_a && cc <= CHAR_z) cc += 64; /* Convert to upper case */
843 c = (c << 4) + cc - ((cc >= CHAR_0)? CHAR_0 : (CHAR_A - 10));
844 #endif
845 }
846
847 #if defined COMPILE_PCRE8
848 if (c > (utf ? 0x10ffff : 0xff))
849 #elif defined COMPILE_PCRE16
850 if (c > (utf ? 0x10ffff : 0xffff))
851 #elif defined COMPILE_PCRE32
852 if (utf && c > 0x10ffff)
853 #endif
854 {
855 *errorcodeptr = ERR76;
856 }
857 else if (utf && c >= 0xd800 && c <= 0xdfff) *errorcodeptr = ERR73;
858 }
859 }
860 else
861 *errorcodeptr = ERR37;
862 break;
863
864 case CHAR_U:
865 /* In JavaScript, \U is an uppercase U letter. */
866 if ((options & PCRE_JAVASCRIPT_COMPAT) == 0) *errorcodeptr = ERR37;
867 break;
868
869 /* In a character class, \g is just a literal "g". Outside a character
870 class, \g must be followed by one of a number of specific things:
871
872 (1) A number, either plain or braced. If positive, it is an absolute
873 backreference. If negative, it is a relative backreference. This is a Perl
874 5.10 feature.
875
876 (2) Perl 5.10 also supports \g{name} as a reference to a named group. This
877 is part of Perl's movement towards a unified syntax for back references. As
878 this is synonymous with \k{name}, we fudge it up by pretending it really
879 was \k.
880
881 (3) For Oniguruma compatibility we also support \g followed by a name or a
882 number either in angle brackets or in single quotes. However, these are
883 (possibly recursive) subroutine calls, _not_ backreferences. Just return
884 the ESC_g code (cf \k). */
885
886 case CHAR_g:
887 if (isclass) break;
888 if (ptr[1] == CHAR_LESS_THAN_SIGN || ptr[1] == CHAR_APOSTROPHE)
889 {
890 escape = ESC_g;
891 break;
892 }
893
894 /* Handle the Perl-compatible cases */
895
896 if (ptr[1] == CHAR_LEFT_CURLY_BRACKET)
897 {
898 const pcre_uchar *p;
899 for (p = ptr+2; *p != 0 && *p != CHAR_RIGHT_CURLY_BRACKET; p++)
900 if (*p != CHAR_MINUS && !IS_DIGIT(*p)) break;
901 if (*p != 0 && *p != CHAR_RIGHT_CURLY_BRACKET)
902 {
903 escape = ESC_k;
904 break;
905 }
906 braced = TRUE;
907 ptr++;
908 }
909 else braced = FALSE;
910
911 if (ptr[1] == CHAR_MINUS)
912 {
913 negated = TRUE;
914 ptr++;
915 }
916 else negated = FALSE;
917
918 /* The integer range is limited by the machine's int representation. */
919 s = 0;
920 overflow = FALSE;
921 while (IS_DIGIT(ptr[1]))
922 {
923 if (s > INT_MAX / 10 - 1) /* Integer overflow */
924 {
925 overflow = TRUE;
926 break;
927 }
928 s = s * 10 + (int)(*(++ptr) - CHAR_0);
929 }
930 if (overflow) /* Integer overflow */
931 {
932 while (IS_DIGIT(ptr[1]))
933 ptr++;
934 *errorcodeptr = ERR61;
935 break;
936 }
937
938 if (braced && *(++ptr) != CHAR_RIGHT_CURLY_BRACKET)
939 {
940 *errorcodeptr = ERR57;
941 break;
942 }
943
944 if (s == 0)
945 {
946 *errorcodeptr = ERR58;
947 break;
948 }
949
950 if (negated)
951 {
952 if (s > bracount)
953 {
954 *errorcodeptr = ERR15;
955 break;
956 }
957 s = bracount - (s - 1);
958 }
959
960 escape = -s;
961 break;
962
963 /* The handling of escape sequences consisting of a string of digits
964 starting with one that is not zero is not straightforward. By experiment,
965 the way Perl works seems to be as follows:
966
967 Outside a character class, the digits are read as a decimal number. If the
968 number is less than 10, or if there are that many previous extracting
969 left brackets, then it is a back reference. Otherwise, up to three octal
970 digits are read to form an escaped byte. Thus \123 is likely to be octal
971 123 (cf \0123, which is octal 012 followed by the literal 3). If the octal
972 value is greater than 377, the least significant 8 bits are taken. Inside a
973 character class, \ followed by a digit is always an octal number. */
974
975 case CHAR_1: case CHAR_2: case CHAR_3: case CHAR_4: case CHAR_5:
976 case CHAR_6: case CHAR_7: case CHAR_8: case CHAR_9:
977
978 if (!isclass)
979 {
980 oldptr = ptr;
981 /* The integer range is limited by the machine's int representation. */
982 s = (int)(c -CHAR_0);
983 overflow = FALSE;
984 while (IS_DIGIT(ptr[1]))
985 {
986 if (s > INT_MAX / 10 - 1) /* Integer overflow */
987 {
988 overflow = TRUE;
989 break;
990 }
991 s = s * 10 + (int)(*(++ptr) - CHAR_0);
992 }
993 if (overflow) /* Integer overflow */
994 {
995 while (IS_DIGIT(ptr[1]))
996 ptr++;
997 *errorcodeptr = ERR61;
998 break;
999 }
1000 if (s < 10 || s <= bracount)
1001 {
1002 escape = -s;
1003 break;
1004 }
1005 ptr = oldptr; /* Put the pointer back and fall through */
1006 }
1007
1008 /* Handle an octal number following \. If the first digit is 8 or 9, Perl
1009 generates a binary zero byte and treats the digit as a following literal.
1010 Thus we have to pull back the pointer by one. */
1011
1012 if ((c = *ptr) >= CHAR_8)
1013 {
1014 ptr--;
1015 c = 0;
1016 break;
1017 }
1018
1019 /* \0 always starts an octal number, but we may drop through to here with a
1020 larger first octal digit. The original code used just to take the least
1021 significant 8 bits of octal numbers (I think this is what early Perls used
1022 to do). Nowadays we allow for larger numbers in UTF-8 mode and 16-bit mode,
1023 but no more than 3 octal digits. */
1024
1025 case CHAR_0:
1026 c -= CHAR_0;
1027 while(i++ < 2 && ptr[1] >= CHAR_0 && ptr[1] <= CHAR_7)
1028 c = c * 8 + *(++ptr) - CHAR_0;
1029 #ifdef COMPILE_PCRE8
1030 if (!utf && c > 0xff) *errorcodeptr = ERR51;
1031 #endif
1032 break;
1033
1034 /* \x is complicated. \x{ddd} is a character number which can be greater
1035 than 0xff in utf or non-8bit mode, but only if the ddd are hex digits.
1036 If not, { is treated as a data character. */
1037
1038 case CHAR_x:
1039 if ((options & PCRE_JAVASCRIPT_COMPAT) != 0)
1040 {
1041 /* In JavaScript, \x must be followed by two hexadecimal numbers.
1042 Otherwise it is a lowercase x letter. */
1043 if (MAX_255(ptr[1]) && (digitab[ptr[1]] & ctype_xdigit) != 0
1044 && MAX_255(ptr[2]) && (digitab[ptr[2]] & ctype_xdigit) != 0)
1045 {
1046 c = 0;
1047 for (i = 0; i < 2; ++i)
1048 {
1049 register pcre_uint32 cc = *(++ptr);
1050 #ifndef EBCDIC /* ASCII/UTF-8 coding */
1051 if (cc >= CHAR_a) cc -= 32; /* Convert to upper case */
1052 c = (c << 4) + cc - ((cc < CHAR_A)? CHAR_0 : (CHAR_A - 10));
1053 #else /* EBCDIC coding */
1054 if (cc >= CHAR_a && cc <= CHAR_z) cc += 64; /* Convert to upper case */
1055 c = (c << 4) + cc - ((cc >= CHAR_0)? CHAR_0 : (CHAR_A - 10));
1056 #endif
1057 }
1058 }
1059 break;
1060 }
1061
1062 if (ptr[1] == CHAR_LEFT_CURLY_BRACKET)
1063 {
1064 const pcre_uchar *pt = ptr + 2;
1065
1066 c = 0;
1067 overflow = FALSE;
1068 while (MAX_255(*pt) && (digitab[*pt] & ctype_xdigit) != 0)
1069 {
1070 register pcre_uint32 cc = *pt++;
1071 if (c == 0 && cc == CHAR_0) continue; /* Leading zeroes */
1072
1073 #ifdef COMPILE_PCRE32
1074 if (c >= 0x10000000l) { overflow = TRUE; break; }
1075 #endif
1076
1077 #ifndef EBCDIC /* ASCII/UTF-8 coding */
1078 if (cc >= CHAR_a) cc -= 32; /* Convert to upper case */
1079 c = (c << 4) + cc - ((cc < CHAR_A)? CHAR_0 : (CHAR_A - 10));
1080 #else /* EBCDIC coding */
1081 if (cc >= CHAR_a && cc <= CHAR_z) cc += 64; /* Convert to upper case */
1082 c = (c << 4) + cc - ((cc >= CHAR_0)? CHAR_0 : (CHAR_A - 10));
1083 #endif
1084
1085 #if defined COMPILE_PCRE8
1086 if (c > (utf ? 0x10ffff : 0xff)) { overflow = TRUE; break; }
1087 #elif defined COMPILE_PCRE16
1088 if (c > (utf ? 0x10ffff : 0xffff)) { overflow = TRUE; break; }
1089 #elif defined COMPILE_PCRE32
1090 if (utf && c > 0x10ffff) { overflow = TRUE; break; }
1091 #endif
1092 }
1093
1094 if (overflow)
1095 {
1096 while (MAX_255(*pt) && (digitab[*pt] & ctype_xdigit) != 0) pt++;
1097 *errorcodeptr = ERR34;
1098 }
1099
1100 if (*pt == CHAR_RIGHT_CURLY_BRACKET)
1101 {
1102 if (utf && c >= 0xd800 && c <= 0xdfff) *errorcodeptr = ERR73;
1103 ptr = pt;
1104 break;
1105 }
1106
1107 /* If the sequence of hex digits does not end with '}', then we don't
1108 recognize this construct; fall through to the normal \x handling. */
1109 }
1110
1111 /* Read just a single-byte hex-defined char */
1112
1113 c = 0;
1114 while (i++ < 2 && MAX_255(ptr[1]) && (digitab[ptr[1]] & ctype_xdigit) != 0)
1115 {
1116 pcre_uint32 cc; /* Some compilers don't like */
1117 cc = *(++ptr); /* ++ in initializers */
1118 #ifndef EBCDIC /* ASCII/UTF-8 coding */
1119 if (cc >= CHAR_a) cc -= 32; /* Convert to upper case */
1120 c = c * 16 + cc - ((cc < CHAR_A)? CHAR_0 : (CHAR_A - 10));
1121 #else /* EBCDIC coding */
1122 if (cc <= CHAR_z) cc += 64; /* Convert to upper case */
1123 c = c * 16 + cc - ((cc >= CHAR_0)? CHAR_0 : (CHAR_A - 10));
1124 #endif
1125 }
1126 break;
1127
1128 /* For \c, a following letter is upper-cased; then the 0x40 bit is flipped.
1129 An error is given if the byte following \c is not an ASCII character. This
1130 coding is ASCII-specific, but then the whole concept of \cx is
1131 ASCII-specific. (However, an EBCDIC equivalent has now been added.) */
1132
1133 case CHAR_c:
1134 c = *(++ptr);
1135 if (c == 0)
1136 {
1137 *errorcodeptr = ERR2;
1138 break;
1139 }
1140 #ifndef EBCDIC /* ASCII/UTF-8 coding */
1141 if (c > 127) /* Excludes all non-ASCII in either mode */
1142 {
1143 *errorcodeptr = ERR68;
1144 break;
1145 }
1146 if (c >= CHAR_a && c <= CHAR_z) c -= 32;
1147 c ^= 0x40;
1148 #else /* EBCDIC coding */
1149 if (c >= CHAR_a && c <= CHAR_z) c += 64;
1150 c ^= 0xC0;
1151 #endif
1152 break;
1153
1154 /* PCRE_EXTRA enables extensions to Perl in the matter of escapes. Any
1155 other alphanumeric following \ is an error if PCRE_EXTRA was set;
1156 otherwise, for Perl compatibility, it is a literal. This code looks a bit
1157 odd, but there used to be some cases other than the default, and there may
1158 be again in future, so I haven't "optimized" it. */
1159
1160 default:
1161 if ((options & PCRE_EXTRA) != 0) switch(c)
1162 {
1163 default:
1164 *errorcodeptr = ERR3;
1165 break;
1166 }
1167 break;
1168 }
1169 }
1170
1171 /* Perl supports \N{name} for character names, as well as plain \N for "not
1172 newline". PCRE does not support \N{name}. However, it does support
1173 quantification such as \N{2,3}. */
1174
1175 if (escape == ESC_N && ptr[1] == CHAR_LEFT_CURLY_BRACKET &&
1176 !is_counted_repeat(ptr+2))
1177 *errorcodeptr = ERR37;
1178
1179 /* If PCRE_UCP is set, we change the values for \d etc. */
1180
1181 if ((options & PCRE_UCP) != 0 && escape >= ESC_D && escape <= ESC_w)
1182 escape += (ESC_DU - ESC_D);
1183
1184 /* Set the pointer to the final character before returning. */
1185
1186 *ptrptr = ptr;
1187 *chptr = c;
1188 return escape;
1189 }
1190
1191 #ifdef SUPPORT_UCP
1192 /*************************************************
1193 * Handle \P and \p *
1194 *************************************************/
1195
1196 /* This function is called after \P or \p has been encountered, provided that
1197 PCRE is compiled with support for Unicode properties. On entry, ptrptr is
1198 pointing at the P or p. On exit, it is pointing at the final character of the
1199 escape sequence.
1200
1201 Argument:
1202 ptrptr points to the pattern position pointer
1203 negptr points to a boolean that is set TRUE for negation else FALSE
1204 dptr points to an int that is set to the detailed property value
1205 errorcodeptr points to the error code variable
1206
1207 Returns: type value from ucp_type_table, or -1 for an invalid type
1208 */
1209
1210 static int
1211 get_ucp(const pcre_uchar **ptrptr, BOOL *negptr, int *dptr, int *errorcodeptr)
1212 {
1213 pcre_uchar c;
1214 int i, bot, top;
1215 const pcre_uchar *ptr = *ptrptr;
1216 pcre_uchar name[32];
1217
1218 c = *(++ptr);
1219 if (c == 0) goto ERROR_RETURN;
1220
1221 *negptr = FALSE;
1222
1223 /* \P or \p can be followed by a name in {}, optionally preceded by ^ for
1224 negation. */
1225
1226 if (c == CHAR_LEFT_CURLY_BRACKET)
1227 {
1228 if (ptr[1] == CHAR_CIRCUMFLEX_ACCENT)
1229 {
1230 *negptr = TRUE;
1231 ptr++;
1232 }
1233 for (i = 0; i < (int)(sizeof(name) / sizeof(pcre_uchar)) - 1; i++)
1234 {
1235 c = *(++ptr);
1236 if (c == 0) goto ERROR_RETURN;
1237 if (c == CHAR_RIGHT_CURLY_BRACKET) break;
1238 name[i] = c;
1239 }
1240 if (c != CHAR_RIGHT_CURLY_BRACKET) goto ERROR_RETURN;
1241 name[i] = 0;
1242 }
1243
1244 /* Otherwise there is just one following character */
1245
1246 else
1247 {
1248 name[0] = c;
1249 name[1] = 0;
1250 }
1251
1252 *ptrptr = ptr;
1253
1254 /* Search for a recognized property name using binary chop */
1255
1256 bot = 0;
1257 top = PRIV(utt_size);
1258
1259 while (bot < top)
1260 {
1261 int r;
1262 i = (bot + top) >> 1;
1263 r = STRCMP_UC_C8(name, PRIV(utt_names) + PRIV(utt)[i].name_offset);
1264 if (r == 0)
1265 {
1266 *dptr = PRIV(utt)[i].value;
1267 return PRIV(utt)[i].type;
1268 }
1269 if (r > 0) bot = i + 1; else top = i;
1270 }
1271
1272 *errorcodeptr = ERR47;
1273 *ptrptr = ptr;
1274 return -1;
1275
1276 ERROR_RETURN:
1277 *errorcodeptr = ERR46;
1278 *ptrptr = ptr;
1279 return -1;
1280 }
1281 #endif
1282
1283
1284
1285
1286 /*************************************************
1287 * Read repeat counts *
1288 *************************************************/
1289
1290 /* Read an item of the form {n,m} and return the values. This is called only
1291 after is_counted_repeat() has confirmed that a repeat-count quantifier exists,
1292 so the syntax is guaranteed to be correct, but we need to check the values.
1293
1294 Arguments:
1295 p pointer to first char after '{'
1296 minp pointer to int for min
1297 maxp pointer to int for max
1298 returned as -1 if no max
1299 errorcodeptr points to error code variable
1300
1301 Returns: pointer to '}' on success;
1302 current ptr on error, with errorcodeptr set non-zero
1303 */
1304
1305 static const pcre_uchar *
1306 read_repeat_counts(const pcre_uchar *p, int *minp, int *maxp, int *errorcodeptr)
1307 {
1308 int min = 0;
1309 int max = -1;
1310
1311 /* Read the minimum value and do a paranoid check: a negative value indicates
1312 an integer overflow. */
1313
1314 while (IS_DIGIT(*p)) min = min * 10 + (int)(*p++ - CHAR_0);
1315 if (min < 0 || min > 65535)
1316 {
1317 *errorcodeptr = ERR5;
1318 return p;
1319 }
1320
1321 /* Read the maximum value if there is one, and again do a paranoid on its size.
1322 Also, max must not be less than min. */
1323
1324 if (*p == CHAR_RIGHT_CURLY_BRACKET) max = min; else
1325 {
1326 if (*(++p) != CHAR_RIGHT_CURLY_BRACKET)
1327 {
1328 max = 0;
1329 while(IS_DIGIT(*p)) max = max * 10 + (int)(*p++ - CHAR_0);
1330 if (max < 0 || max > 65535)
1331 {
1332 *errorcodeptr = ERR5;
1333 return p;
1334 }
1335 if (max < min)
1336 {
1337 *errorcodeptr = ERR4;
1338 return p;
1339 }
1340 }
1341 }
1342
1343 /* Fill in the required variables, and pass back the pointer to the terminating
1344 '}'. */
1345
1346 *minp = min;
1347 *maxp = max;
1348 return p;
1349 }
1350
1351
1352
1353 /*************************************************
1354 * Subroutine for finding forward reference *
1355 *************************************************/
1356
1357 /* This recursive function is called only from find_parens() below. The
1358 top-level call starts at the beginning of the pattern. All other calls must
1359 start at a parenthesis. It scans along a pattern's text looking for capturing
1360 subpatterns, and counting them. If it finds a named pattern that matches the
1361 name it is given, it returns its number. Alternatively, if the name is NULL, it
1362 returns when it reaches a given numbered subpattern. Recursion is used to keep
1363 track of subpatterns that reset the capturing group numbers - the (?| feature.
1364
1365 This function was originally called only from the second pass, in which we know
1366 that if (?< or (?' or (?P< is encountered, the name will be correctly
1367 terminated because that is checked in the first pass. There is now one call to
1368 this function in the first pass, to check for a recursive back reference by
1369 name (so that we can make the whole group atomic). In this case, we need check
1370 only up to the current position in the pattern, and that is still OK because
1371 and previous occurrences will have been checked. To make this work, the test
1372 for "end of pattern" is a check against cd->end_pattern in the main loop,
1373 instead of looking for a binary zero. This means that the special first-pass
1374 call can adjust cd->end_pattern temporarily. (Checks for binary zero while
1375 processing items within the loop are OK, because afterwards the main loop will
1376 terminate.)
1377
1378 Arguments:
1379 ptrptr address of the current character pointer (updated)
1380 cd compile background data
1381 name name to seek, or NULL if seeking a numbered subpattern
1382 lorn name length, or subpattern number if name is NULL
1383 xmode TRUE if we are in /x mode
1384 utf TRUE if we are in UTF-8 / UTF-16 / UTF-32 mode
1385 count pointer to the current capturing subpattern number (updated)
1386
1387 Returns: the number of the named subpattern, or -1 if not found
1388 */
1389
1390 static int
1391 find_parens_sub(pcre_uchar **ptrptr, compile_data *cd, const pcre_uchar *name, int lorn,
1392 BOOL xmode, BOOL utf, int *count)
1393 {
1394 pcre_uchar *ptr = *ptrptr;
1395 int start_count = *count;
1396 int hwm_count = start_count;
1397 BOOL dup_parens = FALSE;
1398
1399 /* If the first character is a parenthesis, check on the type of group we are
1400 dealing with. The very first call may not start with a parenthesis. */
1401
1402 if (ptr[0] == CHAR_LEFT_PARENTHESIS)
1403 {
1404 /* Handle specials such as (*SKIP) or (*UTF8) etc. */
1405
1406 if (ptr[1] == CHAR_ASTERISK) ptr += 2;
1407
1408 /* Handle a normal, unnamed capturing parenthesis. */
1409
1410 else if (ptr[1] != CHAR_QUESTION_MARK)
1411 {
1412 *count += 1;
1413 if (name == NULL && *count == lorn) return *count;
1414 ptr++;
1415 }
1416
1417 /* All cases now have (? at the start. Remember when we are in a group
1418 where the parenthesis numbers are duplicated. */
1419
1420 else if (ptr[2] == CHAR_VERTICAL_LINE)
1421 {
1422 ptr += 3;
1423 dup_parens = TRUE;
1424 }
1425
1426 /* Handle comments; all characters are allowed until a ket is reached. */
1427
1428 else if (ptr[2] == CHAR_NUMBER_SIGN)
1429 {
1430 for (ptr += 3; *ptr != 0; ptr++) if (*ptr == CHAR_RIGHT_PARENTHESIS) break;
1431 goto FAIL_EXIT;
1432 }
1433
1434 /* Handle a condition. If it is an assertion, just carry on so that it
1435 is processed as normal. If not, skip to the closing parenthesis of the
1436 condition (there can't be any nested parens). */
1437
1438 else if (ptr[2] == CHAR_LEFT_PARENTHESIS)
1439 {
1440 ptr += 2;
1441 if (ptr[1] != CHAR_QUESTION_MARK)
1442 {
1443 while (*ptr != 0 && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++;
1444 if (*ptr != 0) ptr++;
1445 }
1446 }
1447
1448 /* Start with (? but not a condition. */
1449
1450 else
1451 {
1452 ptr += 2;
1453 if (*ptr == CHAR_P) ptr++; /* Allow optional P */
1454
1455 /* We have to disambiguate (?<! and (?<= from (?<name> for named groups */
1456
1457 if ((*ptr == CHAR_LESS_THAN_SIGN && ptr[1] != CHAR_EXCLAMATION_MARK &&
1458 ptr[1] != CHAR_EQUALS_SIGN) || *ptr == CHAR_APOSTROPHE)
1459 {
1460 int term;
1461 const pcre_uchar *thisname;
1462 *count += 1;
1463 if (name == NULL && *count == lorn) return *count;
1464 term = *ptr++;
1465 if (term == CHAR_LESS_THAN_SIGN) term = CHAR_GREATER_THAN_SIGN;
1466 thisname = ptr;
1467 while (*ptr != term) ptr++;
1468 if (name != NULL && lorn == ptr - thisname &&
1469 STRNCMP_UC_UC(name, thisname, lorn) == 0)
1470 return *count;
1471 term++;
1472 }
1473 }
1474 }
1475
1476 /* Past any initial parenthesis handling, scan for parentheses or vertical
1477 bars. Stop if we get to cd->end_pattern. Note that this is important for the
1478 first-pass call when this value is temporarily adjusted to stop at the current
1479 position. So DO NOT change this to a test for binary zero. */
1480
1481 for (; ptr < cd->end_pattern; ptr++)
1482 {
1483 /* Skip over backslashed characters and also entire \Q...\E */
1484
1485 if (*ptr == CHAR_BACKSLASH)
1486 {
1487 if (*(++ptr) == 0) goto FAIL_EXIT;
1488 if (*ptr == CHAR_Q) for (;;)
1489 {
1490 while (*(++ptr) != 0 && *ptr != CHAR_BACKSLASH) {};
1491 if (*ptr == 0) goto FAIL_EXIT;
1492 if (*(++ptr) == CHAR_E) break;
1493 }
1494 continue;
1495 }
1496
1497 /* Skip over character classes; this logic must be similar to the way they
1498 are handled for real. If the first character is '^', skip it. Also, if the
1499 first few characters (either before or after ^) are \Q\E or \E we skip them
1500 too. This makes for compatibility with Perl. Note the use of STR macros to
1501 encode "Q\\E" so that it works in UTF-8 on EBCDIC platforms. */
1502
1503 if (*ptr == CHAR_LEFT_SQUARE_BRACKET)
1504 {
1505 BOOL negate_class = FALSE;
1506 for (;;)
1507 {
1508 if (ptr[1] == CHAR_BACKSLASH)
1509 {
1510 if (ptr[2] == CHAR_E)
1511 ptr+= 2;
1512 else if (STRNCMP_UC_C8(ptr + 2,
1513 STR_Q STR_BACKSLASH STR_E, 3) == 0)
1514 ptr += 4;
1515 else
1516 break;
1517 }
1518 else if (!negate_class && ptr[1] == CHAR_CIRCUMFLEX_ACCENT)
1519 {
1520 negate_class = TRUE;
1521 ptr++;
1522 }
1523 else break;
1524 }
1525
1526 /* If the next character is ']', it is a data character that must be
1527 skipped, except in JavaScript compatibility mode. */
1528
1529 if (ptr[1] == CHAR_RIGHT_SQUARE_BRACKET &&
1530 (cd->external_options & PCRE_JAVASCRIPT_COMPAT) == 0)
1531 ptr++;
1532
1533 while (*(++ptr) != CHAR_RIGHT_SQUARE_BRACKET)
1534 {
1535 if (*ptr == 0) return -1;
1536 if (*ptr == CHAR_BACKSLASH)
1537 {
1538 if (*(++ptr) == 0) goto FAIL_EXIT;
1539 if (*ptr == CHAR_Q) for (;;)
1540 {
1541 while (*(++ptr) != 0 && *ptr != CHAR_BACKSLASH) {};
1542 if (*ptr == 0) goto FAIL_EXIT;
1543 if (*(++ptr) == CHAR_E) break;
1544 }
1545 continue;
1546 }
1547 }
1548 continue;
1549 }
1550
1551 /* Skip comments in /x mode */
1552
1553 if (xmode && *ptr == CHAR_NUMBER_SIGN)
1554 {
1555 ptr++;
1556 while (*ptr != 0)
1557 {
1558 if (IS_NEWLINE(ptr)) { ptr += cd->nllen - 1; break; }
1559 ptr++;
1560 #ifdef SUPPORT_UTF
1561 if (utf) FORWARDCHAR(ptr);
1562 #endif
1563 }
1564 if (*ptr == 0) goto FAIL_EXIT;
1565 continue;
1566 }
1567
1568 /* Check for the special metacharacters */
1569
1570 if (*ptr == CHAR_LEFT_PARENTHESIS)
1571 {
1572 int rc = find_parens_sub(&ptr, cd, name, lorn, xmode, utf, count);
1573 if (rc > 0) return rc;
1574 if (*ptr == 0) goto FAIL_EXIT;
1575 }
1576
1577 else if (*ptr == CHAR_RIGHT_PARENTHESIS)
1578 {
1579 if (dup_parens && *count < hwm_count) *count = hwm_count;
1580 goto FAIL_EXIT;
1581 }
1582
1583 else if (*ptr == CHAR_VERTICAL_LINE && dup_parens)
1584 {
1585 if (*count > hwm_count) hwm_count = *count;
1586 *count = start_count;
1587 }
1588 }
1589
1590 FAIL_EXIT:
1591 *ptrptr = ptr;
1592 return -1;
1593 }
1594
1595
1596
1597
1598 /*************************************************
1599 * Find forward referenced subpattern *
1600 *************************************************/
1601
1602 /* This function scans along a pattern's text looking for capturing
1603 subpatterns, and counting them. If it finds a named pattern that matches the
1604 name it is given, it returns its number. Alternatively, if the name is NULL, it
1605 returns when it reaches a given numbered subpattern. This is used for forward
1606 references to subpatterns. We used to be able to start this scan from the
1607 current compiling point, using the current count value from cd->bracount, and
1608 do it all in a single loop, but the addition of the possibility of duplicate
1609 subpattern numbers means that we have to scan from the very start, in order to
1610 take account of such duplicates, and to use a recursive function to keep track
1611 of the different types of group.
