Annotation of embedaddon/pcre/pcre_compile.c, revision 1.1.1.1
1.1 misho 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) 1997-2011 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_printint() function, which is
57: also used by pcretest. PCRE_DEBUG is not defined when building a production
58: library. */
59:
60: #ifdef PCRE_DEBUG
61: #include "pcre_printint.src"
62: #endif
63:
64:
65: /* Macro for setting individual bits in class bitmaps. */
66:
67: #define SETBIT(a,b) a[b/8] |= (1 << (b%8))
68:
69: /* Maximum length value to check against when making sure that the integer that
70: holds the compiled pattern length does not overflow. We make it a bit less than
71: INT_MAX to allow for adding in group terminating bytes, so that we don't have
72: to check them every time. */
73:
74: #define OFLOW_MAX (INT_MAX - 20)
75:
76:
77: /*************************************************
78: * Code parameters and static tables *
79: *************************************************/
80:
81: /* This value specifies the size of stack workspace that is used during the
82: first pre-compile phase that determines how much memory is required. The regex
83: is partly compiled into this space, but the compiled parts are discarded as
84: soon as they can be, so that hopefully there will never be an overrun. The code
85: does, however, check for an overrun. The largest amount I've seen used is 218,
86: so this number is very generous.
87:
88: The same workspace is used during the second, actual compile phase for
89: remembering forward references to groups so that they can be filled in at the
90: end. Each entry in this list occupies LINK_SIZE bytes, so even when LINK_SIZE
91: is 4 there is plenty of room for most patterns. However, the memory can get
92: filled up by repetitions of forward references, for example patterns like
93: /(?1){0,1999}(b)/, and one user did hit the limit. The code has been changed so
94: that the workspace is expanded using malloc() in this situation. The value
95: below is therefore a minimum, and we put a maximum on it for safety. The
96: minimum is now also defined in terms of LINK_SIZE so that the use of malloc()
97: kicks in at the same number of forward references in all cases. */
98:
99: #define COMPILE_WORK_SIZE (2048*LINK_SIZE)
100: #define COMPILE_WORK_SIZE_MAX (100*COMPILE_WORK_SIZE)
101:
102: /* The overrun tests check for a slightly smaller size so that they detect the
103: overrun before it actually does run off the end of the data block. */
104:
105: #define WORK_SIZE_SAFETY_MARGIN (100)
106:
107:
108: /* Table for handling escaped characters in the range '0'-'z'. Positive returns
109: are simple data values; negative values are for special things like \d and so
110: on. Zero means further processing is needed (for things like \x), or the escape
111: is invalid. */
112:
113: #ifndef EBCDIC
114:
115: /* This is the "normal" table for ASCII systems or for EBCDIC systems running
116: in UTF-8 mode. */
117:
118: static const short int escapes[] = {
119: 0, 0,
120: 0, 0,
121: 0, 0,
122: 0, 0,
123: 0, 0,
124: CHAR_COLON, CHAR_SEMICOLON,
125: CHAR_LESS_THAN_SIGN, CHAR_EQUALS_SIGN,
126: CHAR_GREATER_THAN_SIGN, CHAR_QUESTION_MARK,
127: CHAR_COMMERCIAL_AT, -ESC_A,
128: -ESC_B, -ESC_C,
129: -ESC_D, -ESC_E,
130: 0, -ESC_G,
131: -ESC_H, 0,
132: 0, -ESC_K,
133: 0, 0,
134: -ESC_N, 0,
135: -ESC_P, -ESC_Q,
136: -ESC_R, -ESC_S,
137: 0, 0,
138: -ESC_V, -ESC_W,
139: -ESC_X, 0,
140: -ESC_Z, CHAR_LEFT_SQUARE_BRACKET,
141: CHAR_BACKSLASH, CHAR_RIGHT_SQUARE_BRACKET,
142: CHAR_CIRCUMFLEX_ACCENT, CHAR_UNDERSCORE,
143: CHAR_GRAVE_ACCENT, 7,
144: -ESC_b, 0,
145: -ESC_d, ESC_e,
146: ESC_f, 0,
147: -ESC_h, 0,
148: 0, -ESC_k,
149: 0, 0,
150: ESC_n, 0,
151: -ESC_p, 0,
152: ESC_r, -ESC_s,
153: ESC_tee, 0,
154: -ESC_v, -ESC_w,
155: 0, 0,
156: -ESC_z
157: };
158:
159: #else
160:
161: /* This is the "abnormal" table for EBCDIC systems without UTF-8 support. */
162:
163: static const short int escapes[] = {
164: /* 48 */ 0, 0, 0, '.', '<', '(', '+', '|',
165: /* 50 */ '&', 0, 0, 0, 0, 0, 0, 0,
166: /* 58 */ 0, 0, '!', '$', '*', ')', ';', '~',
167: /* 60 */ '-', '/', 0, 0, 0, 0, 0, 0,
168: /* 68 */ 0, 0, '|', ',', '%', '_', '>', '?',
169: /* 70 */ 0, 0, 0, 0, 0, 0, 0, 0,
170: /* 78 */ 0, '`', ':', '#', '@', '\'', '=', '"',
171: /* 80 */ 0, 7, -ESC_b, 0, -ESC_d, ESC_e, ESC_f, 0,
172: /* 88 */-ESC_h, 0, 0, '{', 0, 0, 0, 0,
173: /* 90 */ 0, 0, -ESC_k, 'l', 0, ESC_n, 0, -ESC_p,
174: /* 98 */ 0, ESC_r, 0, '}', 0, 0, 0, 0,
175: /* A0 */ 0, '~', -ESC_s, ESC_tee, 0,-ESC_v, -ESC_w, 0,
176: /* A8 */ 0,-ESC_z, 0, 0, 0, '[', 0, 0,
177: /* B0 */ 0, 0, 0, 0, 0, 0, 0, 0,
178: /* B8 */ 0, 0, 0, 0, 0, ']', '=', '-',
179: /* C0 */ '{',-ESC_A, -ESC_B, -ESC_C, -ESC_D,-ESC_E, 0, -ESC_G,
180: /* C8 */-ESC_H, 0, 0, 0, 0, 0, 0, 0,
181: /* D0 */ '}', 0, -ESC_K, 0, 0,-ESC_N, 0, -ESC_P,
182: /* D8 */-ESC_Q,-ESC_R, 0, 0, 0, 0, 0, 0,
183: /* E0 */ '\\', 0, -ESC_S, 0, 0,-ESC_V, -ESC_W, -ESC_X,
184: /* E8 */ 0,-ESC_Z, 0, 0, 0, 0, 0, 0,
185: /* F0 */ 0, 0, 0, 0, 0, 0, 0, 0,
186: /* F8 */ 0, 0, 0, 0, 0, 0, 0, 0
187: };
188: #endif
189:
190:
191: /* Table of special "verbs" like (*PRUNE). This is a short table, so it is
192: searched linearly. Put all the names into a single string, in order to reduce
193: the number of relocations when a shared library is dynamically linked. The
194: string is built from string macros so that it works in UTF-8 mode on EBCDIC
195: platforms. */
196:
197: typedef struct verbitem {
198: int len; /* Length of verb name */
199: int op; /* Op when no arg, or -1 if arg mandatory */
200: int op_arg; /* Op when arg present, or -1 if not allowed */
201: } verbitem;
202:
203: static const char verbnames[] =
204: "\0" /* Empty name is a shorthand for MARK */
205: STRING_MARK0
206: STRING_ACCEPT0
207: STRING_COMMIT0
208: STRING_F0
209: STRING_FAIL0
210: STRING_PRUNE0
211: STRING_SKIP0
212: STRING_THEN;
213:
214: static const verbitem verbs[] = {
215: { 0, -1, OP_MARK },
216: { 4, -1, OP_MARK },
217: { 6, OP_ACCEPT, -1 },
218: { 6, OP_COMMIT, -1 },
219: { 1, OP_FAIL, -1 },
220: { 4, OP_FAIL, -1 },
221: { 5, OP_PRUNE, OP_PRUNE_ARG },
222: { 4, OP_SKIP, OP_SKIP_ARG },
223: { 4, OP_THEN, OP_THEN_ARG }
224: };
225:
226: static const int verbcount = sizeof(verbs)/sizeof(verbitem);
227:
228:
229: /* Tables of names of POSIX character classes and their lengths. The names are
230: now all in a single string, to reduce the number of relocations when a shared
231: library is dynamically loaded. The list of lengths is terminated by a zero
232: length entry. The first three must be alpha, lower, upper, as this is assumed
233: for handling case independence. */
234:
235: static const char posix_names[] =
236: STRING_alpha0 STRING_lower0 STRING_upper0 STRING_alnum0
237: STRING_ascii0 STRING_blank0 STRING_cntrl0 STRING_digit0
238: STRING_graph0 STRING_print0 STRING_punct0 STRING_space0
239: STRING_word0 STRING_xdigit;
240:
241: static const uschar posix_name_lengths[] = {
242: 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 4, 6, 0 };
243:
244: /* Table of class bit maps for each POSIX class. Each class is formed from a
245: base map, with an optional addition or removal of another map. Then, for some
246: classes, there is some additional tweaking: for [:blank:] the vertical space
247: characters are removed, and for [:alpha:] and [:alnum:] the underscore
248: character is removed. The triples in the table consist of the base map offset,
249: second map offset or -1 if no second map, and a non-negative value for map
250: addition or a negative value for map subtraction (if there are two maps). The
251: absolute value of the third field has these meanings: 0 => no tweaking, 1 =>
252: remove vertical space characters, 2 => remove underscore. */
253:
254: static const int posix_class_maps[] = {
255: cbit_word, cbit_digit, -2, /* alpha */
256: cbit_lower, -1, 0, /* lower */
257: cbit_upper, -1, 0, /* upper */
258: cbit_word, -1, 2, /* alnum - word without underscore */
259: cbit_print, cbit_cntrl, 0, /* ascii */
260: cbit_space, -1, 1, /* blank - a GNU extension */
261: cbit_cntrl, -1, 0, /* cntrl */
262: cbit_digit, -1, 0, /* digit */
263: cbit_graph, -1, 0, /* graph */
264: cbit_print, -1, 0, /* print */
265: cbit_punct, -1, 0, /* punct */
266: cbit_space, -1, 0, /* space */
267: cbit_word, -1, 0, /* word - a Perl extension */
268: cbit_xdigit,-1, 0 /* xdigit */
269: };
270:
271: /* Table of substitutes for \d etc when PCRE_UCP is set. The POSIX class
272: substitutes must be in the order of the names, defined above, and there are
273: both positive and negative cases. NULL means no substitute. */
274:
275: #ifdef SUPPORT_UCP
276: static const uschar *substitutes[] = {
277: (uschar *)"\\P{Nd}", /* \D */
278: (uschar *)"\\p{Nd}", /* \d */
279: (uschar *)"\\P{Xsp}", /* \S */ /* NOTE: Xsp is Perl space */
280: (uschar *)"\\p{Xsp}", /* \s */
281: (uschar *)"\\P{Xwd}", /* \W */
282: (uschar *)"\\p{Xwd}" /* \w */
283: };
284:
285: static const uschar *posix_substitutes[] = {
286: (uschar *)"\\p{L}", /* alpha */
287: (uschar *)"\\p{Ll}", /* lower */
288: (uschar *)"\\p{Lu}", /* upper */
289: (uschar *)"\\p{Xan}", /* alnum */
290: NULL, /* ascii */
291: (uschar *)"\\h", /* blank */
292: NULL, /* cntrl */
293: (uschar *)"\\p{Nd}", /* digit */
294: NULL, /* graph */
295: NULL, /* print */
296: NULL, /* punct */
297: (uschar *)"\\p{Xps}", /* space */ /* NOTE: Xps is POSIX space */
298: (uschar *)"\\p{Xwd}", /* word */
299: NULL, /* xdigit */
300: /* Negated cases */
301: (uschar *)"\\P{L}", /* ^alpha */
302: (uschar *)"\\P{Ll}", /* ^lower */
303: (uschar *)"\\P{Lu}", /* ^upper */
304: (uschar *)"\\P{Xan}", /* ^alnum */
305: NULL, /* ^ascii */
306: (uschar *)"\\H", /* ^blank */
307: NULL, /* ^cntrl */
308: (uschar *)"\\P{Nd}", /* ^digit */
309: NULL, /* ^graph */
310: NULL, /* ^print */
311: NULL, /* ^punct */
312: (uschar *)"\\P{Xps}", /* ^space */ /* NOTE: Xps is POSIX space */
313: (uschar *)"\\P{Xwd}", /* ^word */
314: NULL /* ^xdigit */
315: };
316: #define POSIX_SUBSIZE (sizeof(posix_substitutes)/sizeof(uschar *))
317: #endif
318:
319: #define STRING(a) # a
320: #define XSTRING(s) STRING(s)
321:
322: /* The texts of compile-time error messages. These are "char *" because they
323: are passed to the outside world. Do not ever re-use any error number, because
324: they are documented. Always add a new error instead. Messages marked DEAD below
325: are no longer used. This used to be a table of strings, but in order to reduce
326: the number of relocations needed when a shared library is loaded dynamically,
327: it is now one long string. We cannot use a table of offsets, because the
328: lengths of inserts such as XSTRING(MAX_NAME_SIZE) are not known. Instead, we
329: simply count through to the one we want - this isn't a performance issue
330: because these strings are used only when there is a compilation error.
331:
332: Each substring ends with \0 to insert a null character. This includes the final
333: substring, so that the whole string ends with \0\0, which can be detected when
334: counting through. */
335:
336: static const char error_texts[] =
337: "no error\0"
338: "\\ at end of pattern\0"
339: "\\c at end of pattern\0"
340: "unrecognized character follows \\\0"
341: "numbers out of order in {} quantifier\0"
342: /* 5 */
343: "number too big in {} quantifier\0"
344: "missing terminating ] for character class\0"
345: "invalid escape sequence in character class\0"
346: "range out of order in character class\0"
347: "nothing to repeat\0"
348: /* 10 */
349: "operand of unlimited repeat could match the empty string\0" /** DEAD **/
350: "internal error: unexpected repeat\0"
351: "unrecognized character after (? or (?-\0"
352: "POSIX named classes are supported only within a class\0"
353: "missing )\0"
354: /* 15 */
355: "reference to non-existent subpattern\0"
356: "erroffset passed as NULL\0"
357: "unknown option bit(s) set\0"
358: "missing ) after comment\0"
359: "parentheses nested too deeply\0" /** DEAD **/
360: /* 20 */
361: "regular expression is too large\0"
362: "failed to get memory\0"
363: "unmatched parentheses\0"
364: "internal error: code overflow\0"
365: "unrecognized character after (?<\0"
366: /* 25 */
367: "lookbehind assertion is not fixed length\0"
368: "malformed number or name after (?(\0"
369: "conditional group contains more than two branches\0"
370: "assertion expected after (?(\0"
371: "(?R or (?[+-]digits must be followed by )\0"
372: /* 30 */
373: "unknown POSIX class name\0"
374: "POSIX collating elements are not supported\0"
375: "this version of PCRE is not compiled with PCRE_UTF8 support\0"
376: "spare error\0" /** DEAD **/
377: "character value in \\x{...} sequence is too large\0"
378: /* 35 */
379: "invalid condition (?(0)\0"
380: "\\C not allowed in lookbehind assertion\0"
381: "PCRE does not support \\L, \\l, \\N{name}, \\U, or \\u\0"
382: "number after (?C is > 255\0"
383: "closing ) for (?C expected\0"
384: /* 40 */
385: "recursive call could loop indefinitely\0"
386: "unrecognized character after (?P\0"
387: "syntax error in subpattern name (missing terminator)\0"
388: "two named subpatterns have the same name\0"
389: "invalid UTF-8 string\0"
390: /* 45 */
391: "support for \\P, \\p, and \\X has not been compiled\0"
392: "malformed \\P or \\p sequence\0"
393: "unknown property name after \\P or \\p\0"
394: "subpattern name is too long (maximum " XSTRING(MAX_NAME_SIZE) " characters)\0"
395: "too many named subpatterns (maximum " XSTRING(MAX_NAME_COUNT) ")\0"
396: /* 50 */
397: "repeated subpattern is too long\0" /** DEAD **/
398: "octal value is greater than \\377 (not in UTF-8 mode)\0"
399: "internal error: overran compiling workspace\0"
400: "internal error: previously-checked referenced subpattern not found\0"
401: "DEFINE group contains more than one branch\0"
402: /* 55 */
403: "repeating a DEFINE group is not allowed\0" /** DEAD **/
404: "inconsistent NEWLINE options\0"
405: "\\g is not followed by a braced, angle-bracketed, or quoted name/number or by a plain number\0"
406: "a numbered reference must not be zero\0"
407: "an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)\0"
408: /* 60 */
409: "(*VERB) not recognized\0"
410: "number is too big\0"
411: "subpattern name expected\0"
412: "digit expected after (?+\0"
413: "] is an invalid data character in JavaScript compatibility mode\0"
414: /* 65 */
415: "different names for subpatterns of the same number are not allowed\0"
416: "(*MARK) must have an argument\0"
417: "this version of PCRE is not compiled with PCRE_UCP support\0"
418: "\\c must be followed by an ASCII character\0"
419: "\\k is not followed by a braced, angle-bracketed, or quoted name\0"
420: /* 70 */
421: "internal error: unknown opcode in find_fixedlength()\0"
422: "\\N is not supported in a class\0"
423: "too many forward references\0"
424: ;
425:
426: /* Table to identify digits and hex digits. This is used when compiling
427: patterns. Note that the tables in chartables are dependent on the locale, and
428: may mark arbitrary characters as digits - but the PCRE compiling code expects
429: to handle only 0-9, a-z, and A-Z as digits when compiling. That is why we have
430: a private table here. It costs 256 bytes, but it is a lot faster than doing
431: character value tests (at least in some simple cases I timed), and in some
432: applications one wants PCRE to compile efficiently as well as match
433: efficiently.
434:
435: For convenience, we use the same bit definitions as in chartables:
436:
437: 0x04 decimal digit
438: 0x08 hexadecimal digit
439:
440: Then we can use ctype_digit and ctype_xdigit in the code. */
441:
442: #ifndef EBCDIC
443:
444: /* This is the "normal" case, for ASCII systems, and EBCDIC systems running in
445: UTF-8 mode. */
446:
447: static const unsigned char digitab[] =
448: {
449: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 0- 7 */
450: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 8- 15 */
451: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 16- 23 */
452: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 24- 31 */
453: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - ' */
454: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* ( - / */
455: 0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c, /* 0 - 7 */
456: 0x0c,0x0c,0x00,0x00,0x00,0x00,0x00,0x00, /* 8 - ? */
457: 0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /* @ - G */
458: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* H - O */
459: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* P - W */
460: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* X - _ */
461: 0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /* ` - g */
462: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* h - o */
463: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* p - w */
464: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* x -127 */
465: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 128-135 */
466: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 136-143 */
467: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 144-151 */
468: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 152-159 */
469: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 160-167 */
470: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 168-175 */
471: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 176-183 */
472: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 184-191 */
473: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 192-199 */
474: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 200-207 */
475: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 208-215 */
476: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 216-223 */
477: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 224-231 */
478: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 232-239 */
479: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 240-247 */
480: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00};/* 248-255 */
481:
482: #else
483:
484: /* This is the "abnormal" case, for EBCDIC systems not running in UTF-8 mode. */
485:
486: static const unsigned char digitab[] =
487: {
488: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 0- 7 0 */
489: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 8- 15 */
490: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 16- 23 10 */
491: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 24- 31 */
492: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 32- 39 20 */
493: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 40- 47 */
494: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 48- 55 30 */
495: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 56- 63 */
496: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - 71 40 */
497: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 72- | */
498: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* & - 87 50 */
499: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 88- 95 */
500: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - -103 60 */
501: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 104- ? */
502: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 112-119 70 */
503: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 120- " */
504: 0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /* 128- g 80 */
505: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* h -143 */
506: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 144- p 90 */
507: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* q -159 */
508: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 160- x A0 */
509: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* y -175 */
510: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* ^ -183 B0 */
511: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 184-191 */
512: 0x00,0x08,0x08,0x08,0x08,0x08,0x08,0x00, /* { - G C0 */
513: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* H -207 */
514: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* } - P D0 */
515: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* Q -223 */
516: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* \ - X E0 */
517: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* Y -239 */
518: 0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c,0x0c, /* 0 - 7 F0 */
519: 0x0c,0x0c,0x00,0x00,0x00,0x00,0x00,0x00};/* 8 -255 */
520:
521: static const unsigned char ebcdic_chartab[] = { /* chartable partial dup */
522: 0x80,0x00,0x00,0x00,0x00,0x01,0x00,0x00, /* 0- 7 */
523: 0x00,0x00,0x00,0x00,0x01,0x01,0x00,0x00, /* 8- 15 */
524: 0x00,0x00,0x00,0x00,0x00,0x01,0x00,0x00, /* 16- 23 */
525: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 24- 31 */
526: 0x00,0x00,0x00,0x00,0x00,0x01,0x00,0x00, /* 32- 39 */
527: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 40- 47 */
528: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 48- 55 */
529: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 56- 63 */
530: 0x01,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - 71 */
531: 0x00,0x00,0x00,0x80,0x00,0x80,0x80,0x80, /* 72- | */
532: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* & - 87 */
533: 0x00,0x00,0x00,0x80,0x80,0x80,0x00,0x00, /* 88- 95 */
534: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* - -103 */
535: 0x00,0x00,0x00,0x00,0x00,0x10,0x00,0x80, /* 104- ? */
536: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 112-119 */
537: 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* 120- " */
538: 0x00,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /* 128- g */
539: 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* h -143 */
540: 0x00,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* 144- p */
541: 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* q -159 */
542: 0x00,0x00,0x12,0x12,0x12,0x12,0x12,0x12, /* 160- x */
543: 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* y -175 */
544: 0x80,0x00,0x00,0x00,0x00,0x00,0x00,0x00, /* ^ -183 */
545: 0x00,0x00,0x80,0x00,0x00,0x00,0x00,0x00, /* 184-191 */
546: 0x80,0x1a,0x1a,0x1a,0x1a,0x1a,0x1a,0x12, /* { - G */
547: 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* H -207 */
548: 0x00,0x12,0x12,0x12,0x12,0x12,0x12,0x12, /* } - P */
549: 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* Q -223 */
550: 0x00,0x00,0x12,0x12,0x12,0x12,0x12,0x12, /* \ - X */
551: 0x12,0x12,0x00,0x00,0x00,0x00,0x00,0x00, /* Y -239 */
552: 0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c,0x1c, /* 0 - 7 */
553: 0x1c,0x1c,0x00,0x00,0x00,0x00,0x00,0x00};/* 8 -255 */
554: #endif
555:
556:
557: /* Definition to allow mutual recursion */
558:
559: static BOOL
560: compile_regex(int, uschar **, const uschar **, int *, BOOL, BOOL, int, int,
561: int *, int *, branch_chain *, compile_data *, int *);
562:
563:
564:
565: /*************************************************
566: * Find an error text *
567: *************************************************/
568:
569: /* The error texts are now all in one long string, to save on relocations. As
570: some of the text is of unknown length, we can't use a table of offsets.
571: Instead, just count through the strings. This is not a performance issue
572: because it happens only when there has been a compilation error.
573:
574: Argument: the error number
575: Returns: pointer to the error string
576: */
577:
578: static const char *
579: find_error_text(int n)
580: {
581: const char *s = error_texts;
582: for (; n > 0; n--)
583: {
584: while (*s++ != 0) {};
585: if (*s == 0) return "Error text not found (please report)";
586: }
587: return s;
588: }
589:
590:
591: /*************************************************
592: * Expand the workspace *
593: *************************************************/
594:
595: /* This function is called during the second compiling phase, if the number of
596: forward references fills the existing workspace, which is originally a block on
597: the stack. A larger block is obtained from malloc() unless the ultimate limit
598: has been reached or the increase will be rather small.