1612
1613 Arguments:
1614 cd compile background data
1615 name name to seek, or NULL if seeking a numbered subpattern
1616 lorn name length, or subpattern number if name is NULL
1617 xmode TRUE if we are in /x mode
1618 utf TRUE if we are in UTF-8 / UTF-16 / UTF-32 mode
1619
1620 Returns: the number of the found subpattern, or -1 if not found
1621 */
1622
1623 static int
1624 find_parens(compile_data *cd, const pcre_uchar *name, int lorn, BOOL xmode,
1625 BOOL utf)
1626 {
1627 pcre_uchar *ptr = (pcre_uchar *)cd->start_pattern;
1628 int count = 0;
1629 int rc;
1630
1631 /* If the pattern does not start with an opening parenthesis, the first call
1632 to find_parens_sub() will scan right to the end (if necessary). However, if it
1633 does start with a parenthesis, find_parens_sub() will return when it hits the
1634 matching closing parens. That is why we have to have a loop. */
1635
1636 for (;;)
1637 {
1638 rc = find_parens_sub(&ptr, cd, name, lorn, xmode, utf, &count);
1639 if (rc > 0 || *ptr++ == 0) break;
1640 }
1641
1642 return rc;
1643 }
1644
1645
1646
1647
1648 /*************************************************
1649 * Find first significant op code *
1650 *************************************************/
1651
1652 /* This is called by several functions that scan a compiled expression looking
1653 for a fixed first character, or an anchoring op code etc. It skips over things
1654 that do not influence this. For some calls, it makes sense to skip negative
1655 forward and all backward assertions, and also the \b assertion; for others it
1656 does not.
1657
1658 Arguments:
1659 code pointer to the start of the group
1660 skipassert TRUE if certain assertions are to be skipped
1661
1662 Returns: pointer to the first significant opcode
1663 */
1664
1665 static const pcre_uchar*
1666 first_significant_code(const pcre_uchar *code, BOOL skipassert)
1667 {
1668 for (;;)
1669 {
1670 switch ((int)*code)
1671 {
1672 case OP_ASSERT_NOT:
1673 case OP_ASSERTBACK:
1674 case OP_ASSERTBACK_NOT:
1675 if (!skipassert) return code;
1676 do code += GET(code, 1); while (*code == OP_ALT);
1677 code += PRIV(OP_lengths)[*code];
1678 break;
1679
1680 case OP_WORD_BOUNDARY:
1681 case OP_NOT_WORD_BOUNDARY:
1682 if (!skipassert) return code;
1683 /* Fall through */
1684
1685 case OP_CALLOUT:
1686 case OP_CREF:
1687 case OP_NCREF:
1688 case OP_RREF:
1689 case OP_NRREF:
1690 case OP_DEF:
1691 code += PRIV(OP_lengths)[*code];
1692 break;
1693
1694 default:
1695 return code;
1696 }
1697 }
1698 /* Control never reaches here */
1699 }
1700
1701
1702
1703
1704 /*************************************************
1705 * Find the fixed length of a branch *
1706 *************************************************/
1707
1708 /* Scan a branch and compute the fixed length of subject that will match it,
1709 if the length is fixed. This is needed for dealing with backward assertions.
1710 In UTF8 mode, the result is in characters rather than bytes. The branch is
1711 temporarily terminated with OP_END when this function is called.
1712
1713 This function is called when a backward assertion is encountered, so that if it
1714 fails, the error message can point to the correct place in the pattern.
1715 However, we cannot do this when the assertion contains subroutine calls,
1716 because they can be forward references. We solve this by remembering this case
1717 and doing the check at the end; a flag specifies which mode we are running in.
1718
1719 Arguments:
1720 code points to the start of the pattern (the bracket)
1721 utf TRUE in UTF-8 / UTF-16 / UTF-32 mode
1722 atend TRUE if called when the pattern is complete
1723 cd the "compile data" structure
1724
1725 Returns: the fixed length,
1726 or -1 if there is no fixed length,
1727 or -2 if \C was encountered (in UTF-8 mode only)
1728 or -3 if an OP_RECURSE item was encountered and atend is FALSE
1729 or -4 if an unknown opcode was encountered (internal error)
1730 */
1731
1732 static int
1733 find_fixedlength(pcre_uchar *code, BOOL utf, BOOL atend, compile_data *cd)
1734 {
1735 int length = -1;
1736
1737 register int branchlength = 0;
1738 register pcre_uchar *cc = code + 1 + LINK_SIZE;
1739
1740 /* Scan along the opcodes for this branch. If we get to the end of the
1741 branch, check the length against that of the other branches. */
1742
1743 for (;;)
1744 {
1745 int d;
1746 pcre_uchar *ce, *cs;
1747 register int op = *cc;
1748
1749 switch (op)
1750 {
1751 /* We only need to continue for OP_CBRA (normal capturing bracket) and
1752 OP_BRA (normal non-capturing bracket) because the other variants of these
1753 opcodes are all concerned with unlimited repeated groups, which of course
1754 are not of fixed length. */
1755
1756 case OP_CBRA:
1757 case OP_BRA:
1758 case OP_ONCE:
1759 case OP_ONCE_NC:
1760 case OP_COND:
1761 d = find_fixedlength(cc + ((op == OP_CBRA)? IMM2_SIZE : 0), utf, atend, cd);
1762 if (d < 0) return d;
1763 branchlength += d;
1764 do cc += GET(cc, 1); while (*cc == OP_ALT);
1765 cc += 1 + LINK_SIZE;
1766 break;
1767
1768 /* Reached end of a branch; if it's a ket it is the end of a nested call.
1769 If it's ALT it is an alternation in a nested call. An ACCEPT is effectively
1770 an ALT. If it is END it's the end of the outer call. All can be handled by
1771 the same code. Note that we must not include the OP_KETRxxx opcodes here,
1772 because they all imply an unlimited repeat. */
1773
1774 case OP_ALT:
1775 case OP_KET:
1776 case OP_END:
1777 case OP_ACCEPT:
1778 case OP_ASSERT_ACCEPT:
1779 if (length < 0) length = branchlength;
1780 else if (length != branchlength) return -1;
1781 if (*cc != OP_ALT) return length;
1782 cc += 1 + LINK_SIZE;
1783 branchlength = 0;
1784 break;
1785
1786 /* A true recursion implies not fixed length, but a subroutine call may
1787 be OK. If the subroutine is a forward reference, we can't deal with
1788 it until the end of the pattern, so return -3. */
1789
1790 case OP_RECURSE:
1791 if (!atend) return -3;
1792 cs = ce = (pcre_uchar *)cd->start_code + GET(cc, 1); /* Start subpattern */
1793 do ce += GET(ce, 1); while (*ce == OP_ALT); /* End subpattern */
1794 if (cc > cs && cc < ce) return -1; /* Recursion */
1795 d = find_fixedlength(cs + IMM2_SIZE, utf, atend, cd);
1796 if (d < 0) return d;
1797 branchlength += d;
1798 cc += 1 + LINK_SIZE;
1799 break;
1800
1801 /* Skip over assertive subpatterns */
1802
1803 case OP_ASSERT:
1804 case OP_ASSERT_NOT:
1805 case OP_ASSERTBACK:
1806 case OP_ASSERTBACK_NOT:
1807 do cc += GET(cc, 1); while (*cc == OP_ALT);
1808 cc += PRIV(OP_lengths)[*cc];
1809 break;
1810
1811 /* Skip over things that don't match chars */
1812
1813 case OP_MARK:
1814 case OP_PRUNE_ARG:
1815 case OP_SKIP_ARG:
1816 case OP_THEN_ARG:
1817 cc += cc[1] + PRIV(OP_lengths)[*cc];
1818 break;
1819
1820 case OP_CALLOUT:
1821 case OP_CIRC:
1822 case OP_CIRCM:
1823 case OP_CLOSE:
1824 case OP_COMMIT:
1825 case OP_CREF:
1826 case OP_DEF:
1827 case OP_DOLL:
1828 case OP_DOLLM:
1829 case OP_EOD:
1830 case OP_EODN:
1831 case OP_FAIL:
1832 case OP_NCREF:
1833 case OP_NRREF:
1834 case OP_NOT_WORD_BOUNDARY:
1835 case OP_PRUNE:
1836 case OP_REVERSE:
1837 case OP_RREF:
1838 case OP_SET_SOM:
1839 case OP_SKIP:
1840 case OP_SOD:
1841 case OP_SOM:
1842 case OP_THEN:
1843 case OP_WORD_BOUNDARY:
1844 cc += PRIV(OP_lengths)[*cc];
1845 break;
1846
1847 /* Handle literal characters */
1848
1849 case OP_CHAR:
1850 case OP_CHARI:
1851 case OP_NOT:
1852 case OP_NOTI:
1853 branchlength++;
1854 cc += 2;
1855 #if defined SUPPORT_UTF && !defined COMPILE_PCRE32
1856 if (utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
1857 #endif
1858 break;
1859
1860 /* Handle exact repetitions. The count is already in characters, but we
1861 need to skip over a multibyte character in UTF8 mode. */
1862
1863 case OP_EXACT:
1864 case OP_EXACTI:
1865 case OP_NOTEXACT:
1866 case OP_NOTEXACTI:
1867 branchlength += GET2(cc,1);
1868 cc += 2 + IMM2_SIZE;
1869 #if defined SUPPORT_UTF && !defined COMPILE_PCRE32
1870 if (utf && HAS_EXTRALEN(cc[-1])) cc += GET_EXTRALEN(cc[-1]);
1871 #endif
1872 break;
1873
1874 case OP_TYPEEXACT:
1875 branchlength += GET2(cc,1);
1876 if (cc[1 + IMM2_SIZE] == OP_PROP || cc[1 + IMM2_SIZE] == OP_NOTPROP)
1877 cc += 2;
1878 cc += 1 + IMM2_SIZE + 1;
1879 break;
1880
1881 /* Handle single-char matchers */
1882
1883 case OP_PROP:
1884 case OP_NOTPROP:
1885 cc += 2;
1886 /* Fall through */
1887
1888 case OP_HSPACE:
1889 case OP_VSPACE:
1890 case OP_NOT_HSPACE:
1891 case OP_NOT_VSPACE:
1892 case OP_NOT_DIGIT:
1893 case OP_DIGIT:
1894 case OP_NOT_WHITESPACE:
1895 case OP_WHITESPACE:
1896 case OP_NOT_WORDCHAR:
1897 case OP_WORDCHAR:
1898 case OP_ANY:
1899 case OP_ALLANY:
1900 branchlength++;
1901 cc++;
1902 break;
1903
1904 /* The single-byte matcher isn't allowed. This only happens in UTF-8 mode;
1905 otherwise \C is coded as OP_ALLANY. */
1906
1907 case OP_ANYBYTE:
1908 return -2;
1909
1910 /* Check a class for variable quantification */
1911
1912 #if defined SUPPORT_UTF || defined COMPILE_PCRE16 || defined COMPILE_PCRE32
1913 case OP_XCLASS:
1914 cc += GET(cc, 1) - PRIV(OP_lengths)[OP_CLASS];
1915 /* Fall through */
1916 #endif
1917
1918 case OP_CLASS:
1919 case OP_NCLASS:
1920 cc += PRIV(OP_lengths)[OP_CLASS];
1921
1922 switch (*cc)
1923 {
1924 case OP_CRPLUS:
1925 case OP_CRMINPLUS:
1926 case OP_CRSTAR:
1927 case OP_CRMINSTAR:
1928 case OP_CRQUERY:
1929 case OP_CRMINQUERY:
1930 return -1;
1931
1932 case OP_CRRANGE:
1933 case OP_CRMINRANGE:
1934 if (GET2(cc,1) != GET2(cc,1+IMM2_SIZE)) return -1;
1935 branchlength += GET2(cc,1);
1936 cc += 1 + 2 * IMM2_SIZE;
1937 break;
1938
1939 default:
1940 branchlength++;
1941 }
1942 break;
1943
1944 /* Anything else is variable length */
1945
1946 case OP_ANYNL:
1947 case OP_BRAMINZERO:
1948 case OP_BRAPOS:
1949 case OP_BRAPOSZERO:
1950 case OP_BRAZERO:
1951 case OP_CBRAPOS:
1952 case OP_EXTUNI:
1953 case OP_KETRMAX:
1954 case OP_KETRMIN:
1955 case OP_KETRPOS:
1956 case OP_MINPLUS:
1957 case OP_MINPLUSI:
1958 case OP_MINQUERY:
1959 case OP_MINQUERYI:
1960 case OP_MINSTAR:
1961 case OP_MINSTARI:
1962 case OP_MINUPTO:
1963 case OP_MINUPTOI:
1964 case OP_NOTMINPLUS:
1965 case OP_NOTMINPLUSI:
1966 case OP_NOTMINQUERY:
1967 case OP_NOTMINQUERYI:
1968 case OP_NOTMINSTAR:
1969 case OP_NOTMINSTARI:
1970 case OP_NOTMINUPTO:
1971 case OP_NOTMINUPTOI:
1972 case OP_NOTPLUS:
1973 case OP_NOTPLUSI:
1974 case OP_NOTPOSPLUS:
1975 case OP_NOTPOSPLUSI:
1976 case OP_NOTPOSQUERY:
1977 case OP_NOTPOSQUERYI:
1978 case OP_NOTPOSSTAR:
1979 case OP_NOTPOSSTARI:
1980 case OP_NOTPOSUPTO:
1981 case OP_NOTPOSUPTOI:
1982 case OP_NOTQUERY:
1983 case OP_NOTQUERYI:
1984 case OP_NOTSTAR:
1985 case OP_NOTSTARI:
1986 case OP_NOTUPTO:
1987 case OP_NOTUPTOI:
1988 case OP_PLUS:
1989 case OP_PLUSI:
1990 case OP_POSPLUS:
1991 case OP_POSPLUSI:
1992 case OP_POSQUERY:
1993 case OP_POSQUERYI:
1994 case OP_POSSTAR:
1995 case OP_POSSTARI:
1996 case OP_POSUPTO:
1997 case OP_POSUPTOI:
1998 case OP_QUERY:
1999 case OP_QUERYI:
2000 case OP_REF:
2001 case OP_REFI:
2002 case OP_SBRA:
2003 case OP_SBRAPOS:
2004 case OP_SCBRA:
2005 case OP_SCBRAPOS:
2006 case OP_SCOND:
2007 case OP_SKIPZERO:
2008 case OP_STAR:
2009 case OP_STARI:
2010 case OP_TYPEMINPLUS:
2011 case OP_TYPEMINQUERY:
2012 case OP_TYPEMINSTAR:
2013 case OP_TYPEMINUPTO:
2014 case OP_TYPEPLUS:
2015 case OP_TYPEPOSPLUS:
2016 case OP_TYPEPOSQUERY:
2017 case OP_TYPEPOSSTAR:
2018 case OP_TYPEPOSUPTO:
2019 case OP_TYPEQUERY:
2020 case OP_TYPESTAR:
2021 case OP_TYPEUPTO:
2022 case OP_UPTO:
2023 case OP_UPTOI:
2024 return -1;
2025
2026 /* Catch unrecognized opcodes so that when new ones are added they
2027 are not forgotten, as has happened in the past. */
2028
2029 default:
2030 return -4;
2031 }
2032 }
2033 /* Control never gets here */
2034 }
2035
2036
2037
2038
2039 /*************************************************
2040 * Scan compiled regex for specific bracket *
2041 *************************************************/
2042
2043 /* This little function scans through a compiled pattern until it finds a
2044 capturing bracket with the given number, or, if the number is negative, an
2045 instance of OP_REVERSE for a lookbehind. The function is global in the C sense
2046 so that it can be called from pcre_study() when finding the minimum matching
2047 length.
2048
2049 Arguments:
2050 code points to start of expression
2051 utf TRUE in UTF-8 / UTF-16 / UTF-32 mode
2052 number the required bracket number or negative to find a lookbehind
2053
2054 Returns: pointer to the opcode for the bracket, or NULL if not found
2055 */
2056
2057 const pcre_uchar *
2058 PRIV(find_bracket)(const pcre_uchar *code, BOOL utf, int number)
2059 {
2060 for (;;)
2061 {
2062 register int c = *code;
2063
2064 if (c == OP_END) return NULL;
2065
2066 /* XCLASS is used for classes that cannot be represented just by a bit
2067 map. This includes negated single high-valued characters. The length in
2068 the table is zero; the actual length is stored in the compiled code. */
2069
2070 if (c == OP_XCLASS) code += GET(code, 1);
2071
2072 /* Handle recursion */
2073
2074 else if (c == OP_REVERSE)
2075 {
2076 if (number < 0) return (pcre_uchar *)code;
2077 code += PRIV(OP_lengths)[c];
2078 }
2079
2080 /* Handle capturing bracket */
2081
2082 else if (c == OP_CBRA || c == OP_SCBRA ||
2083 c == OP_CBRAPOS || c == OP_SCBRAPOS)
2084 {
2085 int n = GET2(code, 1+LINK_SIZE);
2086 if (n == number) return (pcre_uchar *)code;
2087 code += PRIV(OP_lengths)[c];
2088 }
2089
2090 /* Otherwise, we can get the item's length from the table, except that for
2091 repeated character types, we have to test for \p and \P, which have an extra
2092 two bytes of parameters, and for MARK/PRUNE/SKIP/THEN with an argument, we
2093 must add in its length. */
2094
2095 else
2096 {
2097 switch(c)
2098 {
2099 case OP_TYPESTAR:
2100 case OP_TYPEMINSTAR:
2101 case OP_TYPEPLUS:
2102 case OP_TYPEMINPLUS:
2103 case OP_TYPEQUERY:
2104 case OP_TYPEMINQUERY:
2105 case OP_TYPEPOSSTAR:
2106 case OP_TYPEPOSPLUS:
2107 case OP_TYPEPOSQUERY:
2108 if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2;
2109 break;
2110
2111 case OP_TYPEUPTO:
2112 case OP_TYPEMINUPTO:
2113 case OP_TYPEEXACT:
2114 case OP_TYPEPOSUPTO:
2115 if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP)
2116 code += 2;
2117 break;
2118
2119 case OP_MARK:
2120 case OP_PRUNE_ARG:
2121 case OP_SKIP_ARG:
2122 code += code[1];
2123 break;
2124
2125 case OP_THEN_ARG:
2126 code += code[1];
2127 break;
2128 }
2129
2130 /* Add in the fixed length from the table */
2131
2132 code += PRIV(OP_lengths)[c];
2133
2134 /* In UTF-8 mode, opcodes that are followed by a character may be followed by
2135 a multi-byte character. The length in the table is a minimum, so we have to
2136 arrange to skip the extra bytes. */
2137
2138 #if defined SUPPORT_UTF && !defined COMPILE_PCRE32
2139 if (utf) switch(c)
2140 {
2141 case OP_CHAR:
2142 case OP_CHARI:
2143 case OP_EXACT:
2144 case OP_EXACTI:
2145 case OP_UPTO:
2146 case OP_UPTOI:
2147 case OP_MINUPTO:
2148 case OP_MINUPTOI:
2149 case OP_POSUPTO:
2150 case OP_POSUPTOI:
2151 case OP_STAR:
2152 case OP_STARI:
2153 case OP_MINSTAR:
2154 case OP_MINSTARI:
2155 case OP_POSSTAR:
2156 case OP_POSSTARI:
2157 case OP_PLUS:
2158 case OP_PLUSI:
2159 case OP_MINPLUS:
2160 case OP_MINPLUSI:
2161 case OP_POSPLUS:
2162 case OP_POSPLUSI:
2163 case OP_QUERY:
2164 case OP_QUERYI:
2165 case OP_MINQUERY:
2166 case OP_MINQUERYI:
2167 case OP_POSQUERY:
2168 case OP_POSQUERYI:
2169 if (HAS_EXTRALEN(code[-1])) code += GET_EXTRALEN(code[-1]);
2170 break;
2171 }
2172 #else
2173 (void)(utf); /* Keep compiler happy by referencing function argument */
2174 #endif
2175 }
2176 }
2177 }
2178
2179
2180
2181 /*************************************************
2182 * Scan compiled regex for recursion reference *
2183 *************************************************/
2184
2185 /* This little function scans through a compiled pattern until it finds an
2186 instance of OP_RECURSE.
2187
2188 Arguments:
2189 code points to start of expression
2190 utf TRUE in UTF-8 / UTF-16 / UTF-32 mode
2191
2192 Returns: pointer to the opcode for OP_RECURSE, or NULL if not found
2193 */
2194
2195 static const pcre_uchar *
2196 find_recurse(const pcre_uchar *code, BOOL utf)
2197 {
2198 for (;;)
2199 {
2200 register int c = *code;
2201 if (c == OP_END) return NULL;
2202 if (c == OP_RECURSE) return code;
2203
2204 /* XCLASS is used for classes that cannot be represented just by a bit
2205 map. This includes negated single high-valued characters. The length in
2206 the table is zero; the actual length is stored in the compiled code. */
2207
2208 if (c == OP_XCLASS) code += GET(code, 1);
2209
2210 /* Otherwise, we can get the item's length from the table, except that for
2211 repeated character types, we have to test for \p and \P, which have an extra
2212 two bytes of parameters, and for MARK/PRUNE/SKIP/THEN with an argument, we
2213 must add in its length. */
2214
2215 else
2216 {
2217 switch(c)
2218 {
2219 case OP_TYPESTAR:
2220 case OP_TYPEMINSTAR:
2221 case OP_TYPEPLUS:
2222 case OP_TYPEMINPLUS:
2223 case OP_TYPEQUERY:
2224 case OP_TYPEMINQUERY:
2225 case OP_TYPEPOSSTAR:
2226 case OP_TYPEPOSPLUS:
2227 case OP_TYPEPOSQUERY:
2228 if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2;
2229 break;
2230
2231 case OP_TYPEPOSUPTO:
2232 case OP_TYPEUPTO:
2233 case OP_TYPEMINUPTO:
2234 case OP_TYPEEXACT:
2235 if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP)
2236 code += 2;
2237 break;
2238
2239 case OP_MARK:
2240 case OP_PRUNE_ARG:
2241 case OP_SKIP_ARG:
2242 code += code[1];
2243 break;
2244
2245 case OP_THEN_ARG:
2246 code += code[1];
2247 break;
2248 }
2249
2250 /* Add in the fixed length from the table */
2251
2252 code += PRIV(OP_lengths)[c];
2253
2254 /* In UTF-8 mode, opcodes that are followed by a character may be followed
2255 by a multi-byte character. The length in the table is a minimum, so we have
2256 to arrange to skip the extra bytes. */
2257
2258 #if defined SUPPORT_UTF && !defined COMPILE_PCRE32
2259 if (utf) switch(c)
2260 {
2261 case OP_CHAR:
2262 case OP_CHARI:
2263 case OP_NOT:
2264 case OP_NOTI:
2265 case OP_EXACT:
2266 case OP_EXACTI:
2267 case OP_NOTEXACT:
2268 case OP_NOTEXACTI:
2269 case OP_UPTO:
2270 case OP_UPTOI:
2271 case OP_NOTUPTO:
2272 case OP_NOTUPTOI:
2273 case OP_MINUPTO:
2274 case OP_MINUPTOI:
2275 case OP_NOTMINUPTO:
2276 case OP_NOTMINUPTOI:
2277 case OP_POSUPTO:
2278 case OP_POSUPTOI:
2279 case OP_NOTPOSUPTO:
2280 case OP_NOTPOSUPTOI:
2281 case OP_STAR:
2282 case OP_STARI:
2283 case OP_NOTSTAR:
2284 case OP_NOTSTARI:
2285 case OP_MINSTAR:
2286 case OP_MINSTARI:
2287 case OP_NOTMINSTAR:
2288 case OP_NOTMINSTARI:
2289 case OP_POSSTAR:
2290 case OP_POSSTARI:
2291 case OP_NOTPOSSTAR:
2292 case OP_NOTPOSSTARI:
2293 case OP_PLUS:
2294 case OP_PLUSI:
2295 case OP_NOTPLUS:
2296 case OP_NOTPLUSI:
2297 case OP_MINPLUS:
2298 case OP_MINPLUSI:
2299 case OP_NOTMINPLUS:
2300 case OP_NOTMINPLUSI:
2301 case OP_POSPLUS:
2302 case OP_POSPLUSI:
2303 case OP_NOTPOSPLUS:
2304 case OP_NOTPOSPLUSI:
2305 case OP_QUERY:
2306 case OP_QUERYI:
2307 case OP_NOTQUERY:
2308 case OP_NOTQUERYI:
2309 case OP_MINQUERY:
2310 case OP_MINQUERYI:
2311 case OP_NOTMINQUERY:
2312 case OP_NOTMINQUERYI:
2313 case OP_POSQUERY:
2314 case OP_POSQUERYI:
2315 case OP_NOTPOSQUERY:
2316 case OP_NOTPOSQUERYI:
2317 if (HAS_EXTRALEN(code[-1])) code += GET_EXTRALEN(code[-1]);
2318 break;
2319 }
2320 #else
2321 (void)(utf); /* Keep compiler happy by referencing function argument */
2322 #endif
2323 }
2324 }
2325 }
2326
2327
2328
2329 /*************************************************
2330 * Scan compiled branch for non-emptiness *
2331 *************************************************/
2332
2333 /* This function scans through a branch of a compiled pattern to see whether it
2334 can match the empty string or not. It is called from could_be_empty()
2335 below and from compile_branch() when checking for an unlimited repeat of a
2336 group that can match nothing. Note that first_significant_code() skips over
2337 backward and negative forward assertions when its final argument is TRUE. If we
2338 hit an unclosed bracket, we return "empty" - this means we've struck an inner
2339 bracket whose current branch will already have been scanned.
2340
2341 Arguments:
2342 code points to start of search
2343 endcode points to where to stop
2344 utf TRUE if in UTF-8 / UTF-16 / UTF-32 mode
2345 cd contains pointers to tables etc.
2346
2347 Returns: TRUE if what is matched could be empty
2348 */
2349
2350 static BOOL
2351 could_be_empty_branch(const pcre_uchar *code, const pcre_uchar *endcode,
2352 BOOL utf, compile_data *cd)
2353 {
2354 register int c;
2355 for (code = first_significant_code(code + PRIV(OP_lengths)[*code], TRUE);
2356 code < endcode;
2357 code = first_significant_code(code + PRIV(OP_lengths)[c], TRUE))
2358 {
2359 const pcre_uchar *ccode;
2360
2361 c = *code;
2362
2363 /* Skip over forward assertions; the other assertions are skipped by
2364 first_significant_code() with a TRUE final argument. */
2365
2366 if (c == OP_ASSERT)
2367 {
2368 do code += GET(code, 1); while (*code == OP_ALT);
2369 c = *code;
2370 continue;
2371 }
2372
2373 /* For a recursion/subroutine call, if its end has been reached, which
2374 implies a backward reference subroutine call, we can scan it. If it's a
2375 forward reference subroutine call, we can't. To detect forward reference
2376 we have to scan up the list that is kept in the workspace. This function is
2377 called only when doing the real compile, not during the pre-compile that
2378 measures the size of the compiled pattern. */
2379
2380 if (c == OP_RECURSE)
2381 {
2382 const pcre_uchar *scode;
2383 BOOL empty_branch;
2384
2385 /* Test for forward reference */
2386
2387 for (scode = cd->start_workspace; scode < cd->hwm; scode += LINK_SIZE)
2388 if (GET(scode, 0) == code + 1 - cd->start_code) return TRUE;
2389
2390 /* Not a forward reference, test for completed backward reference */
2391
2392 empty_branch = FALSE;
2393 scode = cd->start_code + GET(code, 1);
2394 if (GET(scode, 1) == 0) return TRUE; /* Unclosed */
2395
2396 /* Completed backwards reference */
2397
2398 do
2399 {
2400 if (could_be_empty_branch(scode, endcode, utf, cd))
2401 {
2402 empty_branch = TRUE;
2403 break;
2404 }
2405 scode += GET(scode, 1);
2406 }
2407 while (*scode == OP_ALT);
2408
2409 if (!empty_branch) return FALSE; /* All branches are non-empty */
2410 continue;
2411 }
2412
2413 /* Groups with zero repeats can of course be empty; skip them. */
2414
2415 if (c == OP_BRAZERO || c == OP_BRAMINZERO || c == OP_SKIPZERO ||
2416 c == OP_BRAPOSZERO)
2417 {
2418 code += PRIV(OP_lengths)[c];
2419 do code += GET(code, 1); while (*code == OP_ALT);
2420 c = *code;
2421 continue;
2422 }
2423
2424 /* A nested group that is already marked as "could be empty" can just be
2425 skipped. */
2426
2427 if (c == OP_SBRA || c == OP_SBRAPOS ||
2428 c == OP_SCBRA || c == OP_SCBRAPOS)
2429 {
2430 do code += GET(code, 1); while (*code == OP_ALT);
2431 c = *code;
2432 continue;
2433 }
2434
2435 /* For other groups, scan the branches. */
2436
2437 if (c == OP_BRA || c == OP_BRAPOS ||
2438 c == OP_CBRA || c == OP_CBRAPOS ||
2439 c == OP_ONCE || c == OP_ONCE_NC ||
2440 c == OP_COND)
2441 {
2442 BOOL empty_branch;
2443 if (GET(code, 1) == 0) return TRUE; /* Hit unclosed bracket */
2444
2445 /* If a conditional group has only one branch, there is a second, implied,
2446 empty branch, so just skip over the conditional, because it could be empty.
2447 Otherwise, scan the individual branches of the group. */
2448
2449 if (c == OP_COND && code[GET(code, 1)] != OP_ALT)
2450 code += GET(code, 1);
2451 else
2452 {
2453 empty_branch = FALSE;
2454 do
2455 {
2456 if (!empty_branch && could_be_empty_branch(code, endcode, utf, cd))
2457 empty_branch = TRUE;
2458 code += GET(code, 1);
2459 }
2460 while (*code == OP_ALT);
2461 if (!empty_branch) return FALSE; /* All branches are non-empty */
2462 }
2463
2464 c = *code;
2465 continue;
2466 }
2467
2468 /* Handle the other opcodes */
2469
2470 switch (c)
2471 {
2472 /* Check for quantifiers after a class. XCLASS is used for classes that
2473 cannot be represented just by a bit map. This includes negated single
2474 high-valued characters. The length in PRIV(OP_lengths)[] is zero; the
2475 actual length is stored in the compiled code, so we must update "code"
2476 here. */
2477
2478 #if defined SUPPORT_UTF || !defined COMPILE_PCRE8
2479 case OP_XCLASS:
2480 ccode = code += GET(code, 1);
2481 goto CHECK_CLASS_REPEAT;
2482 #endif
2483
2484 case OP_CLASS:
2485 case OP_NCLASS:
2486 ccode = code + PRIV(OP_lengths)[OP_CLASS];
2487
2488 #if defined SUPPORT_UTF || !defined COMPILE_PCRE8
2489 CHECK_CLASS_REPEAT:
2490 #endif
2491
2492 switch (*ccode)
2493 {
2494 case OP_CRSTAR: /* These could be empty; continue */
2495 case OP_CRMINSTAR:
2496 case OP_CRQUERY:
2497 case OP_CRMINQUERY:
2498 break;
2499
2500 default: /* Non-repeat => class must match */
2501 case OP_CRPLUS: /* These repeats aren't empty */
2502 case OP_CRMINPLUS:
2503 return FALSE;
2504
2505 case OP_CRRANGE:
2506 case OP_CRMINRANGE:
2507 if (GET2(ccode, 1) > 0) return FALSE; /* Minimum > 0 */
2508 break;
2509 }
2510 break;
2511
2512 /* Opcodes that must match a character */
2513
2514 case OP_PROP:
2515 case OP_NOTPROP:
2516 case OP_EXTUNI:
2517 case OP_NOT_DIGIT:
2518 case OP_DIGIT:
2519 case OP_NOT_WHITESPACE:
2520 case OP_WHITESPACE:
2521 case OP_NOT_WORDCHAR:
2522 case OP_WORDCHAR:
2523 case OP_ANY:
2524 case OP_ALLANY:
2525 case OP_ANYBYTE:
2526 case OP_CHAR:
2527 case OP_CHARI:
2528 case OP_NOT:
2529 case OP_NOTI:
2530 case OP_PLUS:
2531 case OP_MINPLUS:
2532 case OP_POSPLUS:
2533 case OP_EXACT:
2534 case OP_NOTPLUS:
2535 case OP_NOTMINPLUS:
2536 case OP_NOTPOSPLUS:
2537 case OP_NOTEXACT:
2538 case OP_TYPEPLUS:
2539 case OP_TYPEMINPLUS:
2540 case OP_TYPEPOSPLUS:
2541 case OP_TYPEEXACT:
2542 return FALSE;
2543
2544 /* These are going to continue, as they may be empty, but we have to
2545 fudge the length for the \p and \P cases. */
2546
2547 case OP_TYPESTAR:
2548 case OP_TYPEMINSTAR:
2549 case OP_TYPEPOSSTAR:
2550 case OP_TYPEQUERY:
2551 case OP_TYPEMINQUERY:
2552 case OP_TYPEPOSQUERY:
2553 if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2;
2554 break;
2555
2556 /* Same for these */
2557
2558 case OP_TYPEUPTO:
2559 case OP_TYPEMINUPTO:
2560 case OP_TYPEPOSUPTO:
2561 if (code[1 + IMM2_SIZE] == OP_PROP || code[1 + IMM2_SIZE] == OP_NOTPROP)
2562 code += 2;
2563 break;
2564
2565 /* End of branch */
2566
2567 case OP_KET:
2568 case OP_KETRMAX:
2569 case OP_KETRMIN:
2570 case OP_KETRPOS:
2571 case OP_ALT:
2572 return TRUE;
2573
2574 /* In UTF-8 mode, STAR, MINSTAR, POSSTAR, QUERY, MINQUERY, POSQUERY, UPTO,
2575 MINUPTO, and POSUPTO may be followed by a multibyte character */
2576
2577 #if defined SUPPORT_UTF && !defined COMPILE_PCRE32
2578 case OP_STAR:
2579 case OP_STARI:
2580 case OP_MINSTAR:
2581 case OP_MINSTARI:
2582 case OP_POSSTAR:
2583 case OP_POSSTARI:
2584 case OP_QUERY:
2585 case OP_QUERYI:
2586 case OP_MINQUERY:
2587 case OP_MINQUERYI:
2588 case OP_POSQUERY:
2589 case OP_POSQUERYI:
2590 if (utf && HAS_EXTRALEN(code[1])) code += GET_EXTRALEN(code[1]);
2591 break;
2592
2593 case OP_UPTO:
2594 case OP_UPTOI:
2595 case OP_MINUPTO:
2596 case OP_MINUPTOI:
2597 case OP_POSUPTO:
2598 case OP_POSUPTOI:
2599 if (utf && HAS_EXTRALEN(code[1 + IMM2_SIZE])) code += GET_EXTRALEN(code[1 + IMM2_SIZE]);
2600 break;
2601 #endif
2602
2603 /* MARK, and PRUNE/SKIP/THEN with an argument must skip over the argument
2604 string. */
2605
2606 case OP_MARK:
2607 case OP_PRUNE_ARG:
2608 case OP_SKIP_ARG:
2609 code += code[1];
2610 break;
2611
2612 case OP_THEN_ARG:
2613 code += code[1];
2614 break;
2615
2616 /* None of the remaining opcodes are required to match a character. */
2617
2618 default:
2619 break;
2620 }
2621 }
2622
2623 return TRUE;
2624 }
2625
2626
2627
2628 /*************************************************
2629 * Scan compiled regex for non-emptiness *
2630 *************************************************/
2631
2632 /* This function is called to check for left recursive calls. We want to check
2633 the current branch of the current pattern to see if it could match the empty
2634 string. If it could, we must look outwards for branches at other levels,
2635 stopping when we pass beyond the bracket which is the subject of the recursion.
2636 This function is called only during the real compile, not during the
2637 pre-compile.
2638
2639 Arguments:
2640 code points to start of the recursion
2641 endcode points to where to stop (current RECURSE item)
2642 bcptr points to the chain of current (unclosed) branch starts
2643 utf TRUE if in UTF-8 / UTF-16 / UTF-32 mode
2644 cd pointers to tables etc
2645
2646 Returns: TRUE if what is matched could be empty
2647 */
2648
2649 static BOOL
2650 could_be_empty(const pcre_uchar *code, const pcre_uchar *endcode,
2651 branch_chain *bcptr, BOOL utf, compile_data *cd)
2652 {
2653 while (bcptr != NULL && bcptr->current_branch >= code)
2654 {
2655 if (!could_be_empty_branch(bcptr->current_branch, endcode, utf, cd))
2656 return FALSE;
2657 bcptr = bcptr->outer;
2658 }
2659 return TRUE;
2660 }
2661
2662
2663
2664 /*************************************************
2665 * Check for POSIX class syntax *
2666 *************************************************/
2667
2668 /* This function is called when the sequence "[:" or "[." or "[=" is
2669 encountered in a character class. It checks whether this is followed by a
2670 sequence of characters terminated by a matching ":]" or ".]" or "=]". If we
2671 reach an unescaped ']' without the special preceding character, return FALSE.
2672
2673 Originally, this function only recognized a sequence of letters between the
2674 terminators, but it seems that Perl recognizes any sequence of characters,
2675 though of course unknown POSIX names are subsequently rejected. Perl gives an
2676 "Unknown POSIX class" error for [:f\oo:] for example, where previously PCRE
2677 didn't consider this to be a POSIX class. Likewise for [:1234:].
2678
2679 The problem in trying to be exactly like Perl is in the handling of escapes. We
2680 have to be sure that [abc[:x\]pqr] is *not* treated as containing a POSIX
2681 class, but [abc[:x\]pqr:]] is (so that an error can be generated). The code
2682 below handles the special case of \], but does not try to do any other escape
2683 processing. This makes it different from Perl for cases such as [:l\ower:]
2684 where Perl recognizes it as the POSIX class "lower" but PCRE does not recognize
2685 "l\ower". This is a lesser evil that not diagnosing bad classes when Perl does,
2686 I think.
2687
2688 A user pointed out that PCRE was rejecting [:a[:digit:]] whereas Perl was not.
2689 It seems that the appearance of a nested POSIX class supersedes an apparent
2690 external class. For example, [:a[:digit:]b:] matches "a", "b", ":", or
2691 a digit.
2692
2693 In Perl, unescaped square brackets may also appear as part of class names. For
2694 example, [:a[:abc]b:] gives unknown POSIX class "[:abc]b:]". However, for
2695 [:a[:abc]b][b:] it gives unknown POSIX class "[:abc]b][b:]", which does not
2696 seem right at all. PCRE does not allow closing square brackets in POSIX class
2697 names.
2698
2699 Arguments:
2700 ptr pointer to the initial [
2701 endptr where to return the end pointer
2702
2703 Returns: TRUE or FALSE
2704 */
2705
2706 static BOOL
2707 check_posix_syntax(const pcre_uchar *ptr, const pcre_uchar **endptr)
2708 {
2709 int terminator; /* Don't combine these lines; the Solaris cc */
2710 terminator = *(++ptr); /* compiler warns about "non-constant" initializer. */
2711 for (++ptr; *ptr != 0; ptr++)
2712 {
2713 if (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_RIGHT_SQUARE_BRACKET)
2714 ptr++;
2715 else if (*ptr == CHAR_RIGHT_SQUARE_BRACKET) return FALSE;
2716 else
2717 {
2718 if (*ptr == terminator && ptr[1] == CHAR_RIGHT_SQUARE_BRACKET)
2719 {
2720 *endptr = ptr;
2721 return TRUE;
2722 }
2723 if (*ptr == CHAR_LEFT_SQUARE_BRACKET &&
2724 (ptr[1] == CHAR_COLON || ptr[1] == CHAR_DOT ||
2725 ptr[1] == CHAR_EQUALS_SIGN) &&
2726 check_posix_syntax(ptr, endptr))
2727 return FALSE;
2728 }
2729 }
2730 return FALSE;
2731 }
2732
2733
2734
2735
2736 /*************************************************
2737 * Check POSIX class name *
2738 *************************************************/
2739
2740 /* This function is called to check the name given in a POSIX-style class entry
2741 such as [:alnum:].
2742
2743 Arguments:
2744 ptr points to the first letter
2745 len the length of the name
2746
2747 Returns: a value representing the name, or -1 if unknown
2748 */
2749
2750 static int
2751 check_posix_name(const pcre_uchar *ptr, int len)
2752 {
2753 const char *pn = posix_names;
2754 register int yield = 0;
2755 while (posix_name_lengths[yield] != 0)
2756 {
2757 if (len == posix_name_lengths[yield] &&
2758 STRNCMP_UC_C8(ptr, pn, len) == 0) return yield;
2759 pn += posix_name_lengths[yield] + 1;
2760 yield++;
2761 }
2762 return -1;
2763 }
2764
2765
2766 /*************************************************
2767 * Adjust OP_RECURSE items in repeated group *
2768 *************************************************/
2769
2770 /* OP_RECURSE items contain an offset from the start of the regex to the group
2771 that is referenced. This means that groups can be replicated for fixed
2772 repetition simply by copying (because the recursion is allowed to refer to
2773 earlier groups that are outside the current group). However, when a group is
2774 optional (i.e. the minimum quantifier is zero), OP_BRAZERO or OP_SKIPZERO is
2775 inserted before it, after it has been compiled. This means that any OP_RECURSE
2776 items within it that refer to the group itself or any contained groups have to
2777 have their offsets adjusted. That one of the jobs of this function. Before it
2778 is called, the partially compiled regex must be temporarily terminated with
2779 OP_END.
2780
2781 This function has been extended with the possibility of forward references for
2782 recursions and subroutine calls. It must also check the list of such references
2783 for the group we are dealing with. If it finds that one of the recursions in
2784 the current group is on this list, it adjusts the offset in the list, not the
2785 value in the reference (which is a group number).
2786
2787 Arguments:
2788 group points to the start of the group
2789 adjust the amount by which the group is to be moved
2790 utf TRUE in UTF-8 / UTF-16 / UTF-32 mode
2791 cd contains pointers to tables etc.
2792 save_hwm the hwm forward reference pointer at the start of the group
2793
2794 Returns: nothing
2795 */
2796
2797 static void
2798 adjust_recurse(pcre_uchar *group, int adjust, BOOL utf, compile_data *cd,
2799 pcre_uchar *save_hwm)
2800 {
2801 pcre_uchar *ptr = group;
2802
2803 while ((ptr = (pcre_uchar *)find_recurse(ptr, utf)) != NULL)
2804 {
2805 int offset;
2806 pcre_uchar *hc;
2807
2808 /* See if this recursion is on the forward reference list. If so, adjust the
2809 reference. */
2810
2811 for (hc = save_hwm; hc < cd->hwm; hc += LINK_SIZE)
2812 {
2813 offset = GET(hc, 0);
2814 if (cd->start_code + offset == ptr + 1)
2815 {
2816 PUT(hc, 0, offset + adjust);
2817 break;
2818 }
2819 }
2820
2821 /* Otherwise, adjust the recursion offset if it's after the start of this
2822 group. */
2823
2824 if (hc >= cd->hwm)
2825 {
2826 offset = GET(ptr, 1);
2827 if (cd->start_code + offset >= group) PUT(ptr, 1, offset + adjust);
2828 }
2829
2830 ptr += 1 + LINK_SIZE;
2831 }
2832 }
2833
2834
2835
2836 /*************************************************
2837 * Insert an automatic callout point *
2838 *************************************************/
2839
2840 /* This function is called when the PCRE_AUTO_CALLOUT option is set, to insert
2841 callout points before each pattern item.
2842
2843 Arguments:
2844 code current code pointer
2845 ptr current pattern pointer
2846 cd pointers to tables etc
2847
2848 Returns: new code pointer
2849 */
2850
2851 static pcre_uchar *
2852 auto_callout(pcre_uchar *code, const pcre_uchar *ptr, compile_data *cd)
2853 {
2854 *code++ = OP_CALLOUT;
2855 *code++ = 255;
2856 PUT(code, 0, (int)(ptr - cd->start_pattern)); /* Pattern offset */
2857 PUT(code, LINK_SIZE, 0); /* Default length */
2858 return code + 2 * LINK_SIZE;
2859 }
2860
2861
2862
2863 /*************************************************
2864 * Complete a callout item *
2865 *************************************************/
2866
2867 /* A callout item contains the length of the next item in the pattern, which
2868 we can't fill in till after we have reached the relevant point. This is used
2869 for both automatic and manual callouts.
2870
2871 Arguments:
2872 previous_callout points to previous callout item
2873 ptr current pattern pointer
2874 cd pointers to tables etc
2875
2876 Returns: nothing
2877 */
2878
2879 static void
2880 complete_callout(pcre_uchar *previous_callout, const pcre_uchar *ptr, compile_data *cd)
2881 {
2882 int length = (int)(ptr - cd->start_pattern - GET(previous_callout, 2));
2883 PUT(previous_callout, 2 + LINK_SIZE, length);
2884 }
2885
2886
2887
2888 #ifdef SUPPORT_UCP
2889 /*************************************************
2890 * Get othercase range *
2891 *************************************************/
2892
2893 /* This function is passed the start and end of a class range, in UTF-8 mode
2894 with UCP support. It searches up the characters, looking for ranges of
2895 characters in the "other" case. Each call returns the next one, updating the
2896 start address. A character with multiple other cases is returned on its own
2897 with a special return value.
2898
2899 Arguments:
2900 cptr points to starting character value; updated
2901 d end value
2902 ocptr where to put start of othercase range
2903 odptr where to put end of othercase range
2904
2905 Yield: -1 when no more
2906 0 when a range is returned
2907 >0 the CASESET offset for char with multiple other cases
2908 in this case, ocptr contains the original
2909 */
2910
2911 static int
2912 get_othercase_range(pcre_uint32 *cptr, pcre_uint32 d, pcre_uint32 *ocptr,
2913 pcre_uint32 *odptr)
2914 {
2915 pcre_uint32 c, othercase, next;
2916 int co;
2917
2918 /* Find the first character that has an other case. If it has multiple other
2919 cases, return its case offset value. */
2920
2921 for (c = *cptr; c <= d; c++)
2922 {
2923 if ((co = UCD_CASESET(c)) != 0)
2924 {
2925 *ocptr = c++; /* Character that has the set */
2926 *cptr = c; /* Rest of input range */
2927 return co;
2928 }
2929 if ((othercase = UCD_OTHERCASE(c)) != c) break;
2930 }
2931
2932 if (c > d) return -1; /* Reached end of range */
2933
2934 *ocptr = othercase;
2935 next = othercase + 1;
2936
2937 for (++c; c <= d; c++)
2938 {
2939 if (UCD_OTHERCASE(c) != next) break;
2940 next++;
2941 }
2942
2943 *odptr = next - 1; /* End of othercase range */
2944 *cptr = c; /* Rest of input range */
2945 return 0;
2946 }
2947
2948
2949
2950 /*************************************************
2951 * Check a character and a property *
2952 *************************************************/
2953
2954 /* This function is called by check_auto_possessive() when a property item
2955 is adjacent to a fixed character.
2956
2957 Arguments:
2958 c the character
2959 ptype the property type
2960 pdata the data for the type
2961 negated TRUE if it's a negated property (\P or \p{^)
2962
2963 Returns: TRUE if auto-possessifying is OK
2964 */
2965
2966 static BOOL
2967 check_char_prop(pcre_uint32 c, int ptype, int pdata, BOOL negated)
2968 {
2969 #ifdef SUPPORT_UCP
2970 const pcre_uint32 *p;
2971 #endif
2972
2973 const ucd_record *prop = GET_UCD(c);
2974
2975 switch(ptype)
2976 {
2977 case PT_LAMP:
2978 return (prop->chartype == ucp_Lu ||
2979 prop->chartype == ucp_Ll ||
2980 prop->chartype == ucp_Lt) == negated;
2981
2982 case PT_GC:
2983 return (pdata == PRIV(ucp_gentype)[prop->chartype]) == negated;
2984
2985 case PT_PC:
2986 return (pdata == prop->chartype) == negated;
2987
2988 case PT_SC:
2989 return (pdata == prop->script) == negated;
2990
2991 /* These are specials */
2992
2993 case PT_ALNUM:
2994 return (PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
2995 PRIV(ucp_gentype)[prop->chartype] == ucp_N) == negated;
2996
2997 case PT_SPACE: /* Perl space */
2998 return (PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
2999 c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR)
3000 == negated;
3001
3002 case PT_PXSPACE: /* POSIX space */
3003 return (PRIV(ucp_gentype)[prop->chartype] == ucp_Z ||
3004 c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
3005 c == CHAR_FF || c == CHAR_CR)
3006 == negated;
3007
3008 case PT_WORD:
3009 return (PRIV(ucp_gentype)[prop->chartype] == ucp_L ||
3010 PRIV(ucp_gentype)[prop->chartype] == ucp_N ||
3011 c == CHAR_UNDERSCORE) == negated;
3012
3013 #ifdef SUPPORT_UCP
3014 case PT_CLIST:
3015 p = PRIV(ucd_caseless_sets) + prop->caseset;
3016 for (;;)
3017 {
3018 if ((unsigned int)c < *p) return !negated;
3019 if ((unsigned int)c == *p++) return negated;
3020 }
3021 break; /* Control never reaches here */
3022 #endif
3023 }
3024
3025 return FALSE;
3026 }
3027 #endif /* SUPPORT_UCP */
3028
3029
3030
3031 /*************************************************
3032 * Check if auto-possessifying is possible *
3033 *************************************************/
3034
3035 /* This function is called for unlimited repeats of certain items, to see
3036 whether the next thing could possibly match the repeated item. If not, it makes
3037 sense to automatically possessify the repeated item.
3038
3039 Arguments:
3040 previous pointer to the repeated opcode
3041 utf TRUE in UTF-8 / UTF-16 / UTF-32 mode
3042 ptr next character in pattern
3043 options options bits
3044 cd contains pointers to tables etc.