599:
600: Argument: pointer to the compile data block
601: Returns: 0 if all went well, else an error number
602: */
603:
604: static int
605: expand_workspace(compile_data *cd)
606: {
607: uschar *newspace;
608: int newsize = cd->workspace_size * 2;
609:
610: if (newsize > COMPILE_WORK_SIZE_MAX) newsize = COMPILE_WORK_SIZE_MAX;
611: if (cd->workspace_size >= COMPILE_WORK_SIZE_MAX ||
612: newsize - cd->workspace_size < WORK_SIZE_SAFETY_MARGIN)
613: return ERR72;
614:
615: newspace = (pcre_malloc)(newsize);
616: if (newspace == NULL) return ERR21;
617:
618: memcpy(newspace, cd->start_workspace, cd->workspace_size);
619: cd->hwm = (uschar *)newspace + (cd->hwm - cd->start_workspace);
620: if (cd->workspace_size > COMPILE_WORK_SIZE)
621: (pcre_free)((void *)cd->start_workspace);
622: cd->start_workspace = newspace;
623: cd->workspace_size = newsize;
624: return 0;
625: }
626:
627:
628:
629: /*************************************************
630: * Check for counted repeat *
631: *************************************************/
632:
633: /* This function is called when a '{' is encountered in a place where it might
634: start a quantifier. It looks ahead to see if it really is a quantifier or not.
635: It is only a quantifier if it is one of the forms {ddd} {ddd,} or {ddd,ddd}
636: where the ddds are digits.
637:
638: Arguments:
639: p pointer to the first char after '{'
640:
641: Returns: TRUE or FALSE
642: */
643:
644: static BOOL
645: is_counted_repeat(const uschar *p)
646: {
647: if ((digitab[*p++] & ctype_digit) == 0) return FALSE;
648: while ((digitab[*p] & ctype_digit) != 0) p++;
649: if (*p == CHAR_RIGHT_CURLY_BRACKET) return TRUE;
650:
651: if (*p++ != CHAR_COMMA) return FALSE;
652: if (*p == CHAR_RIGHT_CURLY_BRACKET) return TRUE;
653:
654: if ((digitab[*p++] & ctype_digit) == 0) return FALSE;
655: while ((digitab[*p] & ctype_digit) != 0) p++;
656:
657: return (*p == CHAR_RIGHT_CURLY_BRACKET);
658: }
659:
660:
661:
662: /*************************************************
663: * Handle escapes *
664: *************************************************/
665:
666: /* This function is called when a \ has been encountered. It either returns a
667: positive value for a simple escape such as \n, or a negative value which
668: encodes one of the more complicated things such as \d. A backreference to group
669: n is returned as -(ESC_REF + n); ESC_REF is the highest ESC_xxx macro. When
670: UTF-8 is enabled, a positive value greater than 255 may be returned. On entry,
671: ptr is pointing at the \. On exit, it is on the final character of the escape
672: sequence.
673:
674: Arguments:
675: ptrptr points to the pattern position pointer
676: errorcodeptr points to the errorcode variable
677: bracount number of previous extracting brackets
678: options the options bits
679: isclass TRUE if inside a character class
680:
681: Returns: zero or positive => a data character
682: negative => a special escape sequence
683: on error, errorcodeptr is set
684: */
685:
686: static int
687: check_escape(const uschar **ptrptr, int *errorcodeptr, int bracount,
688: int options, BOOL isclass)
689: {
690: BOOL utf8 = (options & PCRE_UTF8) != 0;
691: const uschar *ptr = *ptrptr + 1;
692: int c, i;
693:
694: GETCHARINCTEST(c, ptr); /* Get character value, increment pointer */
695: ptr--; /* Set pointer back to the last byte */
696:
697: /* If backslash is at the end of the pattern, it's an error. */
698:
699: if (c == 0) *errorcodeptr = ERR1;
700:
701: /* Non-alphanumerics are literals. For digits or letters, do an initial lookup
702: in a table. A non-zero result is something that can be returned immediately.
703: Otherwise further processing may be required. */
704:
705: #ifndef EBCDIC /* ASCII/UTF-8 coding */
706: else if (c < CHAR_0 || c > CHAR_z) {} /* Not alphanumeric */
707: else if ((i = escapes[c - CHAR_0]) != 0) c = i;
708:
709: #else /* EBCDIC coding */
710: else if (c < 'a' || (ebcdic_chartab[c] & 0x0E) == 0) {} /* Not alphanumeric */
711: else if ((i = escapes[c - 0x48]) != 0) c = i;
712: #endif
713:
714: /* Escapes that need further processing, or are illegal. */
715:
716: else
717: {
718: const uschar *oldptr;
719: BOOL braced, negated;
720:
721: switch (c)
722: {
723: /* A number of Perl escapes are not handled by PCRE. We give an explicit
724: error. */
725:
726: case CHAR_l:
727: case CHAR_L:
728: *errorcodeptr = ERR37;
729: break;
730:
731: case CHAR_u:
732: if ((options & PCRE_JAVASCRIPT_COMPAT) != 0)
733: {
734: /* In JavaScript, \u must be followed by four hexadecimal numbers.
735: Otherwise it is a lowercase u letter. */
736: if ((digitab[ptr[1]] & ctype_xdigit) != 0 && (digitab[ptr[2]] & ctype_xdigit) != 0
737: && (digitab[ptr[3]] & ctype_xdigit) != 0 && (digitab[ptr[4]] & ctype_xdigit) != 0)
738: {
739: c = 0;
740: for (i = 0; i < 4; ++i)
741: {
742: register int cc = *(++ptr);
743: #ifndef EBCDIC /* ASCII/UTF-8 coding */
744: if (cc >= CHAR_a) cc -= 32; /* Convert to upper case */
745: c = (c << 4) + cc - ((cc < CHAR_A)? CHAR_0 : (CHAR_A - 10));
746: #else /* EBCDIC coding */
747: if (cc >= CHAR_a && cc <= CHAR_z) cc += 64; /* Convert to upper case */
748: c = (c << 4) + cc - ((cc >= CHAR_0)? CHAR_0 : (CHAR_A - 10));
749: #endif
750: }
751: }
752: }
753: else
754: *errorcodeptr = ERR37;
755: break;
756:
757: case CHAR_U:
758: /* In JavaScript, \U is an uppercase U letter. */
759: if ((options & PCRE_JAVASCRIPT_COMPAT) == 0) *errorcodeptr = ERR37;
760: break;
761:
762: /* In a character class, \g is just a literal "g". Outside a character
763: class, \g must be followed by one of a number of specific things:
764:
765: (1) A number, either plain or braced. If positive, it is an absolute
766: backreference. If negative, it is a relative backreference. This is a Perl
767: 5.10 feature.
768:
769: (2) Perl 5.10 also supports \g{name} as a reference to a named group. This
770: is part of Perl's movement towards a unified syntax for back references. As
771: this is synonymous with \k{name}, we fudge it up by pretending it really
772: was \k.
773:
774: (3) For Oniguruma compatibility we also support \g followed by a name or a
775: number either in angle brackets or in single quotes. However, these are
776: (possibly recursive) subroutine calls, _not_ backreferences. Just return
777: the -ESC_g code (cf \k). */
778:
779: case CHAR_g:
780: if (isclass) break;
781: if (ptr[1] == CHAR_LESS_THAN_SIGN || ptr[1] == CHAR_APOSTROPHE)
782: {
783: c = -ESC_g;
784: break;
785: }
786:
787: /* Handle the Perl-compatible cases */
788:
789: if (ptr[1] == CHAR_LEFT_CURLY_BRACKET)
790: {
791: const uschar *p;
792: for (p = ptr+2; *p != 0 && *p != CHAR_RIGHT_CURLY_BRACKET; p++)
793: if (*p != CHAR_MINUS && (digitab[*p] & ctype_digit) == 0) break;
794: if (*p != 0 && *p != CHAR_RIGHT_CURLY_BRACKET)
795: {
796: c = -ESC_k;
797: break;
798: }
799: braced = TRUE;
800: ptr++;
801: }
802: else braced = FALSE;
803:
804: if (ptr[1] == CHAR_MINUS)
805: {
806: negated = TRUE;
807: ptr++;
808: }
809: else negated = FALSE;
810:
811: c = 0;
812: while ((digitab[ptr[1]] & ctype_digit) != 0)
813: c = c * 10 + *(++ptr) - CHAR_0;
814:
815: if (c < 0) /* Integer overflow */
816: {
817: *errorcodeptr = ERR61;
818: break;
819: }
820:
821: if (braced && *(++ptr) != CHAR_RIGHT_CURLY_BRACKET)
822: {
823: *errorcodeptr = ERR57;
824: break;
825: }
826:
827: if (c == 0)
828: {
829: *errorcodeptr = ERR58;
830: break;
831: }
832:
833: if (negated)
834: {
835: if (c > bracount)
836: {
837: *errorcodeptr = ERR15;
838: break;
839: }
840: c = bracount - (c - 1);
841: }
842:
843: c = -(ESC_REF + c);
844: break;
845:
846: /* The handling of escape sequences consisting of a string of digits
847: starting with one that is not zero is not straightforward. By experiment,
848: the way Perl works seems to be as follows:
849:
850: Outside a character class, the digits are read as a decimal number. If the
851: number is less than 10, or if there are that many previous extracting
852: left brackets, then it is a back reference. Otherwise, up to three octal
853: digits are read to form an escaped byte. Thus \123 is likely to be octal
854: 123 (cf \0123, which is octal 012 followed by the literal 3). If the octal
855: value is greater than 377, the least significant 8 bits are taken. Inside a
856: character class, \ followed by a digit is always an octal number. */
857:
858: case CHAR_1: case CHAR_2: case CHAR_3: case CHAR_4: case CHAR_5:
859: case CHAR_6: case CHAR_7: case CHAR_8: case CHAR_9:
860:
861: if (!isclass)
862: {
863: oldptr = ptr;
864: c -= CHAR_0;
865: while ((digitab[ptr[1]] & ctype_digit) != 0)
866: c = c * 10 + *(++ptr) - CHAR_0;
867: if (c < 0) /* Integer overflow */
868: {
869: *errorcodeptr = ERR61;
870: break;
871: }
872: if (c < 10 || c <= bracount)
873: {
874: c = -(ESC_REF + c);
875: break;
876: }
877: ptr = oldptr; /* Put the pointer back and fall through */
878: }
879:
880: /* Handle an octal number following \. If the first digit is 8 or 9, Perl
881: generates a binary zero byte and treats the digit as a following literal.
882: Thus we have to pull back the pointer by one. */
883:
884: if ((c = *ptr) >= CHAR_8)
885: {
886: ptr--;
887: c = 0;
888: break;
889: }
890:
891: /* \0 always starts an octal number, but we may drop through to here with a
892: larger first octal digit. The original code used just to take the least
893: significant 8 bits of octal numbers (I think this is what early Perls used
894: to do). Nowadays we allow for larger numbers in UTF-8 mode, but no more
895: than 3 octal digits. */
896:
897: case CHAR_0:
898: c -= CHAR_0;
899: while(i++ < 2 && ptr[1] >= CHAR_0 && ptr[1] <= CHAR_7)
900: c = c * 8 + *(++ptr) - CHAR_0;
901: if (!utf8 && c > 255) *errorcodeptr = ERR51;
902: break;
903:
904: /* \x is complicated. \x{ddd} is a character number which can be greater
905: than 0xff in utf8 mode, but only if the ddd are hex digits. If not, { is
906: treated as a data character. */
907:
908: case CHAR_x:
909: if ((options & PCRE_JAVASCRIPT_COMPAT) != 0)
910: {
911: /* In JavaScript, \x must be followed by two hexadecimal numbers.