3045
3046 Returns: TRUE if possessifying is wanted
3047 */
3048
3049 static BOOL
3050 check_auto_possessive(const pcre_uchar *previous, BOOL utf,
3051 const pcre_uchar *ptr, int options, compile_data *cd)
3052 {
3053 pcre_uint32 c = NOTACHAR;
3054 pcre_uint32 next;
3055 int escape;
3056 int op_code = *previous++;
3057
3058 /* Skip whitespace and comments in extended mode */
3059
3060 if ((options & PCRE_EXTENDED) != 0)
3061 {
3062 for (;;)
3063 {
3064 while (MAX_255(*ptr) && (cd->ctypes[*ptr] & ctype_space) != 0) ptr++;
3065 if (*ptr == CHAR_NUMBER_SIGN)
3066 {
3067 ptr++;
3068 while (*ptr != 0)
3069 {
3070 if (IS_NEWLINE(ptr)) { ptr += cd->nllen; break; }
3071 ptr++;
3072 #ifdef SUPPORT_UTF
3073 if (utf) FORWARDCHAR(ptr);
3074 #endif
3075 }
3076 }
3077 else break;
3078 }
3079 }
3080
3081 /* If the next item is one that we can handle, get its value. A non-negative
3082 value is a character, a negative value is an escape value. */
3083
3084 if (*ptr == CHAR_BACKSLASH)
3085 {
3086 int temperrorcode = 0;
3087 escape = check_escape(&ptr, &next, &temperrorcode, cd->bracount, options, FALSE);
3088 if (temperrorcode != 0) return FALSE;
3089 ptr++; /* Point after the escape sequence */
3090 }
3091 else if (!MAX_255(*ptr) || (cd->ctypes[*ptr] & ctype_meta) == 0)
3092 {
3093 escape = 0;
3094 #ifdef SUPPORT_UTF
3095 if (utf) { GETCHARINC(next, ptr); } else
3096 #endif
3097 next = *ptr++;
3098 }
3099 else return FALSE;
3100
3101 /* Skip whitespace and comments in extended mode */
3102
3103 if ((options & PCRE_EXTENDED) != 0)
3104 {
3105 for (;;)
3106 {
3107 while (MAX_255(*ptr) && (cd->ctypes[*ptr] & ctype_space) != 0) ptr++;
3108 if (*ptr == CHAR_NUMBER_SIGN)
3109 {
3110 ptr++;
3111 while (*ptr != 0)
3112 {
3113 if (IS_NEWLINE(ptr)) { ptr += cd->nllen; break; }
3114 ptr++;
3115 #ifdef SUPPORT_UTF
3116 if (utf) FORWARDCHAR(ptr);
3117 #endif
3118 }
3119 }
3120 else break;
3121 }
3122 }
3123
3124 /* If the next thing is itself optional, we have to give up. */
3125
3126 if (*ptr == CHAR_ASTERISK || *ptr == CHAR_QUESTION_MARK ||
3127 STRNCMP_UC_C8(ptr, STR_LEFT_CURLY_BRACKET STR_0 STR_COMMA, 3) == 0)
3128 return FALSE;
3129
3130 /* If the previous item is a character, get its value. */
3131
3132 if (op_code == OP_CHAR || op_code == OP_CHARI ||
3133 op_code == OP_NOT || op_code == OP_NOTI)
3134 //if (escape == 0) switch(op_code)
3135 {
3136 #ifdef SUPPORT_UTF
3137 GETCHARTEST(c, previous);
3138 #else
3139 c = *previous;
3140 #endif
3141 }
3142
3143 /* Now compare the next item with the previous opcode. First, handle cases when
3144 the next item is a character. */
3145
3146 if (escape == 0)
3147 {
3148 /* For a caseless UTF match, the next character may have more than one other
3149 case, which maps to the special PT_CLIST property. Check this first. */
3150
3151 #ifdef SUPPORT_UCP
3152 if (utf && c != NOTACHAR && (options & PCRE_CASELESS) != 0)
3153 {
3154 int ocs = UCD_CASESET(next);
3155 if (ocs > 0) return check_char_prop(c, PT_CLIST, ocs, op_code >= OP_NOT);
3156 }
3157 #endif
3158
3159 switch(op_code)
3160 {
3161 case OP_CHAR:
3162 return c != next;
3163
3164 /* For CHARI (caseless character) we must check the other case. If we have
3165 Unicode property support, we can use it to test the other case of
3166 high-valued characters. We know that next can have only one other case,
3167 because multi-other-case characters are dealt with above. */
3168
3169 case OP_CHARI:
3170 if (c == next) return FALSE;
3171 #ifdef SUPPORT_UTF
3172 if (utf)
3173 {
3174 pcre_uint32 othercase;
3175 if (next < 128) othercase = cd->fcc[next]; else
3176 #ifdef SUPPORT_UCP
3177 othercase = UCD_OTHERCASE(next);
3178 #else
3179 othercase = NOTACHAR;
3180 #endif
3181 return c != othercase;
3182 }
3183 else
3184 #endif /* SUPPORT_UTF */
3185 return (c != TABLE_GET(next, cd->fcc, next)); /* Not UTF */
3186
3187 case OP_NOT:
3188 return c == next;
3189
3190 case OP_NOTI:
3191 if (c == next) return TRUE;
3192 #ifdef SUPPORT_UTF
3193 if (utf)
3194 {
3195 pcre_uint32 othercase;
3196 if (next < 128) othercase = cd->fcc[next]; else
3197 #ifdef SUPPORT_UCP
3198 othercase = UCD_OTHERCASE(next);
3199 #else
3200 othercase = NOTACHAR;
3201 #endif
3202 return c == othercase;
3203 }
3204 else
3205 #endif /* SUPPORT_UTF */
3206 return (c == TABLE_GET(next, cd->fcc, next)); /* Not UTF */
3207
3208 /* Note that OP_DIGIT etc. are generated only when PCRE_UCP is *not* set.
3209 When it is set, \d etc. are converted into OP_(NOT_)PROP codes. */
3210
3211 case OP_DIGIT:
3212 return next > 255 || (cd->ctypes[next] & ctype_digit) == 0;
3213
3214 case OP_NOT_DIGIT:
3215 return next <= 255 && (cd->ctypes[next] & ctype_digit) != 0;
3216
3217 case OP_WHITESPACE:
3218 return next > 255 || (cd->ctypes[next] & ctype_space) == 0;
3219
3220 case OP_NOT_WHITESPACE:
3221 return next <= 255 && (cd->ctypes[next] & ctype_space) != 0;
3222
3223 case OP_WORDCHAR:
3224 return next > 255 || (cd->ctypes[next] & ctype_word) == 0;
3225
3226 case OP_NOT_WORDCHAR:
3227 return next <= 255 && (cd->ctypes[next] & ctype_word) != 0;
3228
3229 case OP_HSPACE:
3230 case OP_NOT_HSPACE:
3231 switch(next)
3232 {
3233 HSPACE_CASES:
3234 return op_code == OP_NOT_HSPACE;
3235
3236 default:
3237 return op_code != OP_NOT_HSPACE;
3238 }
3239
3240 case OP_ANYNL:
3241 case OP_VSPACE:
3242 case OP_NOT_VSPACE:
3243 switch(next)
3244 {
3245 VSPACE_CASES:
3246 return op_code == OP_NOT_VSPACE;
3247
3248 default:
3249 return op_code != OP_NOT_VSPACE;
3250 }
3251
3252 #ifdef SUPPORT_UCP
3253 case OP_PROP:
3254 return check_char_prop(next, previous[0], previous[1], FALSE);
3255
3256 case OP_NOTPROP:
3257 return check_char_prop(next, previous[0], previous[1], TRUE);
3258 #endif
3259
3260 default:
3261 return FALSE;
3262 }
3263 }
3264
3265 /* Handle the case when the next item is \d, \s, etc. Note that when PCRE_UCP
3266 is set, \d turns into ESC_du rather than ESC_d, etc., so ESC_d etc. are
3267 generated only when PCRE_UCP is *not* set, that is, when only ASCII
3268 characteristics are recognized. Similarly, the opcodes OP_DIGIT etc. are
3269 replaced by OP_PROP codes when PCRE_UCP is set. */
3270
3271 switch(op_code)
3272 {
3273 case OP_CHAR:
3274 case OP_CHARI:
3275 switch(escape)
3276 {
3277 case ESC_d:
3278 return c > 255 || (cd->ctypes[c] & ctype_digit) == 0;
3279
3280 case ESC_D:
3281 return c <= 255 && (cd->ctypes[c] & ctype_digit) != 0;
3282
3283 case ESC_s:
3284 return c > 255 || (cd->ctypes[c] & ctype_space) == 0;
3285
3286 case ESC_S:
3287 return c <= 255 && (cd->ctypes[c] & ctype_space) != 0;
3288
3289 case ESC_w:
3290 return c > 255 || (cd->ctypes[c] & ctype_word) == 0;
3291
3292 case ESC_W:
3293 return c <= 255 && (cd->ctypes[c] & ctype_word) != 0;
3294
3295 case ESC_h:
3296 case ESC_H:
3297 switch(c)
3298 {
3299 HSPACE_CASES:
3300 return escape != ESC_h;
3301
3302 default:
3303 return escape == ESC_h;
3304 }
3305
3306 case ESC_v:
3307 case ESC_V:
3308 switch(c)
3309 {
3310 VSPACE_CASES:
3311 return escape != ESC_v;
3312
3313 default:
3314 return escape == ESC_v;
3315 }
3316
3317 /* When PCRE_UCP is set, these values get generated for \d etc. Find
3318 their substitutions and process them. The result will always be either
3319 ESC_p or ESC_P. Then fall through to process those values. */
3320
3321 #ifdef SUPPORT_UCP
3322 case ESC_du:
3323 case ESC_DU:
3324 case ESC_wu:
3325 case ESC_WU:
3326 case ESC_su:
3327 case ESC_SU:
3328 {
3329 int temperrorcode = 0;
3330 ptr = substitutes[escape - ESC_DU];
3331 escape = check_escape(&ptr, &next, &temperrorcode, 0, options, FALSE);
3332 if (temperrorcode != 0) return FALSE;
3333 ptr++; /* For compatibility */
3334 }
3335 /* Fall through */
3336
3337 case ESC_p:
3338 case ESC_P:
3339 {
3340 int ptype, pdata, errorcodeptr;
3341 BOOL negated;
3342
3343 ptr--; /* Make ptr point at the p or P */
3344 ptype = get_ucp(&ptr, &negated, &pdata, &errorcodeptr);
3345 if (ptype < 0) return FALSE;
3346 ptr++; /* Point past the final curly ket */
3347
3348 /* If the property item is optional, we have to give up. (When generated
3349 from \d etc by PCRE_UCP, this test will have been applied much earlier,
3350 to the original \d etc. At this point, ptr will point to a zero byte. */
3351
3352 if (*ptr == CHAR_ASTERISK || *ptr == CHAR_QUESTION_MARK ||
3353 STRNCMP_UC_C8(ptr, STR_LEFT_CURLY_BRACKET STR_0 STR_COMMA, 3) == 0)
3354 return FALSE;
3355
3356 /* Do the property check. */
3357
3358 return check_char_prop(c, ptype, pdata, (escape == ESC_P) != negated);
3359 }
3360 #endif
3361
3362 default:
3363 return FALSE;
3364 }
3365
3366 /* In principle, support for Unicode properties should be integrated here as
3367 well. It means re-organizing the above code so as to get hold of the property
3368 values before switching on the op-code. However, I wonder how many patterns
3369 combine ASCII \d etc with Unicode properties? (Note that if PCRE_UCP is set,
3370 these op-codes are never generated.) */
3371
3372 case OP_DIGIT:
3373 return escape == ESC_D || escape == ESC_s || escape == ESC_W ||
3374 escape == ESC_h || escape == ESC_v || escape == ESC_R;
3375
3376 case OP_NOT_DIGIT:
3377 return escape == ESC_d;
3378
3379 case OP_WHITESPACE:
3380 return escape == ESC_S || escape == ESC_d || escape == ESC_w;
3381
3382 case OP_NOT_WHITESPACE:
3383 return escape == ESC_s || escape == ESC_h || escape == ESC_v || escape == ESC_R;
3384
3385 case OP_HSPACE:
3386 return escape == ESC_S || escape == ESC_H || escape == ESC_d ||
3387 escape == ESC_w || escape == ESC_v || escape == ESC_R;
3388
3389 case OP_NOT_HSPACE:
3390 return escape == ESC_h;
3391
3392 /* Can't have \S in here because VT matches \S (Perl anomaly) */
3393 case OP_ANYNL:
3394 case OP_VSPACE:
3395 return escape == ESC_V || escape == ESC_d || escape == ESC_w;
3396
3397 case OP_NOT_VSPACE:
3398 return escape == ESC_v || escape == ESC_R;
3399
3400 case OP_WORDCHAR:
3401 return escape == ESC_W || escape == ESC_s || escape == ESC_h ||
3402 escape == ESC_v || escape == ESC_R;
3403
3404 case OP_NOT_WORDCHAR:
3405 return escape == ESC_w || escape == ESC_d;
3406
3407 default:
3408 return FALSE;
3409 }
3410
3411 /* Control does not reach here */
3412 }
3413
3414
3415
3416 /*************************************************
3417 * Add a character or range to a class *
3418 *************************************************/
3419
3420 /* This function packages up the logic of adding a character or range of
3421 characters to a class. The character values in the arguments will be within the
3422 valid values for the current mode (8-bit, 16-bit, UTF, etc). This function is
3423 mutually recursive with the function immediately below.
3424
3425 Arguments:
3426 classbits the bit map for characters < 256
3427 uchardptr points to the pointer for extra data
3428 options the options word
3429 cd contains pointers to tables etc.
3430 start start of range character
3431 end end of range character
3432
3433 Returns: the number of < 256 characters added
3434 the pointer to extra data is updated
3435 */
3436
3437 static int
3438 add_to_class(pcre_uint8 *classbits, pcre_uchar **uchardptr, int options,
3439 compile_data *cd, pcre_uint32 start, pcre_uint32 end)
3440 {
3441 pcre_uint32 c;
3442 int n8 = 0;
3443
3444 /* If caseless matching is required, scan the range and process alternate
3445 cases. In Unicode, there are 8-bit characters that have alternate cases that
3446 are greater than 255 and vice-versa. Sometimes we can just extend the original
3447 range. */
3448
3449 if ((options & PCRE_CASELESS) != 0)
3450 {
3451 #ifdef SUPPORT_UCP
3452 if ((options & PCRE_UTF8) != 0)
3453 {
3454 int rc;
3455 pcre_uint32 oc, od;
3456
3457 options &= ~PCRE_CASELESS; /* Remove for recursive calls */
3458 c = start;
3459
3460 while ((rc = get_othercase_range(&c, end, &oc, &od)) >= 0)
3461 {
3462 /* Handle a single character that has more than one other case. */
3463
3464 if (rc > 0) n8 += add_list_to_class(classbits, uchardptr, options, cd,
3465 PRIV(ucd_caseless_sets) + rc, oc);
3466
3467 /* Do nothing if the other case range is within the original range. */
3468
3469 else if (oc >= start && od <= end) continue;
3470
3471 /* Extend the original range if there is overlap, noting that if oc < c, we
3472 can't have od > end because a subrange is always shorter than the basic
3473 range. Otherwise, use a recursive call to add the additional range. */
3474
3475 else if (oc < start && od >= start - 1) start = oc; /* Extend downwards */
3476 else if (od > end && oc <= end + 1) end = od; /* Extend upwards */
3477 else n8 += add_to_class(classbits, uchardptr, options, cd, oc, od);
3478 }
3479 }
3480 else
3481 #endif /* SUPPORT_UCP */
3482
3483 /* Not UTF-mode, or no UCP */
3484
3485 for (c = start; c <= end && c < 256; c++)
3486 {
3487 SETBIT(classbits, cd->fcc[c]);
3488 n8++;
3489 }
3490 }
3491
3492 /* Now handle the original range. Adjust the final value according to the bit
3493 length - this means that the same lists of (e.g.) horizontal spaces can be used
3494 in all cases. */
3495
3496 #if defined COMPILE_PCRE8
3497 #ifdef SUPPORT_UTF
3498 if ((options & PCRE_UTF8) == 0)
3499 #endif
3500 if (end > 0xff) end = 0xff;
3501
3502 #elif defined COMPILE_PCRE16
3503 #ifdef SUPPORT_UTF
3504 if ((options & PCRE_UTF16) == 0)
3505 #endif
3506 if (end > 0xffff) end = 0xffff;
3507
3508 #endif /* COMPILE_PCRE[8|16] */
3509
3510 /* If all characters are less than 256, use the bit map. Otherwise use extra
3511 data. */
3512
3513 if (end < 0x100)
3514 {
3515 for (c = start; c <= end; c++)
3516 {
3517 n8++;
3518 SETBIT(classbits, c);
3519 }
3520 }
3521
3522 else
3523 {
3524 pcre_uchar *uchardata = *uchardptr;
3525
3526 #ifdef SUPPORT_UTF
3527 if ((options & PCRE_UTF8) != 0) /* All UTFs use the same flag bit */
3528 {
3529 if (start < end)
3530 {
3531 *uchardata++ = XCL_RANGE;
3532 uchardata += PRIV(ord2utf)(start, uchardata);
3533 uchardata += PRIV(ord2utf)(end, uchardata);
3534 }
3535 else if (start == end)
3536 {
3537 *uchardata++ = XCL_SINGLE;
3538 uchardata += PRIV(ord2utf)(start, uchardata);
3539 }
3540 }
3541 else
3542 #endif /* SUPPORT_UTF */
3543
3544 /* Without UTF support, character values are constrained by the bit length,
3545 and can only be > 256 for 16-bit and 32-bit libraries. */
3546
3547 #ifdef COMPILE_PCRE8
3548 {}
3549 #else
3550 if (start < end)
3551 {
3552 *uchardata++ = XCL_RANGE;
3553 *uchardata++ = start;
3554 *uchardata++ = end;
3555 }
3556 else if (start == end)
3557 {
3558 *uchardata++ = XCL_SINGLE;
3559 *uchardata++ = start;
3560 }
3561 #endif
3562
3563 *uchardptr = uchardata; /* Updata extra data pointer */
3564 }
3565
3566 return n8; /* Number of 8-bit characters */
3567 }
3568
3569
3570
3571
3572 /*************************************************
3573 * Add a list of characters to a class *
3574 *************************************************/
3575
3576 /* This function is used for adding a list of case-equivalent characters to a
3577 class, and also for adding a list of horizontal or vertical whitespace. If the
3578 list is in order (which it should be), ranges of characters are detected and
3579 handled appropriately. This function is mutually recursive with the function
3580 above.
3581
3582 Arguments:
3583 classbits the bit map for characters < 256
3584 uchardptr points to the pointer for extra data
3585 options the options word
3586 cd contains pointers to tables etc.
3587 p points to row of 32-bit values, terminated by NOTACHAR
3588 except character to omit; this is used when adding lists of
3589 case-equivalent characters to avoid including the one we
3590 already know about
3591
3592 Returns: the number of < 256 characters added
3593 the pointer to extra data is updated
3594 */
3595
3596 static int
3597 add_list_to_class(pcre_uint8 *classbits, pcre_uchar **uchardptr, int options,
3598 compile_data *cd, const pcre_uint32 *p, unsigned int except)
3599 {
3600 int n8 = 0;
3601 while (p[0] < NOTACHAR)
3602 {
3603 int n = 0;
3604 if (p[0] != except)
3605 {
3606 while(p[n+1] == p[0] + n + 1) n++;
3607 n8 += add_to_class(classbits, uchardptr, options, cd, p[0], p[n]);
3608 }
3609 p += n + 1;
3610 }
3611 return n8;
3612 }
3613
3614
3615
3616 /*************************************************
3617 * Add characters not in a list to a class *
3618 *************************************************/
3619
3620 /* This function is used for adding the complement of a list of horizontal or
3621 vertical whitespace to a class. The list must be in order.
3622
3623 Arguments:
3624 classbits the bit map for characters < 256
3625 uchardptr points to the pointer for extra data
3626 options the options word
3627 cd contains pointers to tables etc.
3628 p points to row of 32-bit values, terminated by NOTACHAR
3629
3630 Returns: the number of < 256 characters added
3631 the pointer to extra data is updated
3632 */
3633
3634 static int
3635 add_not_list_to_class(pcre_uint8 *classbits, pcre_uchar **uchardptr,
3636 int options, compile_data *cd, const pcre_uint32 *p)
3637 {
3638 BOOL utf = (options & PCRE_UTF8) != 0;
3639 int n8 = 0;
3640 if (p[0] > 0)
3641 n8 += add_to_class(classbits, uchardptr, options, cd, 0, p[0] - 1);
3642 while (p[0] < NOTACHAR)
3643 {
3644 while (p[1] == p[0] + 1) p++;
3645 n8 += add_to_class(classbits, uchardptr, options, cd, p[0] + 1,
3646 (p[1] == NOTACHAR) ? (utf ? 0x10ffffu : 0xffffffffu) : p[1] - 1);
3647 p++;
3648 }
3649 return n8;
3650 }
3651
3652
3653
3654 /*************************************************
3655 * Compile one branch *
3656 *************************************************/
3657
3658 /* Scan the pattern, compiling it into the a vector. If the options are
3659 changed during the branch, the pointer is used to change the external options
3660 bits. This function is used during the pre-compile phase when we are trying
3661 to find out the amount of memory needed, as well as during the real compile
3662 phase. The value of lengthptr distinguishes the two phases.
3663
3664 Arguments:
3665 optionsptr pointer to the option bits
3666 codeptr points to the pointer to the current code point
3667 ptrptr points to the current pattern pointer
3668 errorcodeptr points to error code variable
3669 firstcharptr set to initial literal character, or < 0 (REQ_UNSET, REQ_NONE)
3670 reqcharptr set to the last literal character required, else < 0
3671 bcptr points to current branch chain
3672 cond_depth conditional nesting depth
3673 cd contains pointers to tables etc.
3674 lengthptr NULL during the real compile phase
3675 points to length accumulator during pre-compile phase
3676
3677 Returns: TRUE on success
3678 FALSE, with *errorcodeptr set non-zero on error
3679 */
3680
3681 static BOOL
3682 compile_branch(int *optionsptr, pcre_uchar **codeptr,
3683 const pcre_uchar **ptrptr, int *errorcodeptr, pcre_int32 *firstcharptr,
3684 pcre_int32 *reqcharptr, branch_chain *bcptr, int cond_depth,
3685 compile_data *cd, int *lengthptr)
3686 {
3687 int repeat_type, op_type;
3688 int repeat_min = 0, repeat_max = 0; /* To please picky compilers */
3689 int bravalue = 0;
3690 int greedy_default, greedy_non_default;
3691 pcre_int32 firstchar, reqchar;
3692 pcre_int32 zeroreqchar, zerofirstchar;
3693 pcre_int32 req_caseopt, reqvary, tempreqvary;
3694 int options = *optionsptr; /* May change dynamically */
3695 int after_manual_callout = 0;
3696 int length_prevgroup = 0;
3697 register pcre_uint32 c;
3698 int escape;
3699 register pcre_uchar *code = *codeptr;
3700 pcre_uchar *last_code = code;
3701 pcre_uchar *orig_code = code;
3702 pcre_uchar *tempcode;
3703 BOOL inescq = FALSE;
3704 BOOL groupsetfirstchar = FALSE;
3705 const pcre_uchar *ptr = *ptrptr;
3706 const pcre_uchar *tempptr;
3707 const pcre_uchar *nestptr = NULL;
3708 pcre_uchar *previous = NULL;
3709 pcre_uchar *previous_callout = NULL;
3710 pcre_uchar *save_hwm = NULL;
3711 pcre_uint8 classbits[32];
3712
3713 /* We can fish out the UTF-8 setting once and for all into a BOOL, but we
3714 must not do this for other options (e.g. PCRE_EXTENDED) because they may change
3715 dynamically as we process the pattern. */
3716
3717 #ifdef SUPPORT_UTF
3718 /* PCRE_UTF[16|32] have the same value as PCRE_UTF8. */
3719 BOOL utf = (options & PCRE_UTF8) != 0;
3720 pcre_uchar utf_chars[6];
3721 #else
3722 BOOL utf = FALSE;
3723 #endif
3724
3725 /* Helper variables for OP_XCLASS opcode (for characters > 255). We define
3726 class_uchardata always so that it can be passed to add_to_class() always,
3727 though it will not be used in non-UTF 8-bit cases. This avoids having to supply
3728 alternative calls for the different cases. */
3729
3730 pcre_uchar *class_uchardata;
3731 #if defined SUPPORT_UTF || !defined COMPILE_PCRE8
3732 BOOL xclass;
3733 pcre_uchar *class_uchardata_base;
3734 #endif
3735
3736 #ifdef PCRE_DEBUG
3737 if (lengthptr != NULL) DPRINTF((">> start branch\n"));
3738 #endif
3739
3740 /* Set up the default and non-default settings for greediness */
3741
3742 greedy_default = ((options & PCRE_UNGREEDY) != 0);
3743 greedy_non_default = greedy_default ^ 1;
3744
3745 /* Initialize no first byte, no required byte. REQ_UNSET means "no char
3746 matching encountered yet". It gets changed to REQ_NONE if we hit something that
3747 matches a non-fixed char first char; reqchar just remains unset if we never
3748 find one.
3749
3750 When we hit a repeat whose minimum is zero, we may have to adjust these values
3751 to take the zero repeat into account. This is implemented by setting them to
3752 zerofirstbyte and zeroreqchar when such a repeat is encountered. The individual
3753 item types that can be repeated set these backoff variables appropriately. */
3754
3755 firstchar = reqchar = zerofirstchar = zeroreqchar = REQ_UNSET;
3756
3757 /* The variable req_caseopt contains either the REQ_CASELESS value
3758 or zero, according to the current setting of the caseless flag. The
3759 REQ_CASELESS leaves the lower 28 bit empty. It is added into the
3760 firstchar or reqchar variables to record the case status of the
3761 value. This is used only for ASCII characters. */
3762
3763 req_caseopt = ((options & PCRE_CASELESS) != 0)? REQ_CASELESS:0;
3764
3765 /* Switch on next character until the end of the branch */
3766
3767 for (;; ptr++)
3768 {
3769 BOOL negate_class;
3770 BOOL should_flip_negation;
3771 BOOL possessive_quantifier;
3772 BOOL is_quantifier;
3773 BOOL is_recurse;
3774 BOOL reset_bracount;
3775 int class_has_8bitchar;
3776 int class_one_char;
3777 int newoptions;
3778 int recno;
3779 int refsign;
3780 int skipbytes;
3781 int subreqchar;
3782 int subfirstchar;
3783 int terminator;
3784 int mclength;
3785 int tempbracount;
3786 int ec; // FIXMEchpe pcre_uint32
3787 pcre_uchar mcbuffer[8];
3788
3789 /* Get next character in the pattern */
3790
3791 c = *ptr;
3792
3793 /* If we are at the end of a nested substitution, revert to the outer level
3794 string. Nesting only happens one level deep. */
3795
3796 if (c == 0 && nestptr != NULL)
3797 {
3798 ptr = nestptr;
3799 nestptr = NULL;
3800 c = *ptr;
3801 }
3802
3803 /* If we are in the pre-compile phase, accumulate the length used for the
3804 previous cycle of this loop. */
3805
3806 if (lengthptr != NULL)
3807 {
3808 #ifdef PCRE_DEBUG
3809 if (code > cd->hwm) cd->hwm = code; /* High water info */
3810 #endif
3811 if (code > cd->start_workspace + cd->workspace_size -
3812 WORK_SIZE_SAFETY_MARGIN) /* Check for overrun */
3813 {
3814 *errorcodeptr = ERR52;
3815 goto FAILED;
3816 }
3817
3818 /* There is at least one situation where code goes backwards: this is the
3819 case of a zero quantifier after a class (e.g. [ab]{0}). At compile time,
3820 the class is simply eliminated. However, it is created first, so we have to
3821 allow memory for it. Therefore, don't ever reduce the length at this point.