912: Otherwise it is a lowercase x letter. */
913: if ((digitab[ptr[1]] & ctype_xdigit) != 0 && (digitab[ptr[2]] & ctype_xdigit) != 0)
914: {
915: c = 0;
916: for (i = 0; i < 2; ++i)
917: {
918: register int cc = *(++ptr);
919: #ifndef EBCDIC /* ASCII/UTF-8 coding */
920: if (cc >= CHAR_a) cc -= 32; /* Convert to upper case */
921: c = (c << 4) + cc - ((cc < CHAR_A)? CHAR_0 : (CHAR_A - 10));
922: #else /* EBCDIC coding */
923: if (cc >= CHAR_a && cc <= CHAR_z) cc += 64; /* Convert to upper case */
924: c = (c << 4) + cc - ((cc >= CHAR_0)? CHAR_0 : (CHAR_A - 10));
925: #endif
926: }
927: }
928: break;
929: }
930:
931: if (ptr[1] == CHAR_LEFT_CURLY_BRACKET)
932: {
933: const uschar *pt = ptr + 2;
934: int count = 0;
935:
936: c = 0;
937: while ((digitab[*pt] & ctype_xdigit) != 0)
938: {
939: register int cc = *pt++;
940: if (c == 0 && cc == CHAR_0) continue; /* Leading zeroes */
941: count++;
942:
943: #ifndef EBCDIC /* ASCII/UTF-8 coding */
944: if (cc >= CHAR_a) cc -= 32; /* Convert to upper case */
945: c = (c << 4) + cc - ((cc < CHAR_A)? CHAR_0 : (CHAR_A - 10));
946: #else /* EBCDIC coding */
947: if (cc >= CHAR_a && cc <= CHAR_z) cc += 64; /* Convert to upper case */
948: c = (c << 4) + cc - ((cc >= CHAR_0)? CHAR_0 : (CHAR_A - 10));
949: #endif
950: }
951:
952: if (*pt == CHAR_RIGHT_CURLY_BRACKET)
953: {
954: if (c < 0 || count > (utf8? 8 : 2)) *errorcodeptr = ERR34;
955: ptr = pt;
956: break;
957: }
958:
959: /* If the sequence of hex digits does not end with '}', then we don't
960: recognize this construct; fall through to the normal \x handling. */
961: }
962:
963: /* Read just a single-byte hex-defined char */
964:
965: c = 0;
966: while (i++ < 2 && (digitab[ptr[1]] & ctype_xdigit) != 0)
967: {
968: int cc; /* Some compilers don't like */
969: cc = *(++ptr); /* ++ in initializers */
970: #ifndef EBCDIC /* ASCII/UTF-8 coding */
971: if (cc >= CHAR_a) cc -= 32; /* Convert to upper case */
972: c = c * 16 + cc - ((cc < CHAR_A)? CHAR_0 : (CHAR_A - 10));
973: #else /* EBCDIC coding */
974: if (cc <= CHAR_z) cc += 64; /* Convert to upper case */
975: c = c * 16 + cc - ((cc >= CHAR_0)? CHAR_0 : (CHAR_A - 10));
976: #endif
977: }
978: break;
979:
980: /* For \c, a following letter is upper-cased; then the 0x40 bit is flipped.
981: An error is given if the byte following \c is not an ASCII character. This
982: coding is ASCII-specific, but then the whole concept of \cx is
983: ASCII-specific. (However, an EBCDIC equivalent has now been added.) */
984:
985: case CHAR_c:
986: c = *(++ptr);
987: if (c == 0)
988: {
989: *errorcodeptr = ERR2;
990: break;
991: }
992: #ifndef EBCDIC /* ASCII/UTF-8 coding */
993: if (c > 127) /* Excludes all non-ASCII in either mode */
994: {
995: *errorcodeptr = ERR68;
996: break;
997: }
998: if (c >= CHAR_a && c <= CHAR_z) c -= 32;
999: c ^= 0x40;
1000: #else /* EBCDIC coding */
1001: if (c >= CHAR_a && c <= CHAR_z) c += 64;
1002: c ^= 0xC0;
1003: #endif
1004: break;
1005:
1006: /* PCRE_EXTRA enables extensions to Perl in the matter of escapes. Any
1007: other alphanumeric following \ is an error if PCRE_EXTRA was set;
1008: otherwise, for Perl compatibility, it is a literal. This code looks a bit
1009: odd, but there used to be some cases other than the default, and there may
1010: be again in future, so I haven't "optimized" it. */
1011:
1012: default:
1013: if ((options & PCRE_EXTRA) != 0) switch(c)
1014: {
1015: default:
1016: *errorcodeptr = ERR3;
1017: break;
1018: }
1019: break;
1020: }
1021: }
1022:
1023: /* Perl supports \N{name} for character names, as well as plain \N for "not
1024: newline". PCRE does not support \N{name}. However, it does support
1025: quantification such as \N{2,3}. */
1026:
1027: if (c == -ESC_N && ptr[1] == CHAR_LEFT_CURLY_BRACKET &&
1028: !is_counted_repeat(ptr+2))
1029: *errorcodeptr = ERR37;
1030:
1031: /* If PCRE_UCP is set, we change the values for \d etc. */
1032:
1033: if ((options & PCRE_UCP) != 0 && c <= -ESC_D && c >= -ESC_w)
1034: c -= (ESC_DU - ESC_D);
1035:
1036: /* Set the pointer to the final character before returning. */
1037:
1038: *ptrptr = ptr;
1039: return c;
1040: }
1041:
1042:
1043:
1044: #ifdef SUPPORT_UCP
1045: /*************************************************
1046: * Handle \P and \p *
1047: *************************************************/
1048:
1049: /* This function is called after \P or \p has been encountered, provided that
1050: PCRE is compiled with support for Unicode properties. On entry, ptrptr is
1051: pointing at the P or p. On exit, it is pointing at the final character of the
1052: escape sequence.
1053:
1054: Argument:
1055: ptrptr points to the pattern position pointer
1056: negptr points to a boolean that is set TRUE for negation else FALSE
1057: dptr points to an int that is set to the detailed property value
1058: errorcodeptr points to the error code variable
1059:
1060: Returns: type value from ucp_type_table, or -1 for an invalid type
1061: */
1062:
1063: static int
1064: get_ucp(const uschar **ptrptr, BOOL *negptr, int *dptr, int *errorcodeptr)
1065: {
1066: int c, i, bot, top;
1067: const uschar *ptr = *ptrptr;
1068: char name[32];
1069:
1070: c = *(++ptr);
1071: if (c == 0) goto ERROR_RETURN;
1072:
1073: *negptr = FALSE;
1074:
1075: /* \P or \p can be followed by a name in {}, optionally preceded by ^ for
1076: negation. */
1077:
1078: if (c == CHAR_LEFT_CURLY_BRACKET)
1079: {
1080: if (ptr[1] == CHAR_CIRCUMFLEX_ACCENT)
1081: {
1082: *negptr = TRUE;
1083: ptr++;
1084: }
1085: for (i = 0; i < (int)sizeof(name) - 1; i++)
1086: {
1087: c = *(++ptr);
1088: if (c == 0) goto ERROR_RETURN;
1089: if (c == CHAR_RIGHT_CURLY_BRACKET) break;
1090: name[i] = c;
1091: }
1092: if (c != CHAR_RIGHT_CURLY_BRACKET) goto ERROR_RETURN;
1093: name[i] = 0;
1094: }
1095:
1096: /* Otherwise there is just one following character */
1097:
1098: else
1099: {
1100: name[0] = c;
1101: name[1] = 0;
1102: }
1103:
1104: *ptrptr = ptr;
1105:
1106: /* Search for a recognized property name using binary chop */
1107:
1108: bot = 0;
1109: top = _pcre_utt_size;
1110:
1111: while (bot < top)
1112: {
1113: i = (bot + top) >> 1;
1114: c = strcmp(name, _pcre_utt_names + _pcre_utt[i].name_offset);
1115: if (c == 0)
1116: {
1117: *dptr = _pcre_utt[i].value;
1118: return _pcre_utt[i].type;
1119: }
1120: if (c > 0) bot = i + 1; else top = i;
1121: }
1122:
1123: *errorcodeptr = ERR47;
1124: *ptrptr = ptr;
1125: return -1;
1126:
1127: ERROR_RETURN:
1128: *errorcodeptr = ERR46;
1129: *ptrptr = ptr;
1130: return -1;
1131: }
1132: #endif
1133:
1134:
1135:
1136:
1137: /*************************************************
1138: * Read repeat counts *
1139: *************************************************/
1140:
1141: /* Read an item of the form {n,m} and return the values. This is called only
1142: after is_counted_repeat() has confirmed that a repeat-count quantifier exists,
1143: so the syntax is guaranteed to be correct, but we need to check the values.
1144:
1145: Arguments:
1146: p pointer to first char after '{'
1147: minp pointer to int for min
1148: maxp pointer to int for max
1149: returned as -1 if no max
1150: errorcodeptr points to error code variable
1151:
1152: Returns: pointer to '}' on success;
1153: current ptr on error, with errorcodeptr set non-zero
1154: */
1155:
1156: static const uschar *
1157: read_repeat_counts(const uschar *p, int *minp, int *maxp, int *errorcodeptr)
1158: {
1159: int min = 0;
1160: int max = -1;
1161:
1162: /* Read the minimum value and do a paranoid check: a negative value indicates
1163: an integer overflow. */
1164:
1165: while ((digitab[*p] & ctype_digit) != 0) min = min * 10 + *p++ - CHAR_0;
1166: if (min < 0 || min > 65535)
1167: {
1168: *errorcodeptr = ERR5;
1169: return p;
1170: }
1171:
1172: /* Read the maximum value if there is one, and again do a paranoid on its size.
1173: Also, max must not be less than min. */
1174:
1175: if (*p == CHAR_RIGHT_CURLY_BRACKET) max = min; else
1176: {
1177: if (*(++p) != CHAR_RIGHT_CURLY_BRACKET)
1178: {
1179: max = 0;
1180: while((digitab[*p] & ctype_digit) != 0) max = max * 10 + *p++ - CHAR_0;
1181: if (max < 0 || max > 65535)
1182: {
1183: *errorcodeptr = ERR5;
1184: return p;
1185: }
1186: if (max < min)
1187: {
1188: *errorcodeptr = ERR4;
1189: return p;
1190: }
1191: }
1192: }
1193:
1194: /* Fill in the required variables, and pass back the pointer to the terminating
1195: '}'. */
1196:
1197: *minp = min;
1198: *maxp = max;
1199: return p;
1200: }
1201:
1202:
1203:
1204: /*************************************************
1205: * Subroutine for finding forward reference *
1206: *************************************************/
1207:
1208: /* This recursive function is called only from find_parens() below. The
1209: top-level call starts at the beginning of the pattern. All other calls must
1210: start at a parenthesis. It scans along a pattern's text looking for capturing
1211: subpatterns, and counting them. If it finds a named pattern that matches the
1212: name it is given, it returns its number. Alternatively, if the name is NULL, it
1213: returns when it reaches a given numbered subpattern. Recursion is used to keep
1214: track of subpatterns that reset the capturing group numbers - the (?| feature.
1215:
1216: This function was originally called only from the second pass, in which we know
1217: that if (?< or (?' or (?P< is encountered, the name will be correctly
1218: terminated because that is checked in the first pass. There is now one call to
1219: this function in the first pass, to check for a recursive back reference by
1220: name (so that we can make the whole group atomic). In this case, we need check
1221: only up to the current position in the pattern, and that is still OK because
1222: and previous occurrences will have been checked. To make this work, the test
1223: for "end of pattern" is a check against cd->end_pattern in the main loop,
1224: instead of looking for a binary zero. This means that the special first-pass
1225: call can adjust cd->end_pattern temporarily. (Checks for binary zero while
1226: processing items within the loop are OK, because afterwards the main loop will
1227: terminate.)