3822 */
3823
3824 if (code < last_code) code = last_code;
3825
3826 /* Paranoid check for integer overflow */
3827
3828 if (OFLOW_MAX - *lengthptr < code - last_code)
3829 {
3830 *errorcodeptr = ERR20;
3831 goto FAILED;
3832 }
3833
3834 *lengthptr += (int)(code - last_code);
3835 DPRINTF(("length=%d added %d c=%c (0x%x)\n", *lengthptr,
3836 (int)(code - last_code), c, c));
3837
3838 /* If "previous" is set and it is not at the start of the work space, move
3839 it back to there, in order to avoid filling up the work space. Otherwise,
3840 if "previous" is NULL, reset the current code pointer to the start. */
3841
3842 if (previous != NULL)
3843 {
3844 if (previous > orig_code)
3845 {
3846 memmove(orig_code, previous, IN_UCHARS(code - previous));
3847 code -= previous - orig_code;
3848 previous = orig_code;
3849 }
3850 }
3851 else code = orig_code;
3852
3853 /* Remember where this code item starts so we can pick up the length
3854 next time round. */
3855
3856 last_code = code;
3857 }
3858
3859 /* In the real compile phase, just check the workspace used by the forward
3860 reference list. */
3861
3862 else if (cd->hwm > cd->start_workspace + cd->workspace_size -
3863 WORK_SIZE_SAFETY_MARGIN)
3864 {
3865 *errorcodeptr = ERR52;
3866 goto FAILED;
3867 }
3868
3869 /* If in \Q...\E, check for the end; if not, we have a literal */
3870
3871 if (inescq && c != 0)
3872 {
3873 if (c == CHAR_BACKSLASH && ptr[1] == CHAR_E)
3874 {
3875 inescq = FALSE;
3876 ptr++;
3877 continue;
3878 }
3879 else
3880 {
3881 if (previous_callout != NULL)
3882 {
3883 if (lengthptr == NULL) /* Don't attempt in pre-compile phase */
3884 complete_callout(previous_callout, ptr, cd);
3885 previous_callout = NULL;
3886 }
3887 if ((options & PCRE_AUTO_CALLOUT) != 0)
3888 {
3889 previous_callout = code;
3890 code = auto_callout(code, ptr, cd);
3891 }
3892 goto NORMAL_CHAR;
3893 }
3894 }
3895
3896 /* Fill in length of a previous callout, except when the next thing is
3897 a quantifier. */
3898
3899 is_quantifier =
3900 c == CHAR_ASTERISK || c == CHAR_PLUS || c == CHAR_QUESTION_MARK ||
3901 (c == CHAR_LEFT_CURLY_BRACKET && is_counted_repeat(ptr+1));
3902
3903 if (!is_quantifier && previous_callout != NULL &&
3904 after_manual_callout-- <= 0)
3905 {
3906 if (lengthptr == NULL) /* Don't attempt in pre-compile phase */
3907 complete_callout(previous_callout, ptr, cd);
3908 previous_callout = NULL;
3909 }
3910
3911 /* In extended mode, skip white space and comments. */
3912
3913 if ((options & PCRE_EXTENDED) != 0)
3914 {
3915 if (MAX_255(*ptr) && (cd->ctypes[c] & ctype_space) != 0) continue;
3916 if (c == CHAR_NUMBER_SIGN)
3917 {
3918 ptr++;
3919 while (*ptr != 0)
3920 {
3921 if (IS_NEWLINE(ptr)) { ptr += cd->nllen - 1; break; }
3922 ptr++;
3923 #ifdef SUPPORT_UTF
3924 if (utf) FORWARDCHAR(ptr);
3925 #endif
3926 }
3927 if (*ptr != 0) continue;
3928
3929 /* Else fall through to handle end of string */
3930 c = 0;
3931 }
3932 }
3933
3934 /* No auto callout for quantifiers. */
3935
3936 if ((options & PCRE_AUTO_CALLOUT) != 0 && !is_quantifier)
3937 {
3938 previous_callout = code;
3939 code = auto_callout(code, ptr, cd);
3940 }
3941
3942 switch(c)
3943 {
3944 /* ===================================================================*/
3945 case 0: /* The branch terminates at string end */
3946 case CHAR_VERTICAL_LINE: /* or | or ) */
3947 case CHAR_RIGHT_PARENTHESIS:
3948 *firstcharptr = firstchar;
3949 *reqcharptr = reqchar;
3950 *codeptr = code;
3951 *ptrptr = ptr;
3952 if (lengthptr != NULL)
3953 {
3954 if (OFLOW_MAX - *lengthptr < code - last_code)
3955 {
3956 *errorcodeptr = ERR20;
3957 goto FAILED;
3958 }
3959 *lengthptr += (int)(code - last_code); /* To include callout length */
3960 DPRINTF((">> end branch\n"));
3961 }
3962 return TRUE;
3963
3964
3965 /* ===================================================================*/
3966 /* Handle single-character metacharacters. In multiline mode, ^ disables
3967 the setting of any following char as a first character. */
3968
3969 case CHAR_CIRCUMFLEX_ACCENT:
3970 previous = NULL;
3971 if ((options & PCRE_MULTILINE) != 0)
3972 {
3973 if (firstchar == REQ_UNSET) firstchar = REQ_NONE;
3974 *code++ = OP_CIRCM;
3975 }
3976 else *code++ = OP_CIRC;
3977 break;
3978
3979 case CHAR_DOLLAR_SIGN:
3980 previous = NULL;
3981 *code++ = ((options & PCRE_MULTILINE) != 0)? OP_DOLLM : OP_DOLL;
3982 break;
3983
3984 /* There can never be a first char if '.' is first, whatever happens about
3985 repeats. The value of reqchar doesn't change either. */
3986
3987 case CHAR_DOT:
3988 if (firstchar == REQ_UNSET) firstchar = REQ_NONE;
3989 zerofirstchar = firstchar;
3990 zeroreqchar = reqchar;
3991 previous = code;
3992 *code++ = ((options & PCRE_DOTALL) != 0)? OP_ALLANY: OP_ANY;
3993 break;
3994
3995
3996 /* ===================================================================*/
3997 /* Character classes. If the included characters are all < 256, we build a
3998 32-byte bitmap of the permitted characters, except in the special case
3999 where there is only one such character. For negated classes, we build the
4000 map as usual, then invert it at the end. However, we use a different opcode
4001 so that data characters > 255 can be handled correctly.
4002
4003 If the class contains characters outside the 0-255 range, a different
4004 opcode is compiled. It may optionally have a bit map for characters < 256,
4005 but those above are are explicitly listed afterwards. A flag byte tells
4006 whether the bitmap is present, and whether this is a negated class or not.
4007
4008 In JavaScript compatibility mode, an isolated ']' causes an error. In
4009 default (Perl) mode, it is treated as a data character. */
4010
4011 case CHAR_RIGHT_SQUARE_BRACKET:
4012 if ((cd->external_options & PCRE_JAVASCRIPT_COMPAT) != 0)
4013 {
4014 *errorcodeptr = ERR64;
4015 goto FAILED;
4016 }
4017 goto NORMAL_CHAR;
4018
4019 case CHAR_LEFT_SQUARE_BRACKET:
4020 previous = code;
4021
4022 /* PCRE supports POSIX class stuff inside a class. Perl gives an error if
4023 they are encountered at the top level, so we'll do that too. */
4024
4025 if ((ptr[1] == CHAR_COLON || ptr[1] == CHAR_DOT ||
4026 ptr[1] == CHAR_EQUALS_SIGN) &&
4027 check_posix_syntax(ptr, &tempptr))
4028 {
4029 *errorcodeptr = (ptr[1] == CHAR_COLON)? ERR13 : ERR31;
4030 goto FAILED;
4031 }
4032
4033 /* If the first character is '^', set the negation flag and skip it. Also,
4034 if the first few characters (either before or after ^) are \Q\E or \E we
4035 skip them too. This makes for compatibility with Perl. */
4036
4037 negate_class = FALSE;
4038 for (;;)
4039 {
4040 c = *(++ptr);
4041 if (c == CHAR_BACKSLASH)
4042 {
4043 if (ptr[1] == CHAR_E)
4044 ptr++;
4045 else if (STRNCMP_UC_C8(ptr + 1, STR_Q STR_BACKSLASH STR_E, 3) == 0)
4046 ptr += 3;
4047 else
4048 break;
4049 }
4050 else if (!negate_class && c == CHAR_CIRCUMFLEX_ACCENT)
4051 negate_class = TRUE;
4052 else break;
4053 }
4054
4055 /* Empty classes are allowed in JavaScript compatibility mode. Otherwise,
4056 an initial ']' is taken as a data character -- the code below handles
4057 that. In JS mode, [] must always fail, so generate OP_FAIL, whereas
4058 [^] must match any character, so generate OP_ALLANY. */
4059
4060 if (c == CHAR_RIGHT_SQUARE_BRACKET &&
4061 (cd->external_options & PCRE_JAVASCRIPT_COMPAT) != 0)
4062 {
4063 *code++ = negate_class? OP_ALLANY : OP_FAIL;
4064 if (firstchar == REQ_UNSET) firstchar = REQ_NONE;
4065 zerofirstchar = firstchar;
4066 break;
4067 }
4068
4069 /* If a class contains a negative special such as \S, we need to flip the
4070 negation flag at the end, so that support for characters > 255 works
4071 correctly (they are all included in the class). */
4072
4073 should_flip_negation = FALSE;
4074
4075 /* For optimization purposes, we track some properties of the class:
4076 class_has_8bitchar will be non-zero if the class contains at least one <
4077 256 character; class_one_char will be 1 if the class contains just one
4078 character. */
4079
4080 class_has_8bitchar = 0;
4081 class_one_char = 0;
4082
4083 /* Initialize the 32-char bit map to all zeros. We build the map in a
4084 temporary bit of memory, in case the class contains fewer than two
4085 8-bit characters because in that case the compiled code doesn't use the bit
4086 map. */
4087
4088 memset(classbits, 0, 32 * sizeof(pcre_uint8));
4089
4090 #if defined SUPPORT_UTF || !defined COMPILE_PCRE8
4091 xclass = FALSE;
4092 class_uchardata = code + LINK_SIZE + 2; /* For XCLASS items */
4093 class_uchardata_base = class_uchardata; /* Save the start */
4094 #endif
4095
4096 /* Process characters until ] is reached. By writing this as a "do" it
4097 means that an initial ] is taken as a data character. At the start of the
4098 loop, c contains the first byte of the character. */
4099
4100 if (c != 0) do
4101 {
4102 const pcre_uchar *oldptr;
4103
4104 #if defined SUPPORT_UTF && !defined COMPILE_PCRE32
4105 if (utf && HAS_EXTRALEN(c))
4106 { /* Braces are required because the */
4107 GETCHARLEN(c, ptr, ptr); /* macro generates multiple statements */
4108 }
4109 #endif
4110
4111 #if defined SUPPORT_UTF || !defined COMPILE_PCRE8
4112 /* In the pre-compile phase, accumulate the length of any extra
4113 data and reset the pointer. This is so that very large classes that
4114 contain a zillion > 255 characters no longer overwrite the work space
4115 (which is on the stack). We have to remember that there was XCLASS data,
4116 however. */
4117
4118 if (lengthptr != NULL && class_uchardata > class_uchardata_base)
4119 {
4120 xclass = TRUE;
4121 *lengthptr += class_uchardata - class_uchardata_base;
4122 class_uchardata = class_uchardata_base;
4123 }
4124 #endif
4125
4126 /* Inside \Q...\E everything is literal except \E */
4127
4128 if (inescq)
4129 {
4130 if (c == CHAR_BACKSLASH && ptr[1] == CHAR_E) /* If we are at \E */
4131 {
4132 inescq = FALSE; /* Reset literal state */
4133 ptr++; /* Skip the 'E' */
4134 continue; /* Carry on with next */
4135 }
4136 goto CHECK_RANGE; /* Could be range if \E follows */
4137 }
4138
4139 /* Handle POSIX class names. Perl allows a negation extension of the
4140 form [:^name:]. A square bracket that doesn't match the syntax is
4141 treated as a literal. We also recognize the POSIX constructions
4142 [.ch.] and [=ch=] ("collating elements") and fault them, as Perl
4143 5.6 and 5.8 do. */
4144
4145 if (c == CHAR_LEFT_SQUARE_BRACKET &&
4146 (ptr[1] == CHAR_COLON || ptr[1] == CHAR_DOT ||
4147 ptr[1] == CHAR_EQUALS_SIGN) && check_posix_syntax(ptr, &tempptr))
4148 {
4149 BOOL local_negate = FALSE;
4150 int posix_class, taboffset, tabopt;
4151 register const pcre_uint8 *cbits = cd->cbits;
4152 pcre_uint8 pbits[32];
4153
4154 if (ptr[1] != CHAR_COLON)
4155 {
4156 *errorcodeptr = ERR31;
4157 goto FAILED;
4158 }
4159
4160 ptr += 2;
4161 if (*ptr == CHAR_CIRCUMFLEX_ACCENT)
4162 {
4163 local_negate = TRUE;
4164 should_flip_negation = TRUE; /* Note negative special */
4165 ptr++;
4166 }
4167
4168 posix_class = check_posix_name(ptr, (int)(tempptr - ptr));
4169 if (posix_class < 0)
4170 {
4171 *errorcodeptr = ERR30;
4172 goto FAILED;
4173 }
4174
4175 /* If matching is caseless, upper and lower are converted to
4176 alpha. This relies on the fact that the class table starts with
4177 alpha, lower, upper as the first 3 entries. */
4178
4179 if ((options & PCRE_CASELESS) != 0 && posix_class <= 2)
4180 posix_class = 0;
4181
4182 /* When PCRE_UCP is set, some of the POSIX classes are converted to
4183 different escape sequences that use Unicode properties. */
4184
4185 #ifdef SUPPORT_UCP
4186 if ((options & PCRE_UCP) != 0)
4187 {
4188 int pc = posix_class + ((local_negate)? POSIX_SUBSIZE/2 : 0);
4189 if (posix_substitutes[pc] != NULL)
4190 {
4191 nestptr = tempptr + 1;
4192 ptr = posix_substitutes[pc] - 1;
4193 continue;
4194 }
4195 }
4196 #endif
4197 /* In the non-UCP case, we build the bit map for the POSIX class in a
4198 chunk of local store because we may be adding and subtracting from it,
4199 and we don't want to subtract bits that may be in the main map already.
4200 At the end we or the result into the bit map that is being built. */
4201
4202 posix_class *= 3;
4203
4204 /* Copy in the first table (always present) */
4205
4206 memcpy(pbits, cbits + posix_class_maps[posix_class],
4207 32 * sizeof(pcre_uint8));
4208
4209 /* If there is a second table, add or remove it as required. */
4210
4211 taboffset = posix_class_maps[posix_class + 1];
4212 tabopt = posix_class_maps[posix_class + 2];
4213
4214 if (taboffset >= 0)
4215 {
4216 if (tabopt >= 0)
4217 for (c = 0; c < 32; c++) pbits[c] |= cbits[c + taboffset];
4218 else
4219 for (c = 0; c < 32; c++) pbits[c] &= ~cbits[c + taboffset];
4220 }
4221
4222 /* Now see if we need to remove any special characters. An option
4223 value of 1 removes vertical space and 2 removes underscore. */
4224
4225 if (tabopt < 0) tabopt = -tabopt;
4226 if (tabopt == 1) pbits[1] &= ~0x3c;
4227 else if (tabopt == 2) pbits[11] &= 0x7f;
4228
4229 /* Add the POSIX table or its complement into the main table that is
4230 being built and we are done. */
4231
4232 if (local_negate)
4233 for (c = 0; c < 32; c++) classbits[c] |= ~pbits[c];
4234 else
4235 for (c = 0; c < 32; c++) classbits[c] |= pbits[c];
4236
4237 ptr = tempptr + 1;
4238 /* Every class contains at least one < 256 character. */
4239 class_has_8bitchar = 1;
4240 /* Every class contains at least two characters. */
4241 class_one_char = 2;
4242 continue; /* End of POSIX syntax handling */
4243 }
4244
4245 /* Backslash may introduce a single character, or it may introduce one
4246 of the specials, which just set a flag. The sequence \b is a special
4247 case. Inside a class (and only there) it is treated as backspace. We
4248 assume that other escapes have more than one character in them, so
4249 speculatively set both class_has_8bitchar and class_one_char bigger
4250 than one. Unrecognized escapes fall through and are either treated
4251 as literal characters (by default), or are faulted if
4252 PCRE_EXTRA is set. */
4253
4254 if (c == CHAR_BACKSLASH)
4255 {
4256 escape = check_escape(&ptr, &ec, errorcodeptr, cd->bracount, options, TRUE);
4257
4258 if (*errorcodeptr != 0) goto FAILED;
4259
4260 if (escape == 0)
4261 c = ec;
4262 else if (escape == ESC_b) c = CHAR_BS; /* \b is backspace in a class */
4263 else if (escape == ESC_N) /* \N is not supported in a class */
4264 {
4265 *errorcodeptr = ERR71;
4266 goto FAILED;
4267 }
4268 else if (escape == ESC_Q) /* Handle start of quoted string */
4269 {
4270 if (ptr[1] == CHAR_BACKSLASH && ptr[2] == CHAR_E)
4271 {
4272 ptr += 2; /* avoid empty string */
4273 }
4274 else inescq = TRUE;
4275 continue;
4276 }
4277 else if (escape == ESC_E) continue; /* Ignore orphan \E */
4278
4279 else
4280 {
4281 register const pcre_uint8 *cbits = cd->cbits;
4282 /* Every class contains at least two < 256 characters. */
4283 class_has_8bitchar++;
4284 /* Every class contains at least two characters. */
4285 class_one_char += 2;
4286
4287 switch (escape)
4288 {
4289 #ifdef SUPPORT_UCP
4290 case ESC_du: /* These are the values given for \d etc */
4291 case ESC_DU: /* when PCRE_UCP is set. We replace the */
4292 case ESC_wu: /* escape sequence with an appropriate \p */
4293 case ESC_WU: /* or \P to test Unicode properties instead */
4294 case ESC_su: /* of the default ASCII testing. */
4295 case ESC_SU:
4296 nestptr = ptr;
4297 ptr = substitutes[escape - ESC_DU] - 1; /* Just before substitute */
4298 class_has_8bitchar--; /* Undo! */
4299 continue;
4300 #endif
4301 case ESC_d:
4302 for (c = 0; c < 32; c++) classbits[c] |= cbits[c+cbit_digit];
4303 continue;
4304
4305 case ESC_D:
4306 should_flip_negation = TRUE;
4307 for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_digit];
4308 continue;
4309
4310 case ESC_w:
4311 for (c = 0; c < 32; c++) classbits[c] |= cbits[c+cbit_word];
4312 continue;
4313
4314 case ESC_W:
4315 should_flip_negation = TRUE;
4316 for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_word];
4317 continue;
4318
4319 /* Perl 5.004 onwards omits VT from \s, but we must preserve it
4320 if it was previously set by something earlier in the character
4321 class. Luckily, the value of CHAR_VT is 0x0b in both ASCII and
4322 EBCDIC, so we lazily just adjust the appropriate bit. */
4323
4324 case ESC_s:
4325 classbits[0] |= cbits[cbit_space];
4326 classbits[1] |= cbits[cbit_space+1] & ~0x08;
4327 for (c = 2; c < 32; c++) classbits[c] |= cbits[c+cbit_space];
4328 continue;
4329
4330 case ESC_S:
4331 should_flip_negation = TRUE;
4332 for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_space];
4333 classbits[1] |= 0x08; /* Perl 5.004 onwards omits VT from \s */
4334 continue;
4335
4336 /* The rest apply in both UCP and non-UCP cases. */
4337
4338 case ESC_h:
4339 (void)add_list_to_class(classbits, &class_uchardata, options, cd,
4340 PRIV(hspace_list), NOTACHAR);
4341 continue;
4342
4343 case ESC_H:
4344 (void)add_not_list_to_class(classbits, &class_uchardata, options,
4345 cd, PRIV(hspace_list));
4346 continue;
4347
4348 case ESC_v:
4349 (void)add_list_to_class(classbits, &class_uchardata, options, cd,
4350 PRIV(vspace_list), NOTACHAR);
4351 continue;
4352
4353 case ESC_V:
4354 (void)add_not_list_to_class(classbits, &class_uchardata, options,
4355 cd, PRIV(vspace_list));
4356 continue;
4357
4358 #ifdef SUPPORT_UCP
4359 case ESC_p:
4360 case ESC_P:
4361 {
4362 BOOL negated;
4363 int pdata;
4364 int ptype = get_ucp(&ptr, &negated, &pdata, errorcodeptr);
4365 if (ptype < 0) goto FAILED;
4366 *class_uchardata++ = ((escape == ESC_p) != negated)?
4367 XCL_PROP : XCL_NOTPROP;
4368 *class_uchardata++ = ptype;
4369 *class_uchardata++ = pdata;
4370 class_has_8bitchar--; /* Undo! */
4371 continue;
4372 }
4373 #endif
4374 /* Unrecognized escapes are faulted if PCRE is running in its
4375 strict mode. By default, for compatibility with Perl, they are
4376 treated as literals. */
4377
4378 default:
4379 if ((options & PCRE_EXTRA) != 0)
4380 {
4381 *errorcodeptr = ERR7;
4382 goto FAILED;
4383 }
4384 class_has_8bitchar--; /* Undo the speculative increase. */
4385 class_one_char -= 2; /* Undo the speculative increase. */
4386 c = *ptr; /* Get the final character and fall through */
4387 break;
4388 }
4389 }
4390
4391 /* Fall through if the escape just defined a single character (c >= 0).
4392 This may be greater than 256. */
4393
4394 escape = 0;
4395
4396 } /* End of backslash handling */
4397
4398 /* A character may be followed by '-' to form a range. However, Perl does
4399 not permit ']' to be the end of the range. A '-' character at the end is
4400 treated as a literal. Perl ignores orphaned \E sequences entirely. The
4401 code for handling \Q and \E is messy. */
4402
4403 CHECK_RANGE:
4404 while (ptr[1] == CHAR_BACKSLASH && ptr[2] == CHAR_E)
4405 {
4406 inescq = FALSE;
4407 ptr += 2;
4408 }
4409 oldptr = ptr;
4410
4411 /* Remember if \r or \n were explicitly used */
4412
4413 if (c == CHAR_CR || c == CHAR_NL) cd->external_flags |= PCRE_HASCRORLF;
4414
4415 /* Check for range */
4416
4417 if (!inescq && ptr[1] == CHAR_MINUS)
4418 {
4419 int d;
4420 ptr += 2;
4421 while (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_E) ptr += 2;
4422
4423 /* If we hit \Q (not followed by \E) at this point, go into escaped
4424 mode. */
4425
4426 while (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_Q)
4427 {
4428 ptr += 2;
4429 if (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_E)
4430 { ptr += 2; continue; }
4431 inescq = TRUE;
4432 break;
4433 }
4434
4435 /* Minus (hyphen) at the end of a class is treated as a literal, so put
4436 back the pointer and jump to handle the character that preceded it. */
4437
4438 if (*ptr == 0 || (!inescq && *ptr == CHAR_RIGHT_SQUARE_BRACKET))
4439 {
4440 ptr = oldptr;
4441 goto CLASS_SINGLE_CHARACTER;
4442 }
4443
4444 /* Otherwise, we have a potential range; pick up the next character */
4445
4446 #ifdef SUPPORT_UTF
4447 if (utf)
4448 { /* Braces are required because the */
4449 GETCHARLEN(d, ptr, ptr); /* macro generates multiple statements */
4450 }
4451 else
4452 #endif
4453 d = *ptr; /* Not UTF-8 mode */
4454
4455 /* The second part of a range can be a single-character escape, but
4456 not any of the other escapes. Perl 5.6 treats a hyphen as a literal
4457 in such circumstances. */
4458
4459 if (!inescq && d == CHAR_BACKSLASH)
4460 {
4461 int descape;
4462 descape = check_escape(&ptr, &d, errorcodeptr, cd->bracount, options, TRUE);
4463 if (*errorcodeptr != 0) goto FAILED;
4464
4465 /* \b is backspace; any other special means the '-' was literal. */
4466
4467 if (descape != 0)
4468 {
4469 if (descape == ESC_b) d = CHAR_BS; else
4470 {
4471 ptr = oldptr;
4472 goto CLASS_SINGLE_CHARACTER; /* A few lines below */
4473 }
4474 }
4475 }
4476
4477 /* Check that the two values are in the correct order. Optimize
4478 one-character ranges. */
4479
4480 if (d < c)
4481 {
4482 *errorcodeptr = ERR8;
4483 goto FAILED;
4484 }
4485 if (d == c) goto CLASS_SINGLE_CHARACTER; /* A few lines below */
4486
4487 /* We have found a character range, so single character optimizations
4488 cannot be done anymore. Any value greater than 1 indicates that there
4489 is more than one character. */
4490
4491 class_one_char = 2;
4492
4493 /* Remember an explicit \r or \n, and add the range to the class. */
4494
4495 if (d == CHAR_CR || d == CHAR_NL) cd->external_flags |= PCRE_HASCRORLF;
4496
4497 class_has_8bitchar +=
4498 add_to_class(classbits, &class_uchardata, options, cd, c, d);
4499
4500 continue; /* Go get the next char in the class */
4501 }
4502
4503 /* Handle a single character - we can get here for a normal non-escape
4504 char, or after \ that introduces a single character or for an apparent
4505 range that isn't. Only the value 1 matters for class_one_char, so don't
4506 increase it if it is already 2 or more ... just in case there's a class
4507 with a zillion characters in it. */
4508
4509 CLASS_SINGLE_CHARACTER:
4510 if (class_one_char < 2) class_one_char++;
4511
4512 /* If class_one_char is 1, we have the first single character in the
4513 class, and there have been no prior ranges, or XCLASS items generated by
4514 escapes. If this is the final character in the class, we can optimize by
4515 turning the item into a 1-character OP_CHAR[I] if it's positive, or
4516 OP_NOT[I] if it's negative. In the positive case, it can cause firstchar
4517 to be set. Otherwise, there can be no first char if this item is first,
4518 whatever repeat count may follow. In the case of reqchar, save the
4519 previous value for reinstating. */
4520
4521 if (class_one_char == 1 && ptr[1] == CHAR_RIGHT_SQUARE_BRACKET)
4522 {
4523 ptr++;
4524 zeroreqchar = reqchar;
4525
4526 if (negate_class)
4527 {
4528 #ifdef SUPPORT_UCP
4529 // FIXMEchpe pcreuint32?
4530 int d;
4531 #endif
4532 if (firstchar == REQ_UNSET) firstchar = REQ_NONE;
4533 zerofirstchar = firstchar;
4534
4535 /* For caseless UTF-8 mode when UCP support is available, check
4536 whether this character has more than one other case. If so, generate
4537 a special OP_NOTPROP item instead of OP_NOTI. */
4538
4539 #ifdef SUPPORT_UCP
4540 if (utf && (options & PCRE_CASELESS) != 0 &&
4541 (d = UCD_CASESET(c)) != 0)
4542 {
4543 *code++ = OP_NOTPROP;
4544 *code++ = PT_CLIST;
4545 *code++ = d;
4546 }
4547 else
4548 #endif
4549 /* Char has only one other case, or UCP not available */
4550
4551 {
4552 *code++ = ((options & PCRE_CASELESS) != 0)? OP_NOTI: OP_NOT;
4553 #if defined SUPPORT_UTF && !defined COMPILE_PCRE32
4554 if (utf && c > MAX_VALUE_FOR_SINGLE_CHAR)
4555 code += PRIV(ord2utf)(c, code);
4556 else
4557 #endif
4558 *code++ = c;
4559 }
4560
4561 /* We are finished with this character class */
4562
4563 goto END_CLASS;
4564 }
4565
4566 /* For a single, positive character, get the value into mcbuffer, and
4567 then we can handle this with the normal one-character code. */
4568
4569 #if defined SUPPORT_UTF && !defined COMPILE_PCRE32
4570 if (utf && c > MAX_VALUE_FOR_SINGLE_CHAR)
4571 mclength = PRIV(ord2utf)(c, mcbuffer);
4572 else
4573 #endif
4574 {
4575 mcbuffer[0] = c;
4576 mclength = 1;
4577 }
4578 goto ONE_CHAR;
4579 } /* End of 1-char optimization */
4580
4581 /* There is more than one character in the class, or an XCLASS item
4582 has been generated. Add this character to the class. */
4583
4584 class_has_8bitchar +=
4585 add_to_class(classbits, &class_uchardata, options, cd, c, c);
4586 }
4587
4588 /* Loop until ']' reached. This "while" is the end of the "do" far above.
4589 If we are at the end of an internal nested string, revert to the outer
4590 string. */
4591
4592 while (((c = *(++ptr)) != 0 ||
4593 (nestptr != NULL &&
4594 (ptr = nestptr, nestptr = NULL, c = *(++ptr)) != 0)) &&
4595 (c != CHAR_RIGHT_SQUARE_BRACKET || inescq));
4596
4597 /* Check for missing terminating ']' */
4598
4599 if (c == 0)
4600 {
4601 *errorcodeptr = ERR6;
4602 goto FAILED;
4603 }
4604
4605 /* We will need an XCLASS if data has been placed in class_uchardata. In
4606 the second phase this is a sufficient test. However, in the pre-compile
4607 phase, class_uchardata gets emptied to prevent workspace overflow, so it
4608 only if the very last character in the class needs XCLASS will it contain
4609 anything at this point. For this reason, xclass gets set TRUE above when
4610 uchar_classdata is emptied, and that's why this code is the way it is here
4611 instead of just doing a test on class_uchardata below. */
4612
4613 #if defined SUPPORT_UTF || !defined COMPILE_PCRE8
4614 if (class_uchardata > class_uchardata_base) xclass = TRUE;
4615 #endif
4616
4617 /* If this is the first thing in the branch, there can be no first char
4618 setting, whatever the repeat count. Any reqchar setting must remain
4619 unchanged after any kind of repeat. */
4620
4621 if (firstchar == REQ_UNSET) firstchar = REQ_NONE;
4622 zerofirstchar = firstchar;
4623 zeroreqchar = reqchar;
4624
4625 /* If there are characters with values > 255, we have to compile an
4626 extended class, with its own opcode, unless there was a negated special
4627 such as \S in the class, and PCRE_UCP is not set, because in that case all
4628 characters > 255 are in the class, so any that were explicitly given as
4629 well can be ignored. If (when there are explicit characters > 255 that must
4630 be listed) there are no characters < 256, we can omit the bitmap in the
4631 actual compiled code. */
4632
4633 #ifdef SUPPORT_UTF
4634 if (xclass && (!should_flip_negation || (options & PCRE_UCP) != 0))
4635 #elif !defined COMPILE_PCRE8
4636 if (xclass && !should_flip_negation)
4637 #endif
4638 #if defined SUPPORT_UTF || !defined COMPILE_PCRE8
4639 {
4640 *class_uchardata++ = XCL_END; /* Marks the end of extra data */
4641 *code++ = OP_XCLASS;
4642 code += LINK_SIZE;
4643 *code = negate_class? XCL_NOT:0;
4644
4645 /* If the map is required, move up the extra data to make room for it;
4646 otherwise just move the code pointer to the end of the extra data. */
4647
4648 if (class_has_8bitchar > 0)
4649 {
4650 *code++ |= XCL_MAP;
4651 memmove(code + (32 / sizeof(pcre_uchar)), code,
4652 IN_UCHARS(class_uchardata - code));
4653 memcpy(code, classbits, 32);
4654 code = class_uchardata + (32 / sizeof(pcre_uchar));
4655 }
4656 else code = class_uchardata;
4657
4658 /* Now fill in the complete length of the item */
4659
4660 PUT(previous, 1, (int)(code - previous));
4661 break; /* End of class handling */
4662 }
4663 #endif
4664
4665 /* If there are no characters > 255, or they are all to be included or
4666 excluded, set the opcode to OP_CLASS or OP_NCLASS, depending on whether the
4667 whole class was negated and whether there were negative specials such as \S
4668 (non-UCP) in the class. Then copy the 32-byte map into the code vector,
4669 negating it if necessary. */
4670
4671 *code++ = (negate_class == should_flip_negation) ? OP_CLASS : OP_NCLASS;
4672 if (lengthptr == NULL) /* Save time in the pre-compile phase */
4673 {
4674 if (negate_class)
4675 for (c = 0; c < 32; c++) classbits[c] = ~classbits[c];
4676 memcpy(code, classbits, 32);
4677 }
4678 code += 32 / sizeof(pcre_uchar);
4679
4680 END_CLASS:
4681 break;
4682
4683
4684 /* ===================================================================*/
4685 /* Various kinds of repeat; '{' is not necessarily a quantifier, but this
4686 has been tested above. */
4687
4688 case CHAR_LEFT_CURLY_BRACKET:
4689 if (!is_quantifier) goto NORMAL_CHAR;
4690 ptr = read_repeat_counts(ptr+1, &repeat_min, &repeat_max, errorcodeptr);
4691 if (*errorcodeptr != 0) goto FAILED;
4692 goto REPEAT;
4693
4694 case CHAR_ASTERISK:
4695 repeat_min = 0;
4696 repeat_max = -1;
4697 goto REPEAT;
4698
4699 case CHAR_PLUS:
4700 repeat_min = 1;
4701 repeat_max = -1;
4702 goto REPEAT;
4703
4704 case CHAR_QUESTION_MARK:
4705 repeat_min = 0;
4706 repeat_max = 1;
4707
4708 REPEAT:
4709 if (previous == NULL)
4710 {
4711 *errorcodeptr = ERR9;
4712 goto FAILED;
4713 }
4714
4715 if (repeat_min == 0)
4716 {
4717 firstchar = zerofirstchar; /* Adjust for zero repeat */
4718 reqchar = zeroreqchar; /* Ditto */
4719 }
4720
4721 /* Remember whether this is a variable length repeat */
4722
4723 reqvary = (repeat_min == repeat_max)? 0 : REQ_VARY;
4724
4725 op_type = 0; /* Default single-char op codes */
4726 possessive_quantifier = FALSE; /* Default not possessive quantifier */
4727
4728 /* Save start of previous item, in case we have to move it up in order to
4729 insert something before it. */
4730
4731 tempcode = previous;
4732
4733 /* If the next character is '+', we have a possessive quantifier. This
4734 implies greediness, whatever the setting of the PCRE_UNGREEDY option.
4735 If the next character is '?' this is a minimizing repeat, by default,
4736 but if PCRE_UNGREEDY is set, it works the other way round. We change the
4737 repeat type to the non-default. */
4738
4739 if (ptr[1] == CHAR_PLUS)
4740 {
4741 repeat_type = 0; /* Force greedy */
4742 possessive_quantifier = TRUE;
4743 ptr++;
4744 }
4745 else if (ptr[1] == CHAR_QUESTION_MARK)
4746 {
4747 repeat_type = greedy_non_default;
4748 ptr++;
4749 }
4750 else repeat_type = greedy_default;
4751
4752 /* If previous was a recursion call, wrap it in atomic brackets so that
4753 previous becomes the atomic group. All recursions were so wrapped in the
4754 past, but it no longer happens for non-repeated recursions. In fact, the
4755 repeated ones could be re-implemented independently so as not to need this,
4756 but for the moment we rely on the code for repeating groups. */
4757
4758 if (*previous == OP_RECURSE)
4759 {
4760 memmove(previous + 1 + LINK_SIZE, previous, IN_UCHARS(1 + LINK_SIZE));
4761 *previous = OP_ONCE;
4762 PUT(previous, 1, 2 + 2*LINK_SIZE);
4763 previous[2 + 2*LINK_SIZE] = OP_KET;
4764 PUT(previous, 3 + 2*LINK_SIZE, 2 + 2*LINK_SIZE);
4765 code += 2 + 2 * LINK_SIZE;
4766 length_prevgroup = 3 + 3*LINK_SIZE;
4767
4768 /* When actually compiling, we need to check whether this was a forward
4769 reference, and if so, adjust the offset. */
4770
4771 if (lengthptr == NULL && cd->hwm >= cd->start_workspace + LINK_SIZE)
4772 {
4773 int offset = GET(cd->hwm, -LINK_SIZE);
4774 if (offset == previous + 1 - cd->start_code)
4775 PUT(cd->hwm, -LINK_SIZE, offset + 1 + LINK_SIZE);
4776 }
4777 }
4778
4779 /* Now handle repetition for the different types of item. */
4780
4781 /* If previous was a character or negated character match, abolish the item
4782 and generate a repeat item instead. If a char item has a minimum of more
4783 than one, ensure that it is set in reqchar - it might not be if a sequence
4784 such as x{3} is the first thing in a branch because the x will have gone
4785 into firstchar instead. */
4786
4787 if (*previous == OP_CHAR || *previous == OP_CHARI
4788 || *previous == OP_NOT || *previous == OP_NOTI)
4789 {
4790 switch (*previous)
4791 {
4792 default: /* Make compiler happy. */
4793 case OP_CHAR: op_type = OP_STAR - OP_STAR; break;
4794 case OP_CHARI: op_type = OP_STARI - OP_STAR; break;
4795 case OP_NOT: op_type = OP_NOTSTAR - OP_STAR; break;
4796 case OP_NOTI: op_type = OP_NOTSTARI - OP_STAR; break;
4797 }
4798
4799 /* Deal with UTF characters that take up more than one character. It's
4800 easier to write this out separately than try to macrify it. Use c to
4801 hold the length of the character in bytes, plus UTF_LENGTH to flag that
4802 it's a length rather than a small character. */
4803
4804 #if defined SUPPORT_UTF && !defined COMPILE_PCRE32
4805 if (utf && NOT_FIRSTCHAR(code[-1]))
4806 {
4807 pcre_uchar *lastchar = code - 1;
4808 BACKCHAR(lastchar);
4809 c = (int)(code - lastchar); /* Length of UTF-8 character */
4810 memcpy(utf_chars, lastchar, IN_UCHARS(c)); /* Save the char */
4811 c |= UTF_LENGTH; /* Flag c as a length */
4812 }
4813 else
4814 #endif /* SUPPORT_UTF */
4815
4816 /* Handle the case of a single charater - either with no UTF support, or
4817 with UTF disabled, or for a single character UTF character. */
4818 {
4819 c = code[-1];
4820 if (*previous <= OP_CHARI && repeat_min > 1)
4821 reqchar = c | req_caseopt | cd->req_varyopt;
4822 }
4823
4824 /* If the repetition is unlimited, it pays to see if the next thing on
4825 the line is something that cannot possibly match this character. If so,
4826 automatically possessifying this item gains some performance in the case
4827 where the match fails. */
4828
4829 if (!possessive_quantifier &&
4830 repeat_max < 0 &&
4831 check_auto_possessive(previous, utf, ptr + 1, options, cd))
4832 {
4833 repeat_type = 0; /* Force greedy */
4834 possessive_quantifier = TRUE;
4835 }
4836
4837 goto OUTPUT_SINGLE_REPEAT; /* Code shared with single character types */
4838 }
4839
4840 /* If previous was a character type match (\d or similar), abolish it and
4841 create a suitable repeat item. The code is shared with single-character
4842 repeats by setting op_type to add a suitable offset into repeat_type. Note
4843 the the Unicode property types will be present only when SUPPORT_UCP is
4844 defined, but we don't wrap the little bits of code here because it just
4845 makes it horribly messy. */
4846
4847 else if (*previous < OP_EODN)
4848 {
4849 pcre_uchar *oldcode;
4850 int prop_type, prop_value;
4851 op_type = OP_TYPESTAR - OP_STAR; /* Use type opcodes */
4852 c = *previous;
4853
4854 if (!possessive_quantifier &&
4855 repeat_max < 0 &&
4856 check_auto_possessive(previous, utf, ptr + 1, options, cd))
4857 {
4858 repeat_type = 0; /* Force greedy */
4859 possessive_quantifier = TRUE;
4860 }
4861
4862 OUTPUT_SINGLE_REPEAT:
4863 if (*previous == OP_PROP || *previous == OP_NOTPROP)
4864 {
4865 prop_type = previous[1];
4866 prop_value = previous[2];
4867 }
4868 else prop_type = prop_value = -1;
4869
4870 oldcode = code;
4871 code = previous; /* Usually overwrite previous item */
4872
4873 /* If the maximum is zero then the minimum must also be zero; Perl allows
4874 this case, so we do too - by simply omitting the item altogether. */
4875
4876 if (repeat_max == 0) goto END_REPEAT;
4877
4878 /*--------------------------------------------------------------------*/
4879 /* This code is obsolete from release 8.00; the restriction was finally
4880 removed: */
4881
4882 /* All real repeats make it impossible to handle partial matching (maybe
4883 one day we will be able to remove this restriction). */
4884
4885 /* if (repeat_max != 1) cd->external_flags |= PCRE_NOPARTIAL; */
4886 /*--------------------------------------------------------------------*/
4887
4888 /* Combine the op_type with the repeat_type */
4889
4890 repeat_type += op_type;
4891
4892 /* A minimum of zero is handled either as the special case * or ?, or as
4893 an UPTO, with the maximum given. */
4894
4895 if (repeat_min == 0)
4896 {
4897 if (repeat_max == -1) *code++ = OP_STAR + repeat_type;
4898 else if (repeat_max == 1) *code++ = OP_QUERY + repeat_type;
4899 else
4900 {
4901 *code++ = OP_UPTO + repeat_type;
4902 PUT2INC(code, 0, repeat_max);
4903 }
4904 }
4905
4906 /* A repeat minimum of 1 is optimized into some special cases. If the
4907 maximum is unlimited, we use OP_PLUS. Otherwise, the original item is
4908 left in place and, if the maximum is greater than 1, we use OP_UPTO with
4909 one less than the maximum. */
4910
4911 else if (repeat_min == 1)
4912 {
4913 if (repeat_max == -1)
4914 *code++ = OP_PLUS + repeat_type;
4915 else
4916 {
4917 code = oldcode; /* leave previous item in place */
4918 if (repeat_max == 1) goto END_REPEAT;
4919 *code++ = OP_UPTO + repeat_type;
4920 PUT2INC(code, 0, repeat_max - 1);
4921 }
4922 }
4923
4924 /* The case {n,n} is just an EXACT, while the general case {n,m} is
4925 handled as an EXACT followed by an UPTO. */
4926
4927 else
4928 {
4929 *code++ = OP_EXACT + op_type; /* NB EXACT doesn't have repeat_type */
4930 PUT2INC(code, 0, repeat_min);
4931
4932 /* If the maximum is unlimited, insert an OP_STAR. Before doing so,
4933 we have to insert the character for the previous code. For a repeated
4934 Unicode property match, there are two extra bytes that define the
4935 required property. In UTF-8 mode, long characters have their length in
4936 c, with the UTF_LENGTH bit as a flag. */
4937
4938 if (repeat_max < 0)
4939 {
4940 #if defined SUPPORT_UTF && !defined COMPILE_PCRE32
4941 if (utf && (c & UTF_LENGTH) != 0)
4942 {
4943 memcpy(code, utf_chars, IN_UCHARS(c & 7));
4944 code += c & 7;
4945 }
4946 else
4947 #endif
4948 {
4949 *code++ = c;
4950 if (prop_type >= 0)
4951 {
4952 *code++ = prop_type;
4953 *code++ = prop_value;
4954 }
4955 }
4956 *code++ = OP_STAR + repeat_type;
4957 }
4958
4959 /* Else insert an UPTO if the max is greater than the min, again
4960 preceded by the character, for the previously inserted code. If the
4961 UPTO is just for 1 instance, we can use QUERY instead. */
4962
4963 else if (repeat_max != repeat_min)
4964 {
4965 #if defined SUPPORT_UTF && !defined COMPILE_PCRE32
4966 if (utf && (c & UTF_LENGTH) != 0)
4967 {
4968 memcpy(code, utf_chars, IN_UCHARS(c & 7));
4969 code += c & 7;
4970 }
4971 else
4972 #endif
4973 *code++ = c;
4974 if (prop_type >= 0)
4975 {
4976 *code++ = prop_type;
4977 *code++ = prop_value;
4978 }
4979 repeat_max -= repeat_min;
4980
4981 if (repeat_max == 1)
4982 {
4983 *code++ = OP_QUERY + repeat_type;
4984 }
4985 else
4986 {
4987 *code++ = OP_UPTO + repeat_type;
4988 PUT2INC(code, 0, repeat_max);
4989 }
4990 }
4991 }
4992
4993 /* The character or character type itself comes last in all cases. */
4994
4995 #if defined SUPPORT_UTF && !defined COMPILE_PCRE32
4996 if (utf && (c & UTF_LENGTH) != 0)
4997 {
4998 memcpy(code, utf_chars, IN_UCHARS(c & 7));
4999 code += c & 7;
5000 }
5001 else
5002 #endif
5003 *code++ = c;
5004
5005 /* For a repeated Unicode property match, there are two extra bytes that
5006 define the required property. */
5007
5008 #ifdef SUPPORT_UCP
5009 if (prop_type >= 0)
5010 {
5011 *code++ = prop_type;
5012 *code++ = prop_value;
5013 }
5014 #endif
5015 }
5016
5017 /* If previous was a character class or a back reference, we put the repeat
5018 stuff after it, but just skip the item if the repeat was {0,0}. */
5019
5020 else if (*previous == OP_CLASS ||
5021 *previous == OP_NCLASS ||
5022 #if defined SUPPORT_UTF || !defined COMPILE_PCRE8
5023 *previous == OP_XCLASS ||
5024 #endif
5025 *previous == OP_REF ||
5026 *previous == OP_REFI)
5027 {
5028 if (repeat_max == 0)
5029 {
5030 code = previous;
5031 goto END_REPEAT;
5032 }
5033
5034 /*--------------------------------------------------------------------*/
5035 /* This code is obsolete from release 8.00; the restriction was finally
5036 removed: */
5037
5038 /* All real repeats make it impossible to handle partial matching (maybe
5039 one day we will be able to remove this restriction). */
5040
5041 /* if (repeat_max != 1) cd->external_flags |= PCRE_NOPARTIAL; */
5042 /*--------------------------------------------------------------------*/
5043
5044 if (repeat_min == 0 && repeat_max == -1)
5045 *code++ = OP_CRSTAR + repeat_type;
5046 else if (repeat_min == 1 && repeat_max == -1)
5047 *code++ = OP_CRPLUS + repeat_type;
5048 else if (repeat_min == 0 && repeat_max == 1)
5049 *code++ = OP_CRQUERY + repeat_type;
5050 else
5051 {
5052 *code++ = OP_CRRANGE + repeat_type;
5053 PUT2INC(code, 0, repeat_min);
5054 if (repeat_max == -1) repeat_max = 0; /* 2-byte encoding for max */
5055 PUT2INC(code, 0, repeat_max);
5056 }
5057 }
5058
5059 /* If previous was a bracket group, we may have to replicate it in certain
5060 cases. Note that at this point we can encounter only the "basic" bracket
5061 opcodes such as BRA and CBRA, as this is the place where they get converted
5062 into the more special varieties such as BRAPOS and SBRA. A test for >=
5063 OP_ASSERT and <= OP_COND includes ASSERT, ASSERT_NOT, ASSERTBACK,
5064 ASSERTBACK_NOT, ONCE, BRA, CBRA, and COND. Originally, PCRE did not allow
5065 repetition of assertions, but now it does, for Perl compatibility. */
5066
5067 else if (*previous >= OP_ASSERT && *previous <= OP_COND)
5068 {
5069 register int i;
5070 int len = (int)(code - previous);
5071 pcre_uchar *bralink = NULL;
5072 pcre_uchar *brazeroptr = NULL;
5073
5074 /* Repeating a DEFINE group is pointless, but Perl allows the syntax, so
5075 we just ignore the repeat. */
5076
5077 if (*previous == OP_COND && previous[LINK_SIZE+1] == OP_DEF)
5078 goto END_REPEAT;
5079
5080 /* There is no sense in actually repeating assertions. The only potential
5081 use of repetition is in cases when the assertion is optional. Therefore,
5082 if the minimum is greater than zero, just ignore the repeat. If the
5083 maximum is not not zero or one, set it to 1. */
5084
5085 if (*previous < OP_ONCE) /* Assertion */
5086 {
5087 if (repeat_min > 0) goto END_REPEAT;
5088 if (repeat_max < 0 || repeat_max > 1) repeat_max = 1;
5089 }
5090
5091 /* The case of a zero minimum is special because of the need to stick
5092 OP_BRAZERO in front of it, and because the group appears once in the
5093 data, whereas in other cases it appears the minimum number of times. For
5094 this reason, it is simplest to treat this case separately, as otherwise
5095 the code gets far too messy. There are several special subcases when the
5096 minimum is zero. */
5097
5098 if (repeat_min == 0)
5099 {
5100 /* If the maximum is also zero, we used to just omit the group from the
5101 output altogether, like this:
5102
5103 ** if (repeat_max == 0)
5104 ** {
5105 ** code = previous;
5106 ** goto END_REPEAT;
5107 ** }
5108
5109 However, that fails when a group or a subgroup within it is referenced
5110 as a subroutine from elsewhere in the pattern, so now we stick in
5111 OP_SKIPZERO in front of it so that it is skipped on execution. As we
5112 don't have a list of which groups are referenced, we cannot do this
5113 selectively.
5114
5115 If the maximum is 1 or unlimited, we just have to stick in the BRAZERO
5116 and do no more at this point. However, we do need to adjust any
5117 OP_RECURSE calls inside the group that refer to the group itself or any
5118 internal or forward referenced group, because the offset is from the
5119 start of the whole regex. Temporarily terminate the pattern while doing
5120 this. */
5121
5122 if (repeat_max <= 1) /* Covers 0, 1, and unlimited */
5123 {
5124 *code = OP_END;
5125 adjust_recurse(previous, 1, utf, cd, save_hwm);
5126 memmove(previous + 1, previous, IN_UCHARS(len));
5127 code++;
5128 if (repeat_max == 0)
5129 {
5130 *previous++ = OP_SKIPZERO;
5131 goto END_REPEAT;
5132 }
5133 brazeroptr = previous; /* Save for possessive optimizing */
5134 *previous++ = OP_BRAZERO + repeat_type;
5135 }
5136
5137 /* If the maximum is greater than 1 and limited, we have to replicate
5138 in a nested fashion, sticking OP_BRAZERO before each set of brackets.