1228:
1229: Arguments:
1230: ptrptr address of the current character pointer (updated)
1231: cd compile background data
1232: name name to seek, or NULL if seeking a numbered subpattern
1233: lorn name length, or subpattern number if name is NULL
1234: xmode TRUE if we are in /x mode
1235: utf8 TRUE if we are in UTF-8 mode
1236: count pointer to the current capturing subpattern number (updated)
1237:
1238: Returns: the number of the named subpattern, or -1 if not found
1239: */
1240:
1241: static int
1242: find_parens_sub(uschar **ptrptr, compile_data *cd, const uschar *name, int lorn,
1243: BOOL xmode, BOOL utf8, int *count)
1244: {
1245: uschar *ptr = *ptrptr;
1246: int start_count = *count;
1247: int hwm_count = start_count;
1248: BOOL dup_parens = FALSE;
1249:
1250: /* If the first character is a parenthesis, check on the type of group we are
1251: dealing with. The very first call may not start with a parenthesis. */
1252:
1253: if (ptr[0] == CHAR_LEFT_PARENTHESIS)
1254: {
1255: /* Handle specials such as (*SKIP) or (*UTF8) etc. */
1256:
1257: if (ptr[1] == CHAR_ASTERISK) ptr += 2;
1258:
1259: /* Handle a normal, unnamed capturing parenthesis. */
1260:
1261: else if (ptr[1] != CHAR_QUESTION_MARK)
1262: {
1263: *count += 1;
1264: if (name == NULL && *count == lorn) return *count;
1265: ptr++;
1266: }
1267:
1268: /* All cases now have (? at the start. Remember when we are in a group
1269: where the parenthesis numbers are duplicated. */
1270:
1271: else if (ptr[2] == CHAR_VERTICAL_LINE)
1272: {
1273: ptr += 3;
1274: dup_parens = TRUE;
1275: }
1276:
1277: /* Handle comments; all characters are allowed until a ket is reached. */
1278:
1279: else if (ptr[2] == CHAR_NUMBER_SIGN)
1280: {
1281: for (ptr += 3; *ptr != 0; ptr++) if (*ptr == CHAR_RIGHT_PARENTHESIS) break;
1282: goto FAIL_EXIT;
1283: }
1284:
1285: /* Handle a condition. If it is an assertion, just carry on so that it
1286: is processed as normal. If not, skip to the closing parenthesis of the
1287: condition (there can't be any nested parens). */
1288:
1289: else if (ptr[2] == CHAR_LEFT_PARENTHESIS)
1290: {
1291: ptr += 2;
1292: if (ptr[1] != CHAR_QUESTION_MARK)
1293: {
1294: while (*ptr != 0 && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++;
1295: if (*ptr != 0) ptr++;
1296: }
1297: }
1298:
1299: /* Start with (? but not a condition. */
1300:
1301: else
1302: {
1303: ptr += 2;
1304: if (*ptr == CHAR_P) ptr++; /* Allow optional P */
1305:
1306: /* We have to disambiguate (?<! and (?<= from (?<name> for named groups */
1307:
1308: if ((*ptr == CHAR_LESS_THAN_SIGN && ptr[1] != CHAR_EXCLAMATION_MARK &&
1309: ptr[1] != CHAR_EQUALS_SIGN) || *ptr == CHAR_APOSTROPHE)
1310: {
1311: int term;
1312: const uschar *thisname;
1313: *count += 1;
1314: if (name == NULL && *count == lorn) return *count;
1315: term = *ptr++;
1316: if (term == CHAR_LESS_THAN_SIGN) term = CHAR_GREATER_THAN_SIGN;
1317: thisname = ptr;
1318: while (*ptr != term) ptr++;
1319: if (name != NULL && lorn == ptr - thisname &&
1320: strncmp((const char *)name, (const char *)thisname, lorn) == 0)
1321: return *count;
1322: term++;
1323: }
1324: }
1325: }
1326:
1327: /* Past any initial parenthesis handling, scan for parentheses or vertical
1328: bars. Stop if we get to cd->end_pattern. Note that this is important for the
1329: first-pass call when this value is temporarily adjusted to stop at the current
1330: position. So DO NOT change this to a test for binary zero. */
1331:
1332: for (; ptr < cd->end_pattern; ptr++)
1333: {
1334: /* Skip over backslashed characters and also entire \Q...\E */
1335:
1336: if (*ptr == CHAR_BACKSLASH)
1337: {
1338: if (*(++ptr) == 0) goto FAIL_EXIT;
1339: if (*ptr == CHAR_Q) for (;;)
1340: {
1341: while (*(++ptr) != 0 && *ptr != CHAR_BACKSLASH) {};
1342: if (*ptr == 0) goto FAIL_EXIT;
1343: if (*(++ptr) == CHAR_E) break;
1344: }
1345: continue;
1346: }
1347:
1348: /* Skip over character classes; this logic must be similar to the way they
1349: are handled for real. If the first character is '^', skip it. Also, if the
1350: first few characters (either before or after ^) are \Q\E or \E we skip them
1351: too. This makes for compatibility with Perl. Note the use of STR macros to
1352: encode "Q\\E" so that it works in UTF-8 on EBCDIC platforms. */
1353:
1354: if (*ptr == CHAR_LEFT_SQUARE_BRACKET)
1355: {
1356: BOOL negate_class = FALSE;
1357: for (;;)
1358: {
1359: if (ptr[1] == CHAR_BACKSLASH)
1360: {
1361: if (ptr[2] == CHAR_E)
1362: ptr+= 2;
1363: else if (strncmp((const char *)ptr+2,
1364: STR_Q STR_BACKSLASH STR_E, 3) == 0)
1365: ptr += 4;
1366: else
1367: break;
1368: }
1369: else if (!negate_class && ptr[1] == CHAR_CIRCUMFLEX_ACCENT)
1370: {
1371: negate_class = TRUE;
1372: ptr++;
1373: }
1374: else break;
1375: }
1376:
1377: /* If the next character is ']', it is a data character that must be
1378: skipped, except in JavaScript compatibility mode. */
1379:
1380: if (ptr[1] == CHAR_RIGHT_SQUARE_BRACKET &&
1381: (cd->external_options & PCRE_JAVASCRIPT_COMPAT) == 0)
1382: ptr++;
1383:
1384: while (*(++ptr) != CHAR_RIGHT_SQUARE_BRACKET)
1385: {
1386: if (*ptr == 0) return -1;
1387: if (*ptr == CHAR_BACKSLASH)
1388: {
1389: if (*(++ptr) == 0) goto FAIL_EXIT;
1390: if (*ptr == CHAR_Q) for (;;)
1391: {
1392: while (*(++ptr) != 0 && *ptr != CHAR_BACKSLASH) {};
1393: if (*ptr == 0) goto FAIL_EXIT;
1394: if (*(++ptr) == CHAR_E) break;
1395: }
1396: continue;
1397: }
1398: }
1399: continue;
1400: }
1401:
1402: /* Skip comments in /x mode */
1403:
1404: if (xmode && *ptr == CHAR_NUMBER_SIGN)
1405: {
1406: ptr++;
1407: while (*ptr != 0)
1408: {
1409: if (IS_NEWLINE(ptr)) { ptr += cd->nllen - 1; break; }
1410: ptr++;
1411: #ifdef SUPPORT_UTF8
1412: if (utf8) while ((*ptr & 0xc0) == 0x80) ptr++;
1413: #endif
1414: }
1415: if (*ptr == 0) goto FAIL_EXIT;
1416: continue;
1417: }
1418:
1419: /* Check for the special metacharacters */
1420:
1421: if (*ptr == CHAR_LEFT_PARENTHESIS)
1422: {
1423: int rc = find_parens_sub(&ptr, cd, name, lorn, xmode, utf8, count);
1424: if (rc > 0) return rc;
1425: if (*ptr == 0) goto FAIL_EXIT;
1426: }
1427:
1428: else if (*ptr == CHAR_RIGHT_PARENTHESIS)
1429: {
1430: if (dup_parens && *count < hwm_count) *count = hwm_count;
1431: goto FAIL_EXIT;
1432: }
1433:
1434: else if (*ptr == CHAR_VERTICAL_LINE && dup_parens)
1435: {
1436: if (*count > hwm_count) hwm_count = *count;
1437: *count = start_count;
1438: }
1439: }
1440:
1441: FAIL_EXIT:
1442: *ptrptr = ptr;
1443: return -1;
1444: }
1445:
1446:
1447:
1448:
1449: /*************************************************
1450: * Find forward referenced subpattern *
1451: *************************************************/
1452:
1453: /* This function scans along a pattern's text looking for capturing
1454: subpatterns, and counting them. If it finds a named pattern that matches the
1455: name it is given, it returns its number. Alternatively, if the name is NULL, it
1456: returns when it reaches a given numbered subpattern. This is used for forward
1457: references to subpatterns. We used to be able to start this scan from the
1458: current compiling point, using the current count value from cd->bracount, and
1459: do it all in a single loop, but the addition of the possibility of duplicate
1460: subpattern numbers means that we have to scan from the very start, in order to
1461: take account of such duplicates, and to use a recursive function to keep track
1462: of the different types of group.
1463:
1464: Arguments:
1465: cd compile background data
1466: name name to seek, or NULL if seeking a numbered subpattern
1467: lorn name length, or subpattern number if name is NULL
1468: xmode TRUE if we are in /x mode
1469: utf8 TRUE if we are in UTF-8 mode
1470:
1471: Returns: the number of the found subpattern, or -1 if not found
1472: */
1473:
1474: static int
1475: find_parens(compile_data *cd, const uschar *name, int lorn, BOOL xmode,
1476: BOOL utf8)
1477: {
1478: uschar *ptr = (uschar *)cd->start_pattern;
1479: int count = 0;
1480: int rc;
1481:
1482: /* If the pattern does not start with an opening parenthesis, the first call
1483: to find_parens_sub() will scan right to the end (if necessary). However, if it
1484: does start with a parenthesis, find_parens_sub() will return when it hits the
1485: matching closing parens. That is why we have to have a loop. */
1486:
1487: for (;;)
1488: {
1489: rc = find_parens_sub(&ptr, cd, name, lorn, xmode, utf8, &count);
1490: if (rc > 0 || *ptr++ == 0) break;
1491: }
1492:
1493: return rc;
1494: }
1495:
1496:
1497:
1498:
1499: /*************************************************
1500: * Find first significant op code *
1501: *************************************************/
1502:
1503: /* This is called by several functions that scan a compiled expression looking
1504: for a fixed first character, or an anchoring op code etc. It skips over things
1505: that do not influence this. For some calls, it makes sense to skip negative
1506: forward and all backward assertions, and also the \b assertion; for others it
1507: does not.
1508:
1509: Arguments:
1510: code pointer to the start of the group
1511: skipassert TRUE if certain assertions are to be skipped
1512:
1513: Returns: pointer to the first significant opcode
1514: */
1515:
1516: static const uschar*
1517: first_significant_code(const uschar *code, BOOL skipassert)
1518: {
1519: for (;;)
1520: {
1521: switch ((int)*code)
1522: {
1523: case OP_ASSERT_NOT:
1524: case OP_ASSERTBACK:
1525: case OP_ASSERTBACK_NOT:
1526: if (!skipassert) return code;
1527: do code += GET(code, 1); while (*code == OP_ALT);
1528: code += _pcre_OP_lengths[*code];
1529: break;
1530:
1531: case OP_WORD_BOUNDARY:
1532: case OP_NOT_WORD_BOUNDARY:
1533: if (!skipassert) return code;
1534: /* Fall through */
1535:
1536: case OP_CALLOUT:
1537: case OP_CREF:
1538: case OP_NCREF:
1539: case OP_RREF:
1540: case OP_NRREF:
1541: case OP_DEF:
1542: code += _pcre_OP_lengths[*code];
1543: break;
1544:
1545: default:
1546: return code;
1547: }
1548: }
1549: /* Control never reaches here */
1550: }
1551:
1552:
1553:
1554:
1555: /*************************************************
1556: * Find the fixed length of a branch *
1557: *************************************************/
1558:
1559: /* Scan a branch and compute the fixed length of subject that will match it,
1560: if the length is fixed. This is needed for dealing with backward assertions.
1561: In UTF8 mode, the result is in characters rather than bytes. The branch is
1562: temporarily terminated with OP_END when this function is called.
1563:
1564: This function is called when a backward assertion is encountered, so that if it
1565: fails, the error message can point to the correct place in the pattern.
1566: However, we cannot do this when the assertion contains subroutine calls,
1567: because they can be forward references. We solve this by remembering this case
1568: and doing the check at the end; a flag specifies which mode we are running in.
1569:
1570: Arguments:
1571: code points to the start of the pattern (the bracket)
1572: utf8 TRUE in UTF-8 mode
1573: atend TRUE if called when the pattern is complete
1574: cd the "compile data" structure
1575:
1576: Returns: the fixed length,
1577: or -1 if there is no fixed length,
1578: or -2 if \C was encountered (in UTF-8 mode only)
1579: or -3 if an OP_RECURSE item was encountered and atend is FALSE
1580: or -4 if an unknown opcode was encountered (internal error)
1581: */
1582:
1583: static int
1584: find_fixedlength(uschar *code, BOOL utf8, BOOL atend, compile_data *cd)
1585: {
1586: int length = -1;
1587:
1588: register int branchlength = 0;
1589: register uschar *cc = code + 1 + LINK_SIZE;
1590:
1591: /* Scan along the opcodes for this branch. If we get to the end of the
1592: branch, check the length against that of the other branches. */
1593:
1594: for (;;)
1595: {
1596: int d;
1597: uschar *ce, *cs;
1598: register int op = *cc;
1599: switch (op)
1600: {
1601: /* We only need to continue for OP_CBRA (normal capturing bracket) and
1602: OP_BRA (normal non-capturing bracket) because the other variants of these
1603: opcodes are all concerned with unlimited repeated groups, which of course
1604: are not of fixed length. */
1605:
1606: case OP_CBRA:
1607: case OP_BRA:
1608: case OP_ONCE:
1609: case OP_ONCE_NC:
1610: case OP_COND:
1611: d = find_fixedlength(cc + ((op == OP_CBRA)? 2:0), utf8, atend, cd);
1612: if (d < 0) return d;
1613: branchlength += d;
1614: do cc += GET(cc, 1); while (*cc == OP_ALT);
1615: cc += 1 + LINK_SIZE;
1616: break;
1617:
1618: /* Reached end of a branch; if it's a ket it is the end of a nested call.
1619: If it's ALT it is an alternation in a nested call. An ACCEPT is effectively
1620: an ALT. If it is END it's the end of the outer call. All can be handled by
1621: the same code. Note that we must not include the OP_KETRxxx opcodes here,
1622: because they all imply an unlimited repeat. */
1623:
1624: case OP_ALT:
1625: case OP_KET:
1626: case OP_END:
1627: case OP_ACCEPT:
1628: case OP_ASSERT_ACCEPT:
1629: if (length < 0) length = branchlength;
1630: else if (length != branchlength) return -1;
1631: if (*cc != OP_ALT) return length;
1632: cc += 1 + LINK_SIZE;
1633: branchlength = 0;
1634: break;
1635:
1636: /* A true recursion implies not fixed length, but a subroutine call may
1637: be OK. If the subroutine is a forward reference, we can't deal with
1638: it until the end of the pattern, so return -3. */
1639:
1640: case OP_RECURSE:
1641: if (!atend) return -3;
1642: cs = ce = (uschar *)cd->start_code + GET(cc, 1); /* Start subpattern */
1643: do ce += GET(ce, 1); while (*ce == OP_ALT); /* End subpattern */
1644: if (cc > cs && cc < ce) return -1; /* Recursion */
1645: d = find_fixedlength(cs + 2, utf8, atend, cd);
1646: if (d < 0) return d;
1647: branchlength += d;
1648: cc += 1 + LINK_SIZE;
1649: break;
1650:
1651: /* Skip over assertive subpatterns */
1652:
1653: case OP_ASSERT:
1654: case OP_ASSERT_NOT:
1655: case OP_ASSERTBACK:
1656: case OP_ASSERTBACK_NOT:
1657: do cc += GET(cc, 1); while (*cc == OP_ALT);
1658: /* Fall through */
1659:
1660: /* Skip over things that don't match chars */
1661:
1662: case OP_MARK:
1663: case OP_PRUNE_ARG:
1664: case OP_SKIP_ARG:
1665: case OP_THEN_ARG:
1666: cc += cc[1] + _pcre_OP_lengths[*cc];
1667: break;
1668:
1669: case OP_CALLOUT:
1670: case OP_CIRC:
1671: case OP_CIRCM:
1672: case OP_CLOSE:
1673: case OP_COMMIT:
1674: case OP_CREF:
1675: case OP_DEF:
1676: case OP_DOLL:
1677: case OP_DOLLM:
1678: case OP_EOD:
1679: case OP_EODN:
1680: case OP_FAIL:
1681: case OP_NCREF:
1682: case OP_NRREF:
1683: case OP_NOT_WORD_BOUNDARY:
1684: case OP_PRUNE:
1685: case OP_REVERSE:
1686: case OP_RREF:
1687: case OP_SET_SOM:
1688: case OP_SKIP:
1689: case OP_SOD:
1690: case OP_SOM:
1691: case OP_THEN:
1692: case OP_WORD_BOUNDARY:
1693: cc += _pcre_OP_lengths[*cc];
1694: break;
1695:
1696: /* Handle literal characters */
1697:
1698: case OP_CHAR:
1699: case OP_CHARI:
1700: case OP_NOT:
1701: case OP_NOTI:
1702: branchlength++;
1703: cc += 2;
1704: #ifdef SUPPORT_UTF8
1705: if (utf8 && cc[-1] >= 0xc0) cc += _pcre_utf8_table4[cc[-1] & 0x3f];
1706: #endif
1707: break;
1708:
1709: /* Handle exact repetitions. The count is already in characters, but we
1710: need to skip over a multibyte character in UTF8 mode. */
1711:
1712: case OP_EXACT:
1713: case OP_EXACTI:
1714: case OP_NOTEXACT:
1715: case OP_NOTEXACTI:
1716: branchlength += GET2(cc,1);
1717: cc += 4;
1718: #ifdef SUPPORT_UTF8
1719: if (utf8 && cc[-1] >= 0xc0) cc += _pcre_utf8_table4[cc[-1] & 0x3f];
1720: #endif
1721: break;
1722:
1723: case OP_TYPEEXACT:
1724: branchlength += GET2(cc,1);
1725: if (cc[3] == OP_PROP || cc[3] == OP_NOTPROP) cc += 2;
1726: cc += 4;
1727: break;
1728:
1729: /* Handle single-char matchers */
1730:
1731: case OP_PROP:
1732: case OP_NOTPROP:
1733: cc += 2;
1734: /* Fall through */
1735:
1736: case OP_HSPACE:
1737: case OP_VSPACE:
1738: case OP_NOT_HSPACE:
1739: case OP_NOT_VSPACE:
1740: case OP_NOT_DIGIT:
1741: case OP_DIGIT:
1742: case OP_NOT_WHITESPACE:
1743: case OP_WHITESPACE:
1744: case OP_NOT_WORDCHAR:
1745: case OP_WORDCHAR:
1746: case OP_ANY:
1747: case OP_ALLANY:
1748: branchlength++;
1749: cc++;
1750: break;
1751:
1752: /* The single-byte matcher isn't allowed. This only happens in UTF-8 mode;
1753: otherwise \C is coded as OP_ALLANY. */
1754:
1755: case OP_ANYBYTE:
1756: return -2;
1757:
1758: /* Check a class for variable quantification */
1759:
1760: #ifdef SUPPORT_UTF8
1761: case OP_XCLASS:
1762: cc += GET(cc, 1) - 33;
1763: /* Fall through */
1764: #endif
1765:
1766: case OP_CLASS:
1767: case OP_NCLASS:
1768: cc += 33;
1769:
1770: switch (*cc)
1771: {
1772: case OP_CRPLUS:
1773: case OP_CRMINPLUS:
1774: case OP_CRSTAR:
1775: case OP_CRMINSTAR:
1776: case OP_CRQUERY:
1777: case OP_CRMINQUERY:
1778: return -1;
1779:
1780: case OP_CRRANGE:
1781: case OP_CRMINRANGE:
1782: if (GET2(cc,1) != GET2(cc,3)) return -1;
1783: branchlength += GET2(cc,1);
1784: cc += 5;
1785: break;
1786:
1787: default:
1788: branchlength++;
1789: }
1790: break;
1791:
1792: /* Anything else is variable length */
1793:
1794: case OP_ANYNL:
1795: case OP_BRAMINZERO:
1796: case OP_BRAPOS:
1797: case OP_BRAPOSZERO:
1798: case OP_BRAZERO:
1799: case OP_CBRAPOS:
1800: case OP_EXTUNI:
1801: case OP_KETRMAX:
1802: case OP_KETRMIN:
1803: case OP_KETRPOS:
1804: case OP_MINPLUS:
1805: case OP_MINPLUSI:
1806: case OP_MINQUERY:
1807: case OP_MINQUERYI:
1808: case OP_MINSTAR:
1809: case OP_MINSTARI:
1810: case OP_MINUPTO:
1811: case OP_MINUPTOI:
1812: case OP_NOTMINPLUS:
1813: case OP_NOTMINPLUSI:
1814: case OP_NOTMINQUERY:
1815: case OP_NOTMINQUERYI:
1816: case OP_NOTMINSTAR:
1817: case OP_NOTMINSTARI:
1818: case OP_NOTMINUPTO:
1819: case OP_NOTMINUPTOI:
1820: case OP_NOTPLUS:
1821: case OP_NOTPLUSI:
1822: case OP_NOTPOSPLUS:
1823: case OP_NOTPOSPLUSI:
1824: case OP_NOTPOSQUERY:
1825: case OP_NOTPOSQUERYI:
1826: case OP_NOTPOSSTAR:
1827: case OP_NOTPOSSTARI:
1828: case OP_NOTPOSUPTO:
1829: case OP_NOTPOSUPTOI:
1830: case OP_NOTQUERY:
1831: case OP_NOTQUERYI:
1832: case OP_NOTSTAR:
1833: case OP_NOTSTARI:
1834: case OP_NOTUPTO:
1835: case OP_NOTUPTOI:
1836: case OP_PLUS:
1837: case OP_PLUSI:
1838: case OP_POSPLUS:
1839: case OP_POSPLUSI:
1840: case OP_POSQUERY:
1841: case OP_POSQUERYI:
1842: case OP_POSSTAR:
1843: case OP_POSSTARI:
1844: case OP_POSUPTO:
1845: case OP_POSUPTOI:
1846: case OP_QUERY:
1847: case OP_QUERYI:
1848: case OP_REF:
1849: case OP_REFI:
1850: case OP_SBRA:
1851: case OP_SBRAPOS:
1852: case OP_SCBRA:
1853: case OP_SCBRAPOS:
1854: case OP_SCOND:
1855: case OP_SKIPZERO:
1856: case OP_STAR:
1857: case OP_STARI:
1858: case OP_TYPEMINPLUS:
1859: case OP_TYPEMINQUERY:
1860: case OP_TYPEMINSTAR:
1861: case OP_TYPEMINUPTO:
1862: case OP_TYPEPLUS:
1863: case OP_TYPEPOSPLUS:
1864: case OP_TYPEPOSQUERY:
1865: case OP_TYPEPOSSTAR:
1866: case OP_TYPEPOSUPTO:
1867: case OP_TYPEQUERY:
1868: case OP_TYPESTAR:
1869: case OP_TYPEUPTO:
1870: case OP_UPTO:
1871: case OP_UPTOI:
1872: return -1;
1873:
1874: /* Catch unrecognized opcodes so that when new ones are added they
1875: are not forgotten, as has happened in the past. */
1876:
1877: default:
1878: return -4;
1879: }
1880: }
1881: /* Control never gets here */
1882: }
1883:
1884:
1885:
1886:
1887: /*************************************************
1888: * Scan compiled regex for specific bracket *
1889: *************************************************/
1890:
1891: /* This little function scans through a compiled pattern until it finds a
1892: capturing bracket with the given number, or, if the number is negative, an
1893: instance of OP_REVERSE for a lookbehind. The function is global in the C sense
1894: so that it can be called from pcre_study() when finding the minimum matching
1895: length.