5139 The first one has to be handled carefully because it's the original
5140 copy, which has to be moved up. The remainder can be handled by code
5141 that is common with the non-zero minimum case below. We have to
5142 adjust the value or repeat_max, since one less copy is required. Once
5143 again, we may have to adjust any OP_RECURSE calls inside the group. */
5144
5145 else
5146 {
5147 int offset;
5148 *code = OP_END;
5149 adjust_recurse(previous, 2 + LINK_SIZE, utf, cd, save_hwm);
5150 memmove(previous + 2 + LINK_SIZE, previous, IN_UCHARS(len));
5151 code += 2 + LINK_SIZE;
5152 *previous++ = OP_BRAZERO + repeat_type;
5153 *previous++ = OP_BRA;
5154
5155 /* We chain together the bracket offset fields that have to be
5156 filled in later when the ends of the brackets are reached. */
5157
5158 offset = (bralink == NULL)? 0 : (int)(previous - bralink);
5159 bralink = previous;
5160 PUTINC(previous, 0, offset);
5161 }
5162
5163 repeat_max--;
5164 }
5165
5166 /* If the minimum is greater than zero, replicate the group as many
5167 times as necessary, and adjust the maximum to the number of subsequent
5168 copies that we need. If we set a first char from the group, and didn't
5169 set a required char, copy the latter from the former. If there are any
5170 forward reference subroutine calls in the group, there will be entries on
5171 the workspace list; replicate these with an appropriate increment. */
5172
5173 else
5174 {
5175 if (repeat_min > 1)
5176 {
5177 /* In the pre-compile phase, we don't actually do the replication. We
5178 just adjust the length as if we had. Do some paranoid checks for
5179 potential integer overflow. The INT64_OR_DOUBLE type is a 64-bit
5180 integer type when available, otherwise double. */
5181
5182 if (lengthptr != NULL)
5183 {
5184 int delta = (repeat_min - 1)*length_prevgroup;
5185 if ((INT64_OR_DOUBLE)(repeat_min - 1)*
5186 (INT64_OR_DOUBLE)length_prevgroup >
5187 (INT64_OR_DOUBLE)INT_MAX ||
5188 OFLOW_MAX - *lengthptr < delta)
5189 {
5190 *errorcodeptr = ERR20;
5191 goto FAILED;
5192 }
5193 *lengthptr += delta;
5194 }
5195
5196 /* This is compiling for real. If there is a set first byte for
5197 the group, and we have not yet set a "required byte", set it. Make
5198 sure there is enough workspace for copying forward references before
5199 doing the copy. */
5200
5201 else
5202 {
5203 if (groupsetfirstchar && reqchar < 0) reqchar = firstchar;
5204
5205 for (i = 1; i < repeat_min; i++)
5206 {
5207 pcre_uchar *hc;
5208 pcre_uchar *this_hwm = cd->hwm;
5209 memcpy(code, previous, IN_UCHARS(len));
5210
5211 while (cd->hwm > cd->start_workspace + cd->workspace_size -
5212 WORK_SIZE_SAFETY_MARGIN - (this_hwm - save_hwm))
5213 {
5214 int save_offset = save_hwm - cd->start_workspace;
5215 int this_offset = this_hwm - cd->start_workspace;
5216 *errorcodeptr = expand_workspace(cd);
5217 if (*errorcodeptr != 0) goto FAILED;
5218 save_hwm = (pcre_uchar *)cd->start_workspace + save_offset;
5219 this_hwm = (pcre_uchar *)cd->start_workspace + this_offset;
5220 }
5221
5222 for (hc = save_hwm; hc < this_hwm; hc += LINK_SIZE)
5223 {
5224 PUT(cd->hwm, 0, GET(hc, 0) + len);
5225 cd->hwm += LINK_SIZE;
5226 }
5227 save_hwm = this_hwm;
5228 code += len;
5229 }
5230 }
5231 }
5232
5233 if (repeat_max > 0) repeat_max -= repeat_min;
5234 }
5235
5236 /* This code is common to both the zero and non-zero minimum cases. If
5237 the maximum is limited, it replicates the group in a nested fashion,
5238 remembering the bracket starts on a stack. In the case of a zero minimum,
5239 the first one was set up above. In all cases the repeat_max now specifies
5240 the number of additional copies needed. Again, we must remember to
5241 replicate entries on the forward reference list. */
5242
5243 if (repeat_max >= 0)
5244 {
5245 /* In the pre-compile phase, we don't actually do the replication. We
5246 just adjust the length as if we had. For each repetition we must add 1
5247 to the length for BRAZERO and for all but the last repetition we must
5248 add 2 + 2*LINKSIZE to allow for the nesting that occurs. Do some
5249 paranoid checks to avoid integer overflow. The INT64_OR_DOUBLE type is
5250 a 64-bit integer type when available, otherwise double. */
5251
5252 if (lengthptr != NULL && repeat_max > 0)
5253 {
5254 int delta = repeat_max * (length_prevgroup + 1 + 2 + 2*LINK_SIZE) -
5255 2 - 2*LINK_SIZE; /* Last one doesn't nest */
5256 if ((INT64_OR_DOUBLE)repeat_max *
5257 (INT64_OR_DOUBLE)(length_prevgroup + 1 + 2 + 2*LINK_SIZE)
5258 > (INT64_OR_DOUBLE)INT_MAX ||
5259 OFLOW_MAX - *lengthptr < delta)
5260 {
5261 *errorcodeptr = ERR20;
5262 goto FAILED;
5263 }
5264 *lengthptr += delta;
5265 }
5266
5267 /* This is compiling for real */
5268
5269 else for (i = repeat_max - 1; i >= 0; i--)
5270 {
5271 pcre_uchar *hc;
5272 pcre_uchar *this_hwm = cd->hwm;
5273
5274 *code++ = OP_BRAZERO + repeat_type;
5275
5276 /* All but the final copy start a new nesting, maintaining the
5277 chain of brackets outstanding. */
5278
5279 if (i != 0)
5280 {
5281 int offset;
5282 *code++ = OP_BRA;
5283 offset = (bralink == NULL)? 0 : (int)(code - bralink);
5284 bralink = code;
5285 PUTINC(code, 0, offset);
5286 }
5287
5288 memcpy(code, previous, IN_UCHARS(len));
5289
5290 /* Ensure there is enough workspace for forward references before
5291 copying them. */
5292
5293 while (cd->hwm > cd->start_workspace + cd->workspace_size -
5294 WORK_SIZE_SAFETY_MARGIN - (this_hwm - save_hwm))
5295 {
5296 int save_offset = save_hwm - cd->start_workspace;
5297 int this_offset = this_hwm - cd->start_workspace;
5298 *errorcodeptr = expand_workspace(cd);
5299 if (*errorcodeptr != 0) goto FAILED;
5300 save_hwm = (pcre_uchar *)cd->start_workspace + save_offset;
5301 this_hwm = (pcre_uchar *)cd->start_workspace + this_offset;
5302 }
5303
5304 for (hc = save_hwm; hc < this_hwm; hc += LINK_SIZE)
5305 {
5306 PUT(cd->hwm, 0, GET(hc, 0) + len + ((i != 0)? 2+LINK_SIZE : 1));
5307 cd->hwm += LINK_SIZE;
5308 }
5309 save_hwm = this_hwm;
5310 code += len;
5311 }
5312
5313 /* Now chain through the pending brackets, and fill in their length
5314 fields (which are holding the chain links pro tem). */
5315
5316 while (bralink != NULL)
5317 {
5318 int oldlinkoffset;
5319 int offset = (int)(code - bralink + 1);
5320 pcre_uchar *bra = code - offset;
5321 oldlinkoffset = GET(bra, 1);
5322 bralink = (oldlinkoffset == 0)? NULL : bralink - oldlinkoffset;
5323 *code++ = OP_KET;
5324 PUTINC(code, 0, offset);
5325 PUT(bra, 1, offset);
5326 }
5327 }
5328
5329 /* If the maximum is unlimited, set a repeater in the final copy. For
5330 ONCE brackets, that's all we need to do. However, possessively repeated
5331 ONCE brackets can be converted into non-capturing brackets, as the
5332 behaviour of (?:xx)++ is the same as (?>xx)++ and this saves having to
5333 deal with possessive ONCEs specially.
5334
5335 Otherwise, when we are doing the actual compile phase, check to see
5336 whether this group is one that could match an empty string. If so,
5337 convert the initial operator to the S form (e.g. OP_BRA -> OP_SBRA) so
5338 that runtime checking can be done. [This check is also applied to ONCE
5339 groups at runtime, but in a different way.]
5340
5341 Then, if the quantifier was possessive and the bracket is not a
5342 conditional, we convert the BRA code to the POS form, and the KET code to
5343 KETRPOS. (It turns out to be convenient at runtime to detect this kind of
5344 subpattern at both the start and at the end.) The use of special opcodes
5345 makes it possible to reduce greatly the stack usage in pcre_exec(). If
5346 the group is preceded by OP_BRAZERO, convert this to OP_BRAPOSZERO.
5347
5348 Then, if the minimum number of matches is 1 or 0, cancel the possessive
5349 flag so that the default action below, of wrapping everything inside
5350 atomic brackets, does not happen. When the minimum is greater than 1,
5351 there will be earlier copies of the group, and so we still have to wrap
5352 the whole thing. */
5353
5354 else
5355 {
5356 pcre_uchar *ketcode = code - 1 - LINK_SIZE;
5357 pcre_uchar *bracode = ketcode - GET(ketcode, 1);
5358
5359 /* Convert possessive ONCE brackets to non-capturing */
5360
5361 if ((*bracode == OP_ONCE || *bracode == OP_ONCE_NC) &&
5362 possessive_quantifier) *bracode = OP_BRA;
5363
5364 /* For non-possessive ONCE brackets, all we need to do is to
5365 set the KET. */
5366
5367 if (*bracode == OP_ONCE || *bracode == OP_ONCE_NC)
5368 *ketcode = OP_KETRMAX + repeat_type;
5369
5370 /* Handle non-ONCE brackets and possessive ONCEs (which have been
5371 converted to non-capturing above). */
5372
5373 else
5374 {
5375 /* In the compile phase, check for empty string matching. */
5376
5377 if (lengthptr == NULL)
5378 {
5379 pcre_uchar *scode = bracode;
5380 do
5381 {
5382 if (could_be_empty_branch(scode, ketcode, utf, cd))
5383 {
5384 *bracode += OP_SBRA - OP_BRA;
5385 break;
5386 }
5387 scode += GET(scode, 1);
5388 }
5389 while (*scode == OP_ALT);
5390 }
5391
5392 /* Handle possessive quantifiers. */
5393
5394 if (possessive_quantifier)
5395 {
5396 /* For COND brackets, we wrap the whole thing in a possessively
5397 repeated non-capturing bracket, because we have not invented POS
5398 versions of the COND opcodes. Because we are moving code along, we
5399 must ensure that any pending recursive references are updated. */
5400
5401 if (*bracode == OP_COND || *bracode == OP_SCOND)
5402 {
5403 int nlen = (int)(code - bracode);
5404 *code = OP_END;
5405 adjust_recurse(bracode, 1 + LINK_SIZE, utf, cd, save_hwm);
5406 memmove(bracode + 1 + LINK_SIZE, bracode, IN_UCHARS(nlen));
5407 code += 1 + LINK_SIZE;
5408 nlen += 1 + LINK_SIZE;
5409 *bracode = OP_BRAPOS;
5410 *code++ = OP_KETRPOS;
5411 PUTINC(code, 0, nlen);
5412 PUT(bracode, 1, nlen);
5413 }
5414
5415 /* For non-COND brackets, we modify the BRA code and use KETRPOS. */
5416
5417 else
5418 {
5419 *bracode += 1; /* Switch to xxxPOS opcodes */
5420 *ketcode = OP_KETRPOS;
5421 }
5422
5423 /* If the minimum is zero, mark it as possessive, then unset the
5424 possessive flag when the minimum is 0 or 1. */
5425
5426 if (brazeroptr != NULL) *brazeroptr = OP_BRAPOSZERO;
5427 if (repeat_min < 2) possessive_quantifier = FALSE;
5428 }
5429
5430 /* Non-possessive quantifier */
5431
5432 else *ketcode = OP_KETRMAX + repeat_type;
5433 }
5434 }
5435 }
5436
5437 /* If previous is OP_FAIL, it was generated by an empty class [] in
5438 JavaScript mode. The other ways in which OP_FAIL can be generated, that is
5439 by (*FAIL) or (?!) set previous to NULL, which gives a "nothing to repeat"
5440 error above. We can just ignore the repeat in JS case. */
5441
5442 else if (*previous == OP_FAIL) goto END_REPEAT;
5443
5444 /* Else there's some kind of shambles */
5445
5446 else
5447 {
5448 *errorcodeptr = ERR11;
5449 goto FAILED;
5450 }
5451
5452 /* If the character following a repeat is '+', or if certain optimization
5453 tests above succeeded, possessive_quantifier is TRUE. For some opcodes,
5454 there are special alternative opcodes for this case. For anything else, we
5455 wrap the entire repeated item inside OP_ONCE brackets. Logically, the '+'
5456 notation is just syntactic sugar, taken from Sun's Java package, but the
5457 special opcodes can optimize it.
5458
5459 Some (but not all) possessively repeated subpatterns have already been
5460 completely handled in the code just above. For them, possessive_quantifier
5461 is always FALSE at this stage.
5462
5463 Note that the repeated item starts at tempcode, not at previous, which
5464 might be the first part of a string whose (former) last char we repeated.
5465
5466 Possessifying an 'exact' quantifier has no effect, so we can ignore it. But
5467 an 'upto' may follow. We skip over an 'exact' item, and then test the
5468 length of what remains before proceeding. */
5469
5470 if (possessive_quantifier)
5471 {
5472 int len;
5473
5474 if (*tempcode == OP_TYPEEXACT)
5475 tempcode += PRIV(OP_lengths)[*tempcode] +
5476 ((tempcode[1 + IMM2_SIZE] == OP_PROP
5477 || tempcode[1 + IMM2_SIZE] == OP_NOTPROP)? 2 : 0);
5478
5479 else if (*tempcode == OP_EXACT || *tempcode == OP_NOTEXACT)
5480 {
5481 tempcode += PRIV(OP_lengths)[*tempcode];
5482 #if defined SUPPORT_UTF && !defined COMPILE_PCRE32
5483 if (utf && HAS_EXTRALEN(tempcode[-1]))
5484 tempcode += GET_EXTRALEN(tempcode[-1]);
5485 #endif
5486 }
5487
5488 len = (int)(code - tempcode);
5489 if (len > 0) switch (*tempcode)
5490 {
5491 case OP_STAR: *tempcode = OP_POSSTAR; break;
5492 case OP_PLUS: *tempcode = OP_POSPLUS; break;
5493 case OP_QUERY: *tempcode = OP_POSQUERY; break;
5494 case OP_UPTO: *tempcode = OP_POSUPTO; break;
5495
5496 case OP_STARI: *tempcode = OP_POSSTARI; break;
5497 case OP_PLUSI: *tempcode = OP_POSPLUSI; break;
5498 case OP_QUERYI: *tempcode = OP_POSQUERYI; break;
5499 case OP_UPTOI: *tempcode = OP_POSUPTOI; break;
5500
5501 case OP_NOTSTAR: *tempcode = OP_NOTPOSSTAR; break;
5502 case OP_NOTPLUS: *tempcode = OP_NOTPOSPLUS; break;
5503 case OP_NOTQUERY: *tempcode = OP_NOTPOSQUERY; break;
5504 case OP_NOTUPTO: *tempcode = OP_NOTPOSUPTO; break;
5505
5506 case OP_NOTSTARI: *tempcode = OP_NOTPOSSTARI; break;
5507 case OP_NOTPLUSI: *tempcode = OP_NOTPOSPLUSI; break;
5508 case OP_NOTQUERYI: *tempcode = OP_NOTPOSQUERYI; break;
5509 case OP_NOTUPTOI: *tempcode = OP_NOTPOSUPTOI; break;
5510
5511 case OP_TYPESTAR: *tempcode = OP_TYPEPOSSTAR; break;
5512 case OP_TYPEPLUS: *tempcode = OP_TYPEPOSPLUS; break;
5513 case OP_TYPEQUERY: *tempcode = OP_TYPEPOSQUERY; break;
5514 case OP_TYPEUPTO: *tempcode = OP_TYPEPOSUPTO; break;
5515
5516 /* Because we are moving code along, we must ensure that any
5517 pending recursive references are updated. */
5518
5519 default:
5520 *code = OP_END;
5521 adjust_recurse(tempcode, 1 + LINK_SIZE, utf, cd, save_hwm);
5522 memmove(tempcode + 1 + LINK_SIZE, tempcode, IN_UCHARS(len));
5523 code += 1 + LINK_SIZE;
5524 len += 1 + LINK_SIZE;
5525 tempcode[0] = OP_ONCE;
5526 *code++ = OP_KET;
5527 PUTINC(code, 0, len);
5528 PUT(tempcode, 1, len);
5529 break;
5530 }
5531 }
5532
5533 /* In all case we no longer have a previous item. We also set the
5534 "follows varying string" flag for subsequently encountered reqchars if
5535 it isn't already set and we have just passed a varying length item. */
5536
5537 END_REPEAT:
5538 previous = NULL;
5539 cd->req_varyopt |= reqvary;
5540 break;
5541
5542
5543 /* ===================================================================*/
5544 /* Start of nested parenthesized sub-expression, or comment or lookahead or
5545 lookbehind or option setting or condition or all the other extended
5546 parenthesis forms. */
5547
5548 case CHAR_LEFT_PARENTHESIS:
5549 newoptions = options;
5550 skipbytes = 0;
5551 bravalue = OP_CBRA;
5552 save_hwm = cd->hwm;
5553 reset_bracount = FALSE;
5554
5555 /* First deal with various "verbs" that can be introduced by '*'. */
5556
5557 ptr++;
5558 if (ptr[0] == CHAR_ASTERISK && (ptr[1] == ':'
5559 || (MAX_255(ptr[1]) && ((cd->ctypes[ptr[1]] & ctype_letter) != 0))))
5560 {
5561 int i, namelen;
5562 int arglen = 0;
5563 const char *vn = verbnames;
5564 const pcre_uchar *name = ptr + 1;
5565 const pcre_uchar *arg = NULL;
5566 previous = NULL;
5567 ptr++;
5568 while (MAX_255(*ptr) && (cd->ctypes[*ptr] & ctype_letter) != 0) ptr++;
5569 namelen = (int)(ptr - name);
5570
5571 /* It appears that Perl allows any characters whatsoever, other than
5572 a closing parenthesis, to appear in arguments, so we no longer insist on
5573 letters, digits, and underscores. */
5574
5575 if (*ptr == CHAR_COLON)
5576 {
5577 arg = ++ptr;
5578 while (*ptr != 0 && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++;
5579 arglen = (int)(ptr - arg);
5580 if ((unsigned int)arglen > MAX_MARK)
5581 {
5582 *errorcodeptr = ERR75;
5583 goto FAILED;
5584 }
5585 }
5586
5587 if (*ptr != CHAR_RIGHT_PARENTHESIS)
5588 {
5589 *errorcodeptr = ERR60;
5590 goto FAILED;
5591 }
5592
5593 /* Scan the table of verb names */
5594
5595 for (i = 0; i < verbcount; i++)
5596 {
5597 if (namelen == verbs[i].len &&
5598 STRNCMP_UC_C8(name, vn, namelen) == 0)
5599 {
5600 int setverb;
5601
5602 /* Check for open captures before ACCEPT and convert it to
5603 ASSERT_ACCEPT if in an assertion. */
5604
5605 if (verbs[i].op == OP_ACCEPT)
5606 {
5607 open_capitem *oc;
5608 if (arglen != 0)
5609 {
5610 *errorcodeptr = ERR59;
5611 goto FAILED;
5612 }
5613 cd->had_accept = TRUE;
5614 for (oc = cd->open_caps; oc != NULL; oc = oc->next)
5615 {
5616 *code++ = OP_CLOSE;
5617 PUT2INC(code, 0, oc->number);
5618 }
5619 setverb = *code++ =
5620 (cd->assert_depth > 0)? OP_ASSERT_ACCEPT : OP_ACCEPT;
5621
5622 /* Do not set firstchar after *ACCEPT */
5623 if (firstchar == REQ_UNSET) firstchar = REQ_NONE;
5624 }
5625
5626 /* Handle other cases with/without an argument */
5627
5628 else if (arglen == 0)
5629 {
5630 if (verbs[i].op < 0) /* Argument is mandatory */
5631 {
5632 *errorcodeptr = ERR66;
5633 goto FAILED;
5634 }
5635 setverb = *code++ = verbs[i].op;
5636 }
5637
5638 else
5639 {
5640 if (verbs[i].op_arg < 0) /* Argument is forbidden */
5641 {
5642 *errorcodeptr = ERR59;
5643 goto FAILED;
5644 }
5645 setverb = *code++ = verbs[i].op_arg;
5646 *code++ = arglen;
5647 memcpy(code, arg, IN_UCHARS(arglen));
5648 code += arglen;
5649 *code++ = 0;
5650 }
5651
5652 switch (setverb)
5653 {
5654 case OP_THEN:
5655 case OP_THEN_ARG:
5656 cd->external_flags |= PCRE_HASTHEN;
5657 break;
5658
5659 case OP_PRUNE:
5660 case OP_PRUNE_ARG:
5661 case OP_SKIP:
5662 case OP_SKIP_ARG:
5663 cd->had_pruneorskip = TRUE;
5664 break;
5665 }
5666
5667 break; /* Found verb, exit loop */
5668 }
5669
5670 vn += verbs[i].len + 1;
5671 }
5672
5673 if (i < verbcount) continue; /* Successfully handled a verb */
5674 *errorcodeptr = ERR60; /* Verb not recognized */
5675 goto FAILED;
5676 }
5677
5678 /* Deal with the extended parentheses; all are introduced by '?', and the
5679 appearance of any of them means that this is not a capturing group. */
5680
5681 else if (*ptr == CHAR_QUESTION_MARK)
5682 {
5683 int i, set, unset, namelen;
5684 int *optset;
5685 const pcre_uchar *name;
5686 pcre_uchar *slot;
5687
5688 switch (*(++ptr))
5689 {
5690 case CHAR_NUMBER_SIGN: /* Comment; skip to ket */
5691 ptr++;
5692 while (*ptr != 0 && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++;
5693 if (*ptr == 0)
5694 {
5695 *errorcodeptr = ERR18;
5696 goto FAILED;
5697 }
5698 continue;
5699
5700
5701 /* ------------------------------------------------------------ */
5702 case CHAR_VERTICAL_LINE: /* Reset capture count for each branch */
5703 reset_bracount = TRUE;
5704 /* Fall through */
5705
5706 /* ------------------------------------------------------------ */
5707 case CHAR_COLON: /* Non-capturing bracket */
5708 bravalue = OP_BRA;
5709 ptr++;
5710 break;
5711
5712
5713 /* ------------------------------------------------------------ */
5714 case CHAR_LEFT_PARENTHESIS:
5715 bravalue = OP_COND; /* Conditional group */
5716
5717 /* A condition can be an assertion, a number (referring to a numbered
5718 group), a name (referring to a named group), or 'R', referring to
5719 recursion. R<digits> and R&name are also permitted for recursion tests.
5720
5721 There are several syntaxes for testing a named group: (?(name)) is used
5722 by Python; Perl 5.10 onwards uses (?(<name>) or (?('name')).
5723
5724 There are two unfortunate ambiguities, caused by history. (a) 'R' can
5725 be the recursive thing or the name 'R' (and similarly for 'R' followed
5726 by digits), and (b) a number could be a name that consists of digits.
5727 In both cases, we look for a name first; if not found, we try the other
5728 cases. */
5729
5730 /* For conditions that are assertions, check the syntax, and then exit
5731 the switch. This will take control down to where bracketed groups,
5732 including assertions, are processed. */
5733
5734 if (ptr[1] == CHAR_QUESTION_MARK && (ptr[2] == CHAR_EQUALS_SIGN ||
5735 ptr[2] == CHAR_EXCLAMATION_MARK || ptr[2] == CHAR_LESS_THAN_SIGN))
5736 break;
5737
5738 /* Most other conditions use OP_CREF (a couple change to OP_RREF
5739 below), and all need to skip 1+IMM2_SIZE bytes at the start of the group. */
5740
5741 code[1+LINK_SIZE] = OP_CREF;
5742 skipbytes = 1+IMM2_SIZE;
5743 refsign = -1;
5744
5745 /* Check for a test for recursion in a named group. */
5746
5747 if (ptr[1] == CHAR_R && ptr[2] == CHAR_AMPERSAND)
5748 {
5749 terminator = -1;
5750 ptr += 2;
5751 code[1+LINK_SIZE] = OP_RREF; /* Change the type of test */
5752 }
5753
5754 /* Check for a test for a named group's having been set, using the Perl
5755 syntax (?(<name>) or (?('name') */
5756
5757 else if (ptr[1] == CHAR_LESS_THAN_SIGN)
5758 {
5759 terminator = CHAR_GREATER_THAN_SIGN;
5760 ptr++;
5761 }
5762 else if (ptr[1] == CHAR_APOSTROPHE)
5763 {
5764 terminator = CHAR_APOSTROPHE;
5765 ptr++;
5766 }
5767 else
5768 {
5769 terminator = 0;
5770 if (ptr[1] == CHAR_MINUS || ptr[1] == CHAR_PLUS) refsign = *(++ptr);
5771 }
5772
5773 /* We now expect to read a name; any thing else is an error */
5774
5775 if (!MAX_255(ptr[1]) || (cd->ctypes[ptr[1]] & ctype_word) == 0)
5776 {
5777 ptr += 1; /* To get the right offset */
5778 *errorcodeptr = ERR28;
5779 goto FAILED;
5780 }
5781
5782 /* Read the name, but also get it as a number if it's all digits */
5783
5784 recno = 0;
5785 name = ++ptr;
5786 while (MAX_255(*ptr) && (cd->ctypes[*ptr] & ctype_word) != 0)
5787 {
5788 if (recno >= 0)
5789 recno = (IS_DIGIT(*ptr))? recno * 10 + *ptr - CHAR_0 : -1;
5790 ptr++;
5791 }
5792 namelen = (int)(ptr - name);
5793
5794 if ((terminator > 0 && *ptr++ != terminator) ||
5795 *ptr++ != CHAR_RIGHT_PARENTHESIS)
5796 {
5797 ptr--; /* Error offset */
5798 *errorcodeptr = ERR26;
5799 goto FAILED;
5800 }
5801
5802 /* Do no further checking in the pre-compile phase. */
5803
5804 if (lengthptr != NULL) break;
5805
5806 /* In the real compile we do the work of looking for the actual
5807 reference. If the string started with "+" or "-" we require the rest to
5808 be digits, in which case recno will be set. */
5809
5810 if (refsign > 0)
5811 {
5812 if (recno <= 0)
5813 {
5814 *errorcodeptr = ERR58;
5815 goto FAILED;
5816 }
5817 recno = (refsign == CHAR_MINUS)?