1896:
1897: Arguments:
1898: code points to start of expression
1899: utf8 TRUE in UTF-8 mode
1900: number the required bracket number or negative to find a lookbehind
1901:
1902: Returns: pointer to the opcode for the bracket, or NULL if not found
1903: */
1904:
1905: const uschar *
1906: _pcre_find_bracket(const uschar *code, BOOL utf8, int number)
1907: {
1908: for (;;)
1909: {
1910: register int c = *code;
1911:
1912: if (c == OP_END) return NULL;
1913:
1914: /* XCLASS is used for classes that cannot be represented just by a bit
1915: map. This includes negated single high-valued characters. The length in
1916: the table is zero; the actual length is stored in the compiled code. */
1917:
1918: if (c == OP_XCLASS) code += GET(code, 1);
1919:
1920: /* Handle recursion */
1921:
1922: else if (c == OP_REVERSE)
1923: {
1924: if (number < 0) return (uschar *)code;
1925: code += _pcre_OP_lengths[c];
1926: }
1927:
1928: /* Handle capturing bracket */
1929:
1930: else if (c == OP_CBRA || c == OP_SCBRA ||
1931: c == OP_CBRAPOS || c == OP_SCBRAPOS)
1932: {
1933: int n = GET2(code, 1+LINK_SIZE);
1934: if (n == number) return (uschar *)code;
1935: code += _pcre_OP_lengths[c];
1936: }
1937:
1938: /* Otherwise, we can get the item's length from the table, except that for
1939: repeated character types, we have to test for \p and \P, which have an extra
1940: two bytes of parameters, and for MARK/PRUNE/SKIP/THEN with an argument, we
1941: must add in its length. */
1942:
1943: else
1944: {
1945: switch(c)
1946: {
1947: case OP_TYPESTAR:
1948: case OP_TYPEMINSTAR:
1949: case OP_TYPEPLUS:
1950: case OP_TYPEMINPLUS:
1951: case OP_TYPEQUERY:
1952: case OP_TYPEMINQUERY:
1953: case OP_TYPEPOSSTAR:
1954: case OP_TYPEPOSPLUS:
1955: case OP_TYPEPOSQUERY:
1956: if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2;
1957: break;
1958:
1959: case OP_TYPEUPTO:
1960: case OP_TYPEMINUPTO:
1961: case OP_TYPEEXACT:
1962: case OP_TYPEPOSUPTO:
1963: if (code[3] == OP_PROP || code[3] == OP_NOTPROP) code += 2;
1964: break;
1965:
1966: case OP_MARK:
1967: case OP_PRUNE_ARG:
1968: case OP_SKIP_ARG:
1969: code += code[1];
1970: break;
1971:
1972: case OP_THEN_ARG:
1973: code += code[1];
1974: break;
1975: }
1976:
1977: /* Add in the fixed length from the table */
1978:
1979: code += _pcre_OP_lengths[c];
1980:
1981: /* In UTF-8 mode, opcodes that are followed by a character may be followed by
1982: a multi-byte character. The length in the table is a minimum, so we have to
1983: arrange to skip the extra bytes. */
1984:
1985: #ifdef SUPPORT_UTF8
1986: if (utf8) switch(c)
1987: {
1988: case OP_CHAR:
1989: case OP_CHARI:
1990: case OP_EXACT:
1991: case OP_EXACTI:
1992: case OP_UPTO:
1993: case OP_UPTOI:
1994: case OP_MINUPTO:
1995: case OP_MINUPTOI:
1996: case OP_POSUPTO:
1997: case OP_POSUPTOI:
1998: case OP_STAR:
1999: case OP_STARI:
2000: case OP_MINSTAR:
2001: case OP_MINSTARI:
2002: case OP_POSSTAR:
2003: case OP_POSSTARI:
2004: case OP_PLUS:
2005: case OP_PLUSI:
2006: case OP_MINPLUS:
2007: case OP_MINPLUSI:
2008: case OP_POSPLUS:
2009: case OP_POSPLUSI:
2010: case OP_QUERY:
2011: case OP_QUERYI:
2012: case OP_MINQUERY:
2013: case OP_MINQUERYI:
2014: case OP_POSQUERY:
2015: case OP_POSQUERYI:
2016: if (code[-1] >= 0xc0) code += _pcre_utf8_table4[code[-1] & 0x3f];
2017: break;
2018: }
2019: #else
2020: (void)(utf8); /* Keep compiler happy by referencing function argument */
2021: #endif
2022: }
2023: }
2024: }
2025:
2026:
2027:
2028: /*************************************************
2029: * Scan compiled regex for recursion reference *
2030: *************************************************/
2031:
2032: /* This little function scans through a compiled pattern until it finds an
2033: instance of OP_RECURSE.
2034:
2035: Arguments:
2036: code points to start of expression
2037: utf8 TRUE in UTF-8 mode
2038:
2039: Returns: pointer to the opcode for OP_RECURSE, or NULL if not found
2040: */
2041:
2042: static const uschar *
2043: find_recurse(const uschar *code, BOOL utf8)
2044: {
2045: for (;;)
2046: {
2047: register int c = *code;
2048: if (c == OP_END) return NULL;
2049: if (c == OP_RECURSE) return code;
2050:
2051: /* XCLASS is used for classes that cannot be represented just by a bit
2052: map. This includes negated single high-valued characters. The length in
2053: the table is zero; the actual length is stored in the compiled code. */
2054:
2055: if (c == OP_XCLASS) code += GET(code, 1);
2056:
2057: /* Otherwise, we can get the item's length from the table, except that for
2058: repeated character types, we have to test for \p and \P, which have an extra
2059: two bytes of parameters, and for MARK/PRUNE/SKIP/THEN with an argument, we
2060: must add in its length. */
2061:
2062: else
2063: {
2064: switch(c)
2065: {
2066: case OP_TYPESTAR:
2067: case OP_TYPEMINSTAR:
2068: case OP_TYPEPLUS:
2069: case OP_TYPEMINPLUS:
2070: case OP_TYPEQUERY:
2071: case OP_TYPEMINQUERY:
2072: case OP_TYPEPOSSTAR:
2073: case OP_TYPEPOSPLUS:
2074: case OP_TYPEPOSQUERY:
2075: if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2;
2076: break;
2077:
2078: case OP_TYPEPOSUPTO:
2079: case OP_TYPEUPTO:
2080: case OP_TYPEMINUPTO:
2081: case OP_TYPEEXACT:
2082: if (code[3] == OP_PROP || code[3] == OP_NOTPROP) code += 2;
2083: break;
2084:
2085: case OP_MARK:
2086: case OP_PRUNE_ARG:
2087: case OP_SKIP_ARG:
2088: code += code[1];
2089: break;
2090:
2091: case OP_THEN_ARG:
2092: code += code[1];
2093: break;
2094: }
2095:
2096: /* Add in the fixed length from the table */
2097:
2098: code += _pcre_OP_lengths[c];
2099:
2100: /* In UTF-8 mode, opcodes that are followed by a character may be followed
2101: by a multi-byte character. The length in the table is a minimum, so we have
2102: to arrange to skip the extra bytes. */
2103:
2104: #ifdef SUPPORT_UTF8
2105: if (utf8) switch(c)
2106: {
2107: case OP_CHAR:
2108: case OP_CHARI:
2109: case OP_EXACT:
2110: case OP_EXACTI:
2111: case OP_UPTO:
2112: case OP_UPTOI:
2113: case OP_MINUPTO:
2114: case OP_MINUPTOI:
2115: case OP_POSUPTO:
2116: case OP_POSUPTOI:
2117: case OP_STAR:
2118: case OP_STARI:
2119: case OP_MINSTAR:
2120: case OP_MINSTARI:
2121: case OP_POSSTAR:
2122: case OP_POSSTARI:
2123: case OP_PLUS:
2124: case OP_PLUSI:
2125: case OP_MINPLUS:
2126: case OP_MINPLUSI:
2127: case OP_POSPLUS:
2128: case OP_POSPLUSI:
2129: case OP_QUERY:
2130: case OP_QUERYI:
2131: case OP_MINQUERY:
2132: case OP_MINQUERYI:
2133: case OP_POSQUERY:
2134: case OP_POSQUERYI:
2135: if (code[-1] >= 0xc0) code += _pcre_utf8_table4[code[-1] & 0x3f];
2136: break;
2137: }
2138: #else
2139: (void)(utf8); /* Keep compiler happy by referencing function argument */
2140: #endif
2141: }
2142: }
2143: }
2144:
2145:
2146:
2147: /*************************************************
2148: * Scan compiled branch for non-emptiness *
2149: *************************************************/
2150:
2151: /* This function scans through a branch of a compiled pattern to see whether it
2152: can match the empty string or not. It is called from could_be_empty()
2153: below and from compile_branch() when checking for an unlimited repeat of a
2154: group that can match nothing. Note that first_significant_code() skips over
2155: backward and negative forward assertions when its final argument is TRUE. If we
2156: hit an unclosed bracket, we return "empty" - this means we've struck an inner
2157: bracket whose current branch will already have been scanned.
2158:
2159: Arguments:
2160: code points to start of search
2161: endcode points to where to stop
2162: utf8 TRUE if in UTF8 mode
2163: cd contains pointers to tables etc.
2164:
2165: Returns: TRUE if what is matched could be empty
2166: */
2167:
2168: static BOOL
2169: could_be_empty_branch(const uschar *code, const uschar *endcode, BOOL utf8,
2170: compile_data *cd)
2171: {
2172: register int c;
2173: for (code = first_significant_code(code + _pcre_OP_lengths[*code], TRUE);
2174: code < endcode;
2175: code = first_significant_code(code + _pcre_OP_lengths[c], TRUE))
2176: {
2177: const uschar *ccode;
2178:
2179: c = *code;
2180:
2181: /* Skip over forward assertions; the other assertions are skipped by
2182: first_significant_code() with a TRUE final argument. */
2183:
2184: if (c == OP_ASSERT)
2185: {
2186: do code += GET(code, 1); while (*code == OP_ALT);
2187: c = *code;
2188: continue;
2189: }
2190:
2191: /* For a recursion/subroutine call, if its end has been reached, which
2192: implies a backward reference subroutine call, we can scan it. If it's a
2193: forward reference subroutine call, we can't. To detect forward reference
2194: we have to scan up the list that is kept in the workspace. This function is
2195: called only when doing the real compile, not during the pre-compile that
2196: measures the size of the compiled pattern. */
2197:
2198: if (c == OP_RECURSE)
2199: {
2200: const uschar *scode;
2201: BOOL empty_branch;
2202:
2203: /* Test for forward reference */
2204:
2205: for (scode = cd->start_workspace; scode < cd->hwm; scode += LINK_SIZE)
2206: if (GET(scode, 0) == code + 1 - cd->start_code) return TRUE;
2207:
2208: /* Not a forward reference, test for completed backward reference */
2209:
2210: empty_branch = FALSE;
2211: scode = cd->start_code + GET(code, 1);
2212: if (GET(scode, 1) == 0) return TRUE; /* Unclosed */
2213:
2214: /* Completed backwards reference */
2215:
2216: do
2217: {
2218: if (could_be_empty_branch(scode, endcode, utf8, cd))
2219: {
2220: empty_branch = TRUE;
2221: break;
2222: }
2223: scode += GET(scode, 1);
2224: }
2225: while (*scode == OP_ALT);
2226:
2227: if (!empty_branch) return FALSE; /* All branches are non-empty */
2228: continue;
2229: }
2230:
2231: /* Groups with zero repeats can of course be empty; skip them. */
2232:
2233: if (c == OP_BRAZERO || c == OP_BRAMINZERO || c == OP_SKIPZERO ||
2234: c == OP_BRAPOSZERO)
2235: {
2236: code += _pcre_OP_lengths[c];
2237: do code += GET(code, 1); while (*code == OP_ALT);
2238: c = *code;
2239: continue;
2240: }
2241:
2242: /* A nested group that is already marked as "could be empty" can just be
2243: skipped. */
2244:
2245: if (c == OP_SBRA || c == OP_SBRAPOS ||
2246: c == OP_SCBRA || c == OP_SCBRAPOS)
2247: {
2248: do code += GET(code, 1); while (*code == OP_ALT);
2249: c = *code;
2250: continue;
2251: }
2252:
2253: /* For other groups, scan the branches. */
2254:
2255: if (c == OP_BRA || c == OP_BRAPOS ||
2256: c == OP_CBRA || c == OP_CBRAPOS ||
2257: c == OP_ONCE || c == OP_ONCE_NC ||
2258: c == OP_COND)
2259: {
2260: BOOL empty_branch;
2261: if (GET(code, 1) == 0) return TRUE; /* Hit unclosed bracket */
2262:
2263: /* If a conditional group has only one branch, there is a second, implied,
2264: empty branch, so just skip over the conditional, because it could be empty.