5818 cd->bracount - recno + 1 : recno +cd->bracount;
5819 if (recno <= 0 || recno > cd->final_bracount)
5820 {
5821 *errorcodeptr = ERR15;
5822 goto FAILED;
5823 }
5824 PUT2(code, 2+LINK_SIZE, recno);
5825 break;
5826 }
5827
5828 /* Otherwise (did not start with "+" or "-"), start by looking for the
5829 name. If we find a name, add one to the opcode to change OP_CREF or
5830 OP_RREF into OP_NCREF or OP_NRREF. These behave exactly the same,
5831 except they record that the reference was originally to a name. The
5832 information is used to check duplicate names. */
5833
5834 slot = cd->name_table;
5835 for (i = 0; i < cd->names_found; i++)
5836 {
5837 if (STRNCMP_UC_UC(name, slot+IMM2_SIZE, namelen) == 0) break;
5838 slot += cd->name_entry_size;
5839 }
5840
5841 /* Found a previous named subpattern */
5842
5843 if (i < cd->names_found)
5844 {
5845 recno = GET2(slot, 0);
5846 PUT2(code, 2+LINK_SIZE, recno);
5847 code[1+LINK_SIZE]++;
5848 }
5849
5850 /* Search the pattern for a forward reference */
5851
5852 else if ((i = find_parens(cd, name, namelen,
5853 (options & PCRE_EXTENDED) != 0, utf)) > 0)
5854 {
5855 PUT2(code, 2+LINK_SIZE, i);
5856 code[1+LINK_SIZE]++;
5857 }
5858
5859 /* If terminator == 0 it means that the name followed directly after
5860 the opening parenthesis [e.g. (?(abc)...] and in this case there are
5861 some further alternatives to try. For the cases where terminator != 0
5862 [things like (?(<name>... or (?('name')... or (?(R&name)... ] we have
5863 now checked all the possibilities, so give an error. */
5864
5865 else if (terminator != 0)
5866 {
5867 *errorcodeptr = ERR15;
5868 goto FAILED;
5869 }
5870
5871 /* Check for (?(R) for recursion. Allow digits after R to specify a
5872 specific group number. */
5873
5874 else if (*name == CHAR_R)
5875 {
5876 recno = 0;
5877 for (i = 1; i < namelen; i++)
5878 {
5879 if (!IS_DIGIT(name[i]))
5880 {
5881 *errorcodeptr = ERR15;
5882 goto FAILED;
5883 }
5884 recno = recno * 10 + name[i] - CHAR_0;
5885 }
5886 if (recno == 0) recno = RREF_ANY;
5887 code[1+LINK_SIZE] = OP_RREF; /* Change test type */
5888 PUT2(code, 2+LINK_SIZE, recno);
5889 }
5890
5891 /* Similarly, check for the (?(DEFINE) "condition", which is always
5892 false. */
5893
5894 else if (namelen == 6 && STRNCMP_UC_C8(name, STRING_DEFINE, 6) == 0)
5895 {
5896 code[1+LINK_SIZE] = OP_DEF;
5897 skipbytes = 1;
5898 }
5899
5900 /* Check for the "name" actually being a subpattern number. We are
5901 in the second pass here, so final_bracount is set. */
5902
5903 else if (recno > 0 && recno <= cd->final_bracount)
5904 {
5905 PUT2(code, 2+LINK_SIZE, recno);
5906 }
5907
5908 /* Either an unidentified subpattern, or a reference to (?(0) */
5909
5910 else
5911 {
5912 *errorcodeptr = (recno == 0)? ERR35: ERR15;
5913 goto FAILED;
5914 }
5915 break;
5916
5917
5918 /* ------------------------------------------------------------ */
5919 case CHAR_EQUALS_SIGN: /* Positive lookahead */
5920 bravalue = OP_ASSERT;
5921 cd->assert_depth += 1;
5922 ptr++;
5923 break;
5924
5925
5926 /* ------------------------------------------------------------ */
5927 case CHAR_EXCLAMATION_MARK: /* Negative lookahead */
5928 ptr++;
5929 if (*ptr == CHAR_RIGHT_PARENTHESIS) /* Optimize (?!) */
5930 {
5931 *code++ = OP_FAIL;
5932 previous = NULL;
5933 continue;
5934 }
5935 bravalue = OP_ASSERT_NOT;
5936 cd->assert_depth += 1;
5937 break;
5938
5939
5940 /* ------------------------------------------------------------ */
5941 case CHAR_LESS_THAN_SIGN: /* Lookbehind or named define */
5942 switch (ptr[1])
5943 {
5944 case CHAR_EQUALS_SIGN: /* Positive lookbehind */
5945 bravalue = OP_ASSERTBACK;
5946 cd->assert_depth += 1;
5947 ptr += 2;
5948 break;
5949
5950 case CHAR_EXCLAMATION_MARK: /* Negative lookbehind */
5951 bravalue = OP_ASSERTBACK_NOT;
5952 cd->assert_depth += 1;
5953 ptr += 2;
5954 break;
5955
5956 default: /* Could be name define, else bad */
5957 if (MAX_255(ptr[1]) && (cd->ctypes[ptr[1]] & ctype_word) != 0)
5958 goto DEFINE_NAME;
5959 ptr++; /* Correct offset for error */
5960 *errorcodeptr = ERR24;
5961 goto FAILED;
5962 }
5963 break;
5964
5965
5966 /* ------------------------------------------------------------ */
5967 case CHAR_GREATER_THAN_SIGN: /* One-time brackets */
5968 bravalue = OP_ONCE;
5969 ptr++;
5970 break;
5971
5972
5973 /* ------------------------------------------------------------ */
5974 case CHAR_C: /* Callout - may be followed by digits; */
5975 previous_callout = code; /* Save for later completion */
5976 after_manual_callout = 1; /* Skip one item before completing */
5977 *code++ = OP_CALLOUT;
5978 {
5979 int n = 0;
5980 ptr++;
5981 while(IS_DIGIT(*ptr))
5982 n = n * 10 + *ptr++ - CHAR_0;
5983 if (*ptr != CHAR_RIGHT_PARENTHESIS)
5984 {
5985 *errorcodeptr = ERR39;
5986 goto FAILED;
5987 }
5988 if (n > 255)
5989 {
5990 *errorcodeptr = ERR38;
5991 goto FAILED;
5992 }
5993 *code++ = n;
5994 PUT(code, 0, (int)(ptr - cd->start_pattern + 1)); /* Pattern offset */
5995 PUT(code, LINK_SIZE, 0); /* Default length */
5996 code += 2 * LINK_SIZE;
5997 }
5998 previous = NULL;
5999 continue;
6000
6001
6002 /* ------------------------------------------------------------ */
6003 case CHAR_P: /* Python-style named subpattern handling */
6004 if (*(++ptr) == CHAR_EQUALS_SIGN ||
6005 *ptr == CHAR_GREATER_THAN_SIGN) /* Reference or recursion */
6006 {
6007 is_recurse = *ptr == CHAR_GREATER_THAN_SIGN;
6008 terminator = CHAR_RIGHT_PARENTHESIS;
6009 goto NAMED_REF_OR_RECURSE;
6010 }
6011 else if (*ptr != CHAR_LESS_THAN_SIGN) /* Test for Python-style defn */
6012 {
6013 *errorcodeptr = ERR41;
6014 goto FAILED;
6015 }
6016 /* Fall through to handle (?P< as (?< is handled */
6017
6018
6019 /* ------------------------------------------------------------ */
6020 DEFINE_NAME: /* Come here from (?< handling */
6021 case CHAR_APOSTROPHE:
6022 {
6023 terminator = (*ptr == CHAR_LESS_THAN_SIGN)?
6024 CHAR_GREATER_THAN_SIGN : CHAR_APOSTROPHE;
6025 name = ++ptr;
6026
6027 while (MAX_255(*ptr) && (cd->ctypes[*ptr] & ctype_word) != 0) ptr++;
6028 namelen = (int)(ptr - name);
6029
6030 /* In the pre-compile phase, just do a syntax check. */
6031
6032 if (lengthptr != NULL)
6033 {
6034 if (*ptr != terminator)
6035 {
6036 *errorcodeptr = ERR42;
6037 goto FAILED;
6038 }
6039 if (cd->names_found >= MAX_NAME_COUNT)
6040 {
6041 *errorcodeptr = ERR49;
6042 goto FAILED;
6043 }
6044 if (namelen + IMM2_SIZE + 1 > cd->name_entry_size)
6045 {
6046 cd->name_entry_size = namelen + IMM2_SIZE + 1;
6047 if (namelen > MAX_NAME_SIZE)
6048 {
6049 *errorcodeptr = ERR48;
6050 goto FAILED;
6051 }
6052 }
6053 }
6054
6055 /* In the real compile, create the entry in the table, maintaining
6056 alphabetical order. Duplicate names for different numbers are
6057 permitted only if PCRE_DUPNAMES is set. Duplicate names for the same
6058 number are always OK. (An existing number can be re-used if (?|
6059 appears in the pattern.) In either event, a duplicate name results in
6060 a duplicate entry in the table, even if the number is the same. This
6061 is because the number of names, and hence the table size, is computed
6062 in the pre-compile, and it affects various numbers and pointers which
6063 would all have to be modified, and the compiled code moved down, if
6064 duplicates with the same number were omitted from the table. This
6065 doesn't seem worth the hassle. However, *different* names for the
6066 same number are not permitted. */
6067
6068 else
6069 {
6070 BOOL dupname = FALSE;
6071 slot = cd->name_table;
6072
6073 for (i = 0; i < cd->names_found; i++)
6074 {
6075 int crc = memcmp(name, slot+IMM2_SIZE, IN_UCHARS(namelen));
6076 if (crc == 0)
6077 {
6078 if (slot[IMM2_SIZE+namelen] == 0)
6079 {
6080 if (GET2(slot, 0) != cd->bracount + 1 &&
6081 (options & PCRE_DUPNAMES) == 0)
6082 {
6083 *errorcodeptr = ERR43;
6084 goto FAILED;
6085 }
6086 else dupname = TRUE;
6087 }
6088 else crc = -1; /* Current name is a substring */
6089 }
6090
6091 /* Make space in the table and break the loop for an earlier
6092 name. For a duplicate or later name, carry on. We do this for
6093 duplicates so that in the simple case (when ?(| is not used) they
6094 are in order of their numbers. */
6095
6096 if (crc < 0)
6097 {
6098 memmove(slot + cd->name_entry_size, slot,
6099 IN_UCHARS((cd->names_found - i) * cd->name_entry_size));
6100 break;
6101 }
6102
6103 /* Continue the loop for a later or duplicate name */
6104
6105 slot += cd->name_entry_size;
6106 }
6107
6108 /* For non-duplicate names, check for a duplicate number before
6109 adding the new name. */
6110
6111 if (!dupname)
6112 {
6113 pcre_uchar *cslot = cd->name_table;
6114 for (i = 0; i < cd->names_found; i++)
6115 {
6116 if (cslot != slot)
6117 {
6118 if (GET2(cslot, 0) == cd->bracount + 1)
6119 {
6120 *errorcodeptr = ERR65;
6121 goto FAILED;
6122 }
6123 }
6124 else i--;
6125 cslot += cd->name_entry_size;
6126 }
6127 }
6128
6129 PUT2(slot, 0, cd->bracount + 1);
6130 memcpy(slot + IMM2_SIZE, name, IN_UCHARS(namelen));
6131 slot[IMM2_SIZE + namelen] = 0;
6132 }
6133 }
6134
6135 /* In both pre-compile and compile, count the number of names we've
6136 encountered. */
6137
6138 cd->names_found++;
6139 ptr++; /* Move past > or ' */
6140 goto NUMBERED_GROUP;
6141
6142
6143 /* ------------------------------------------------------------ */
6144 case CHAR_AMPERSAND: /* Perl recursion/subroutine syntax */
6145 terminator = CHAR_RIGHT_PARENTHESIS;
6146 is_recurse = TRUE;
6147 /* Fall through */
6148
6149 /* We come here from the Python syntax above that handles both
6150 references (?P=name) and recursion (?P>name), as well as falling
6151 through from the Perl recursion syntax (?&name). We also come here from
6152 the Perl \k<name> or \k'name' back reference syntax and the \k{name}
6153 .NET syntax, and the Oniguruma \g<...> and \g'...' subroutine syntax. */
6154
6155 NAMED_REF_OR_RECURSE:
6156 name = ++ptr;
6157 while (MAX_255(*ptr) && (cd->ctypes[*ptr] & ctype_word) != 0) ptr++;
6158 namelen = (int)(ptr - name);
6159
6160 /* In the pre-compile phase, do a syntax check. We used to just set
6161 a dummy reference number, because it was not used in the first pass.
6162 However, with the change of recursive back references to be atomic,
6163 we have to look for the number so that this state can be identified, as
6164 otherwise the incorrect length is computed. If it's not a backwards
6165 reference, the dummy number will do. */
6166
6167 if (lengthptr != NULL)
6168 {
6169 const pcre_uchar *temp;
6170
6171 if (namelen == 0)
6172 {
6173 *errorcodeptr = ERR62;
6174 goto FAILED;
6175 }
6176 if (*ptr != terminator)
6177 {
6178 *errorcodeptr = ERR42;
6179 goto FAILED;
6180 }
6181 if (namelen > MAX_NAME_SIZE)
6182 {
6183 *errorcodeptr = ERR48;
6184 goto FAILED;
6185 }
6186
6187 /* The name table does not exist in the first pass, so we cannot
6188 do a simple search as in the code below. Instead, we have to scan the
6189 pattern to find the number. It is important that we scan it only as
6190 far as we have got because the syntax of named subpatterns has not
6191 been checked for the rest of the pattern, and find_parens() assumes
6192 correct syntax. In any case, it's a waste of resources to scan
6193 further. We stop the scan at the current point by temporarily
6194 adjusting the value of cd->endpattern. */
6195
6196 temp = cd->end_pattern;
6197 cd->end_pattern = ptr;
6198 recno = find_parens(cd, name, namelen,
6199 (options & PCRE_EXTENDED) != 0, utf);
6200 cd->end_pattern = temp;
6201 if (recno < 0) recno = 0; /* Forward ref; set dummy number */
6202 }
6203
6204 /* In the real compile, seek the name in the table. We check the name
6205 first, and then check that we have reached the end of the name in the
6206 table. That way, if the name that is longer than any in the table,
6207 the comparison will fail without reading beyond the table entry. */
6208
6209 else
6210 {
6211 slot = cd->name_table;
6212 for (i = 0; i < cd->names_found; i++)
6213 {
6214 if (STRNCMP_UC_UC(name, slot+IMM2_SIZE, namelen) == 0 &&
6215 slot[IMM2_SIZE+namelen] == 0)
6216 break;
6217 slot += cd->name_entry_size;
6218 }
6219
6220 if (i < cd->names_found) /* Back reference */
6221 {
6222 recno = GET2(slot, 0);
6223 }
6224 else if ((recno = /* Forward back reference */
6225 find_parens(cd, name, namelen,
6226 (options & PCRE_EXTENDED) != 0, utf)) <= 0)
6227 {
6228 *errorcodeptr = ERR15;
6229 goto FAILED;
6230 }
6231 }
6232
6233 /* In both phases, we can now go to the code than handles numerical
6234 recursion or backreferences. */
6235
6236 if (is_recurse) goto HANDLE_RECURSION;
6237 else goto HANDLE_REFERENCE;
6238
6239
6240 /* ------------------------------------------------------------ */
6241 case CHAR_R: /* Recursion */
6242 ptr++; /* Same as (?0) */
6243 /* Fall through */
6244
6245
6246 /* ------------------------------------------------------------ */
6247 case CHAR_MINUS: case CHAR_PLUS: /* Recursion or subroutine */
6248 case CHAR_0: case CHAR_1: case CHAR_2: case CHAR_3: case CHAR_4:
6249 case CHAR_5: case CHAR_6: case CHAR_7: case CHAR_8: case CHAR_9:
6250 {
6251 const pcre_uchar *called;
6252 terminator = CHAR_RIGHT_PARENTHESIS;
6253
6254 /* Come here from the \g<...> and \g'...' code (Oniguruma
6255 compatibility). However, the syntax has been checked to ensure that
6256 the ... are a (signed) number, so that neither ERR63 nor ERR29 will
6257 be called on this path, nor with the jump to OTHER_CHAR_AFTER_QUERY
6258 ever be taken. */
6259
6260 HANDLE_NUMERICAL_RECURSION:
6261
6262 if ((refsign = *ptr) == CHAR_PLUS)
6263 {
6264 ptr++;
6265 if (!IS_DIGIT(*ptr))
6266 {
6267 *errorcodeptr = ERR63;
6268 goto FAILED;
6269 }
6270 }
6271 else if (refsign == CHAR_MINUS)
6272 {
6273 if (!IS_DIGIT(ptr[1]))
6274 goto OTHER_CHAR_AFTER_QUERY;
6275 ptr++;
6276 }
6277
6278 recno = 0;
6279 while(IS_DIGIT(*ptr))
6280 recno = recno * 10 + *ptr++ - CHAR_0;
6281
6282 if (*ptr != terminator)
6283 {
6284 *errorcodeptr = ERR29;
6285 goto FAILED;
6286 }
6287
6288 if (refsign == CHAR_MINUS)
6289 {
6290 if (recno == 0)
6291 {
6292 *errorcodeptr = ERR58;
6293 goto FAILED;
6294 }
6295 recno = cd->bracount - recno + 1;
6296 if (recno <= 0)
6297 {
6298 *errorcodeptr = ERR15;
6299 goto FAILED;
6300 }
6301 }
6302 else if (refsign == CHAR_PLUS)
6303 {
6304 if (recno == 0)
6305 {
6306 *errorcodeptr = ERR58;
6307 goto FAILED;
6308 }
6309 recno += cd->bracount;
6310 }
6311
6312 /* Come here from code above that handles a named recursion */
6313
6314 HANDLE_RECURSION:
6315
6316 previous = code;
6317 called = cd->start_code;
6318
6319 /* When we are actually compiling, find the bracket that is being
6320 referenced. Temporarily end the regex in case it doesn't exist before
6321 this point. If we end up with a forward reference, first check that
6322 the bracket does occur later so we can give the error (and position)
6323 now. Then remember this forward reference in the workspace so it can
6324 be filled in at the end. */
6325
6326 if (lengthptr == NULL)
6327 {
6328 *code = OP_END;
6329 if (recno != 0)
6330 called = PRIV(find_bracket)(cd->start_code, utf, recno);
6331
6332 /* Forward reference */
6333
6334 if (called == NULL)
6335 {
6336 if (find_parens(cd, NULL, recno,
6337 (options & PCRE_EXTENDED) != 0, utf) < 0)
6338 {
6339 *errorcodeptr = ERR15;
6340 goto FAILED;
6341 }
6342
6343 /* Fudge the value of "called" so that when it is inserted as an
6344 offset below, what it actually inserted is the reference number
6345 of the group. Then remember the forward reference. */
6346
6347 called = cd->start_code + recno;
6348 if (cd->hwm >= cd->start_workspace + cd->workspace_size -
6349 WORK_SIZE_SAFETY_MARGIN)
6350 {
6351 *errorcodeptr = expand_workspace(cd);
6352 if (*errorcodeptr != 0) goto FAILED;
6353 }
6354 PUTINC(cd->hwm, 0, (int)(code + 1 - cd->start_code));
6355 }
6356
6357 /* If not a forward reference, and the subpattern is still open,
6358 this is a recursive call. We check to see if this is a left
6359 recursion that could loop for ever, and diagnose that case. We
6360 must not, however, do this check if we are in a conditional
6361 subpattern because the condition might be testing for recursion in
6362 a pattern such as /(?(R)a+|(?R)b)/, which is perfectly valid.
6363 Forever loops are also detected at runtime, so those that occur in
6364 conditional subpatterns will be picked up then. */
6365
6366 else if (GET(called, 1) == 0 && cond_depth <= 0 &&
6367 could_be_empty(called, code, bcptr, utf, cd))
6368 {
6369 *errorcodeptr = ERR40;
6370 goto FAILED;
6371 }
6372 }
6373
6374 /* Insert the recursion/subroutine item. It does not have a set first
6375 character (relevant if it is repeated, because it will then be
6376 wrapped with ONCE brackets). */
6377
6378 *code = OP_RECURSE;
6379 PUT(code, 1, (int)(called - cd->start_code));
6380 code += 1 + LINK_SIZE;
6381 groupsetfirstchar = FALSE;
6382 }
6383
6384 /* Can't determine a first byte now */
6385
6386 if (firstchar == REQ_UNSET) firstchar = REQ_NONE;
6387 continue;
6388
6389
6390 /* ------------------------------------------------------------ */
6391 default: /* Other characters: check option setting */
6392 OTHER_CHAR_AFTER_QUERY:
6393 set = unset = 0;
6394 optset = &set;
6395
6396 while (*ptr != CHAR_RIGHT_PARENTHESIS && *ptr != CHAR_COLON)
6397 {
6398 switch (*ptr++)
6399 {
6400 case CHAR_MINUS: optset = &unset; break;
6401
6402 case CHAR_J: /* Record that it changed in the external options */
6403 *optset |= PCRE_DUPNAMES;
6404 cd->external_flags |= PCRE_JCHANGED;
6405 break;
6406
6407 case CHAR_i: *optset |= PCRE_CASELESS; break;
6408 case CHAR_m: *optset |= PCRE_MULTILINE; break;
6409 case CHAR_s: *optset |= PCRE_DOTALL; break;
6410 case CHAR_x: *optset |= PCRE_EXTENDED; break;
6411 case CHAR_U: *optset |= PCRE_UNGREEDY; break;
6412 case CHAR_X: *optset |= PCRE_EXTRA; break;
6413
6414 default: *errorcodeptr = ERR12;
6415 ptr--; /* Correct the offset */
6416 goto FAILED;
6417 }
6418 }
6419
6420 /* Set up the changed option bits, but don't change anything yet. */
6421
6422 newoptions = (options | set) & (~unset);
6423
6424 /* If the options ended with ')' this is not the start of a nested
6425 group with option changes, so the options change at this level. If this
6426 item is right at the start of the pattern, the options can be
6427 abstracted and made external in the pre-compile phase, and ignored in
6428 the compile phase. This can be helpful when matching -- for instance in
6429 caseless checking of required bytes.
6430
6431 If the code pointer is not (cd->start_code + 1 + LINK_SIZE), we are
6432 definitely *not* at the start of the pattern because something has been
6433 compiled. In the pre-compile phase, however, the code pointer can have
6434 that value after the start, because it gets reset as code is discarded
6435 during the pre-compile. However, this can happen only at top level - if
6436 we are within parentheses, the starting BRA will still be present. At
6437 any parenthesis level, the length value can be used to test if anything
6438 has been compiled at that level. Thus, a test for both these conditions
6439 is necessary to ensure we correctly detect the start of the pattern in
6440 both phases.
6441
6442 If we are not at the pattern start, reset the greedy defaults and the
6443 case value for firstchar and reqchar. */
6444
6445 if (*ptr == CHAR_RIGHT_PARENTHESIS)
6446 {
6447 if (code == cd->start_code + 1 + LINK_SIZE &&
6448 (lengthptr == NULL || *lengthptr == 2 + 2*LINK_SIZE))
6449 {
6450 cd->external_options = newoptions;
6451 }
6452 else
6453 {
6454 greedy_default = ((newoptions & PCRE_UNGREEDY) != 0);
6455 greedy_non_default = greedy_default ^ 1;
6456 req_caseopt = ((newoptions & PCRE_CASELESS) != 0)? REQ_CASELESS:0;
6457 }
6458
6459 /* Change options at this level, and pass them back for use
6460 in subsequent branches. */
6461
6462 *optionsptr = options = newoptions;
6463 previous = NULL; /* This item can't be repeated */
6464 continue; /* It is complete */
6465 }
6466
6467 /* If the options ended with ':' we are heading into a nested group
6468 with possible change of options. Such groups are non-capturing and are
6469 not assertions of any kind. All we need to do is skip over the ':';
6470 the newoptions value is handled below. */
6471
6472 bravalue = OP_BRA;
6473 ptr++;
6474 } /* End of switch for character following (? */
6475 } /* End of (? handling */
6476
6477 /* Opening parenthesis not followed by '*' or '?'. If PCRE_NO_AUTO_CAPTURE
6478 is set, all unadorned brackets become non-capturing and behave like (?:...)
6479 brackets. */
6480
6481 else if ((options & PCRE_NO_AUTO_CAPTURE) != 0)
6482 {
6483 bravalue = OP_BRA;
6484 }
6485
6486 /* Else we have a capturing group. */
6487
6488 else
6489 {
6490 NUMBERED_GROUP:
6491 cd->bracount += 1;
6492 PUT2(code, 1+LINK_SIZE, cd->bracount);
6493 skipbytes = IMM2_SIZE;
6494 }
6495
6496 /* Process nested bracketed regex. Assertions used not to be repeatable,
6497 but this was changed for Perl compatibility, so all kinds can now be
6498 repeated. We copy code into a non-register variable (tempcode) in order to
6499 be able to pass its address because some compilers complain otherwise. */
6500
6501 previous = code; /* For handling repetition */
6502 *code = bravalue;
6503 tempcode = code;
6504 tempreqvary = cd->req_varyopt; /* Save value before bracket */
6505 tempbracount = cd->bracount; /* Save value before bracket */
6506 length_prevgroup = 0; /* Initialize for pre-compile phase */
6507
6508 if (!compile_regex(
6509 newoptions, /* The complete new option state */
6510 &tempcode, /* Where to put code (updated) */
6511 &ptr, /* Input pointer (updated) */
6512 errorcodeptr, /* Where to put an error message */
6513 (bravalue == OP_ASSERTBACK ||
6514 bravalue == OP_ASSERTBACK_NOT), /* TRUE if back assert */
6515 reset_bracount, /* True if (?| group */
6516 skipbytes, /* Skip over bracket number */
6517 cond_depth +
6518 ((bravalue == OP_COND)?1:0), /* Depth of condition subpatterns */
6519 &subfirstchar, /* For possible first char */
6520 &subreqchar, /* For possible last char */
6521 bcptr, /* Current branch chain */
6522 cd, /* Tables block */
6523 (lengthptr == NULL)? NULL : /* Actual compile phase */
6524 &length_prevgroup /* Pre-compile phase */
6525 ))
6526 goto FAILED;
6527
6528 /* If this was an atomic group and there are no capturing groups within it,
6529 generate OP_ONCE_NC instead of OP_ONCE. */
6530
6531 if (bravalue == OP_ONCE && cd->bracount <= tempbracount)
6532 *code = OP_ONCE_NC;
6533
6534 if (bravalue >= OP_ASSERT && bravalue <= OP_ASSERTBACK_NOT)
6535 cd->assert_depth -= 1;
6536
6537 /* At the end of compiling, code is still pointing to the start of the
6538 group, while tempcode has been updated to point past the end of the group.
6539 The pattern pointer (ptr) is on the bracket.
6540
6541 If this is a conditional bracket, check that there are no more than
6542 two branches in the group, or just one if it's a DEFINE group. We do this
6543 in the real compile phase, not in the pre-pass, where the whole group may
6544 not be available. */
6545
6546 if (bravalue == OP_COND && lengthptr == NULL)
6547 {
6548 pcre_uchar *tc = code;
6549 int condcount = 0;
6550
6551 do {
6552 condcount++;
6553 tc += GET(tc,1);
6554 }
6555 while (*tc != OP_KET);
6556
6557 /* A DEFINE group is never obeyed inline (the "condition" is always
6558 false). It must have only one branch. */
6559
6560 if (code[LINK_SIZE+1] == OP_DEF)
6561 {
6562 if (condcount > 1)
6563 {
6564 *errorcodeptr = ERR54;
6565 goto FAILED;
6566 }
6567 bravalue = OP_DEF; /* Just a flag to suppress char handling below */
6568 }
6569
6570 /* A "normal" conditional group. If there is just one branch, we must not
6571 make use of its firstchar or reqchar, because this is equivalent to an
6572 empty second branch. */
6573
6574 else
6575 {
6576 if (condcount > 2)
6577 {
6578 *errorcodeptr = ERR27;
6579 goto FAILED;
6580 }
6581 if (condcount == 1) subfirstchar = subreqchar = REQ_NONE;
6582 }
6583 }
6584
6585 /* Error if hit end of pattern */
6586
6587 if (*ptr != CHAR_RIGHT_PARENTHESIS)
6588 {
6589 *errorcodeptr = ERR14;
6590 goto FAILED;
6591 }
6592
6593 /* In the pre-compile phase, update the length by the length of the group,
6594 less the brackets at either end. Then reduce the compiled code to just a
6595 set of non-capturing brackets so that it doesn't use much memory if it is
6596 duplicated by a quantifier.*/
6597
6598 if (lengthptr != NULL)
6599 {
6600 if (OFLOW_MAX - *lengthptr < length_prevgroup - 2 - 2*LINK_SIZE)
6601 {
6602 *errorcodeptr = ERR20;
6603 goto FAILED;
6604 }
6605 *lengthptr += length_prevgroup - 2 - 2*LINK_SIZE;
6606 code++; /* This already contains bravalue */
6607 PUTINC(code, 0, 1 + LINK_SIZE);
6608 *code++ = OP_KET;
6609 PUTINC(code, 0, 1 + LINK_SIZE);
6610 break; /* No need to waste time with special character handling */
6611 }
6612
6613 /* Otherwise update the main code pointer to the end of the group. */
6614
6615 code = tempcode;
6616
6617 /* For a DEFINE group, required and first character settings are not
6618 relevant. */
6619
6620 if (bravalue == OP_DEF) break;
6621
6622 /* Handle updating of the required and first characters for other types of
6623 group. Update for normal brackets of all kinds, and conditions with two
6624 branches (see code above). If the bracket is followed by a quantifier with
6625 zero repeat, we have to back off. Hence the definition of zeroreqchar and
6626 zerofirstchar outside the main loop so that they can be accessed for the
6627 back off. */
6628
6629 zeroreqchar = reqchar;
6630 zerofirstchar = firstchar;
6631 groupsetfirstchar = FALSE;
6632
6633 if (bravalue >= OP_ONCE)
6634 {
6635 /* If we have not yet set a firstchar in this branch, take it from the
6636 subpattern, remembering that it was set here so that a repeat of more
6637 than one can replicate it as reqchar if necessary. If the subpattern has
6638 no firstchar, set "none" for the whole branch. In both cases, a zero
6639 repeat forces firstchar to "none". */
6640
6641 if (firstchar == REQ_UNSET)
6642 {