2265: Otherwise, scan the individual branches of the group. */
2266:
2267: if (c == OP_COND && code[GET(code, 1)] != OP_ALT)
2268: code += GET(code, 1);
2269: else
2270: {
2271: empty_branch = FALSE;
2272: do
2273: {
2274: if (!empty_branch && could_be_empty_branch(code, endcode, utf8, cd))
2275: empty_branch = TRUE;
2276: code += GET(code, 1);
2277: }
2278: while (*code == OP_ALT);
2279: if (!empty_branch) return FALSE; /* All branches are non-empty */
2280: }
2281:
2282: c = *code;
2283: continue;
2284: }
2285:
2286: /* Handle the other opcodes */
2287:
2288: switch (c)
2289: {
2290: /* Check for quantifiers after a class. XCLASS is used for classes that
2291: cannot be represented just by a bit map. This includes negated single
2292: high-valued characters. The length in _pcre_OP_lengths[] is zero; the
2293: actual length is stored in the compiled code, so we must update "code"
2294: here. */
2295:
2296: #ifdef SUPPORT_UTF8
2297: case OP_XCLASS:
2298: ccode = code += GET(code, 1);
2299: goto CHECK_CLASS_REPEAT;
2300: #endif
2301:
2302: case OP_CLASS:
2303: case OP_NCLASS:
2304: ccode = code + 33;
2305:
2306: #ifdef SUPPORT_UTF8
2307: CHECK_CLASS_REPEAT:
2308: #endif
2309:
2310: switch (*ccode)
2311: {
2312: case OP_CRSTAR: /* These could be empty; continue */
2313: case OP_CRMINSTAR:
2314: case OP_CRQUERY:
2315: case OP_CRMINQUERY:
2316: break;
2317:
2318: default: /* Non-repeat => class must match */
2319: case OP_CRPLUS: /* These repeats aren't empty */
2320: case OP_CRMINPLUS:
2321: return FALSE;
2322:
2323: case OP_CRRANGE:
2324: case OP_CRMINRANGE:
2325: if (GET2(ccode, 1) > 0) return FALSE; /* Minimum > 0 */
2326: break;
2327: }
2328: break;
2329:
2330: /* Opcodes that must match a character */
2331:
2332: case OP_PROP:
2333: case OP_NOTPROP:
2334: case OP_EXTUNI:
2335: case OP_NOT_DIGIT:
2336: case OP_DIGIT:
2337: case OP_NOT_WHITESPACE:
2338: case OP_WHITESPACE:
2339: case OP_NOT_WORDCHAR:
2340: case OP_WORDCHAR:
2341: case OP_ANY:
2342: case OP_ALLANY:
2343: case OP_ANYBYTE:
2344: case OP_CHAR:
2345: case OP_CHARI:
2346: case OP_NOT:
2347: case OP_NOTI:
2348: case OP_PLUS:
2349: case OP_MINPLUS:
2350: case OP_POSPLUS:
2351: case OP_EXACT:
2352: case OP_NOTPLUS:
2353: case OP_NOTMINPLUS:
2354: case OP_NOTPOSPLUS:
2355: case OP_NOTEXACT:
2356: case OP_TYPEPLUS:
2357: case OP_TYPEMINPLUS:
2358: case OP_TYPEPOSPLUS:
2359: case OP_TYPEEXACT:
2360: return FALSE;
2361:
2362: /* These are going to continue, as they may be empty, but we have to
2363: fudge the length for the \p and \P cases. */
2364:
2365: case OP_TYPESTAR:
2366: case OP_TYPEMINSTAR:
2367: case OP_TYPEPOSSTAR:
2368: case OP_TYPEQUERY:
2369: case OP_TYPEMINQUERY:
2370: case OP_TYPEPOSQUERY:
2371: if (code[1] == OP_PROP || code[1] == OP_NOTPROP) code += 2;
2372: break;
2373:
2374: /* Same for these */
2375:
2376: case OP_TYPEUPTO:
2377: case OP_TYPEMINUPTO:
2378: case OP_TYPEPOSUPTO:
2379: if (code[3] == OP_PROP || code[3] == OP_NOTPROP) code += 2;
2380: break;
2381:
2382: /* End of branch */
2383:
2384: case OP_KET:
2385: case OP_KETRMAX:
2386: case OP_KETRMIN:
2387: case OP_KETRPOS:
2388: case OP_ALT:
2389: return TRUE;
2390:
2391: /* In UTF-8 mode, STAR, MINSTAR, POSSTAR, QUERY, MINQUERY, POSQUERY, UPTO,
2392: MINUPTO, and POSUPTO may be followed by a multibyte character */
2393:
2394: #ifdef SUPPORT_UTF8
2395: case OP_STAR:
2396: case OP_STARI:
2397: case OP_MINSTAR:
2398: case OP_MINSTARI:
2399: case OP_POSSTAR:
2400: case OP_POSSTARI:
2401: case OP_QUERY:
2402: case OP_QUERYI:
2403: case OP_MINQUERY:
2404: case OP_MINQUERYI:
2405: case OP_POSQUERY:
2406: case OP_POSQUERYI:
2407: if (utf8 && code[1] >= 0xc0) code += _pcre_utf8_table4[code[1] & 0x3f];
2408: break;
2409:
2410: case OP_UPTO:
2411: case OP_UPTOI:
2412: case OP_MINUPTO:
2413: case OP_MINUPTOI:
2414: case OP_POSUPTO:
2415: case OP_POSUPTOI:
2416: if (utf8 && code[3] >= 0xc0) code += _pcre_utf8_table4[code[3] & 0x3f];
2417: break;
2418: #endif
2419:
2420: /* MARK, and PRUNE/SKIP/THEN with an argument must skip over the argument
2421: string. */
2422:
2423: case OP_MARK:
2424: case OP_PRUNE_ARG:
2425: case OP_SKIP_ARG:
2426: code += code[1];
2427: break;
2428:
2429: case OP_THEN_ARG:
2430: code += code[1];
2431: break;
2432:
2433: /* None of the remaining opcodes are required to match a character. */
2434:
2435: default:
2436: break;
2437: }
2438: }
2439:
2440: return TRUE;
2441: }
2442:
2443:
2444:
2445: /*************************************************
2446: * Scan compiled regex for non-emptiness *
2447: *************************************************/
2448:
2449: /* This function is called to check for left recursive calls. We want to check
2450: the current branch of the current pattern to see if it could match the empty
2451: string. If it could, we must look outwards for branches at other levels,
2452: stopping when we pass beyond the bracket which is the subject of the recursion.
2453: This function is called only during the real compile, not during the
2454: pre-compile.
2455:
2456: Arguments:
2457: code points to start of the recursion
2458: endcode points to where to stop (current RECURSE item)
2459: bcptr points to the chain of current (unclosed) branch starts
2460: utf8 TRUE if in UTF-8 mode
2461: cd pointers to tables etc
2462:
2463: Returns: TRUE if what is matched could be empty
2464: */
2465:
2466: static BOOL
2467: could_be_empty(const uschar *code, const uschar *endcode, branch_chain *bcptr,
2468: BOOL utf8, compile_data *cd)
2469: {
2470: while (bcptr != NULL && bcptr->current_branch >= code)
2471: {
2472: if (!could_be_empty_branch(bcptr->current_branch, endcode, utf8, cd))
2473: return FALSE;
2474: bcptr = bcptr->outer;
2475: }
2476: return TRUE;
2477: }
2478:
2479:
2480:
2481: /*************************************************
2482: * Check for POSIX class syntax *
2483: *************************************************/
2484:
2485: /* This function is called when the sequence "[:" or "[." or "[=" is
2486: encountered in a character class. It checks whether this is followed by a
2487: sequence of characters terminated by a matching ":]" or ".]" or "=]". If we
2488: reach an unescaped ']' without the special preceding character, return FALSE.
2489:
2490: Originally, this function only recognized a sequence of letters between the
2491: terminators, but it seems that Perl recognizes any sequence of characters,
2492: though of course unknown POSIX names are subsequently rejected. Perl gives an
2493: "Unknown POSIX class" error for [:f\oo:] for example, where previously PCRE
2494: didn't consider this to be a POSIX class. Likewise for [:1234:].
2495:
2496: The problem in trying to be exactly like Perl is in the handling of escapes. We
2497: have to be sure that [abc[:x\]pqr] is *not* treated as containing a POSIX
2498: class, but [abc[:x\]pqr:]] is (so that an error can be generated). The code
2499: below handles the special case of \], but does not try to do any other escape
2500: processing. This makes it different from Perl for cases such as [:l\ower:]
2501: where Perl recognizes it as the POSIX class "lower" but PCRE does not recognize
2502: "l\ower". This is a lesser evil that not diagnosing bad classes when Perl does,
2503: I think.
2504:
2505: A user pointed out that PCRE was rejecting [:a[:digit:]] whereas Perl was not.
2506: It seems that the appearance of a nested POSIX class supersedes an apparent
2507: external class. For example, [:a[:digit:]b:] matches "a", "b", ":", or
2508: a digit.
2509:
2510: In Perl, unescaped square brackets may also appear as part of class names. For
2511: example, [:a[:abc]b:] gives unknown POSIX class "[:abc]b:]". However, for
2512: [:a[:abc]b][b:] it gives unknown POSIX class "[:abc]b][b:]", which does not
2513: seem right at all. PCRE does not allow closing square brackets in POSIX class
2514: names.
2515:
2516: Arguments:
2517: ptr pointer to the initial [
2518: endptr where to return the end pointer
2519:
2520: Returns: TRUE or FALSE
2521: */
2522:
2523: static BOOL
2524: check_posix_syntax(const uschar *ptr, const uschar **endptr)
2525: {
2526: int terminator; /* Don't combine these lines; the Solaris cc */
2527: terminator = *(++ptr); /* compiler warns about "non-constant" initializer. */
2528: for (++ptr; *ptr != 0; ptr++)
2529: {
2530: if (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_RIGHT_SQUARE_BRACKET)
2531: ptr++;
2532: else if (*ptr == CHAR_RIGHT_SQUARE_BRACKET) return FALSE;
2533: else
2534: {
2535: if (*ptr == terminator && ptr[1] == CHAR_RIGHT_SQUARE_BRACKET)
2536: {
2537: *endptr = ptr;
2538: return TRUE;
2539: }
2540: if (*ptr == CHAR_LEFT_SQUARE_BRACKET &&
2541: (ptr[1] == CHAR_COLON || ptr[1] == CHAR_DOT ||
2542: ptr[1] == CHAR_EQUALS_SIGN) &&
2543: check_posix_syntax(ptr, endptr))
2544: return FALSE;
2545: }
2546: }
2547: return FALSE;
2548: }
2549:
2550:
2551:
2552:
2553: /*************************************************
2554: * Check POSIX class name *
2555: *************************************************/
2556:
2557: /* This function is called to check the name given in a POSIX-style class entry
2558: such as [:alnum:].
2559:
2560: Arguments:
2561: ptr points to the first letter
2562: len the length of the name
2563:
2564: Returns: a value representing the name, or -1 if unknown
2565: */
2566:
2567: static int
2568: check_posix_name(const uschar *ptr, int len)
2569: {
2570: const char *pn = posix_names;
2571: register int yield = 0;
2572: while (posix_name_lengths[yield] != 0)
2573: {
2574: if (len == posix_name_lengths[yield] &&
2575: strncmp((const char *)ptr, pn, len) == 0) return yield;
2576: pn += posix_name_lengths[yield] + 1;
2577: yield++;
2578: }
2579: return -1;
2580: }
2581:
2582:
2583: /*************************************************
2584: * Adjust OP_RECURSE items in repeated group *
2585: *************************************************/
2586:
2587: /* OP_RECURSE items contain an offset from the start of the regex to the group
2588: that is referenced. This means that groups can be replicated for fixed
2589: repetition simply by copying (because the recursion is allowed to refer to
2590: earlier groups that are outside the current group). However, when a group is
2591: optional (i.e. the minimum quantifier is zero), OP_BRAZERO or OP_SKIPZERO is
2592: inserted before it, after it has been compiled. This means that any OP_RECURSE
2593: items within it that refer to the group itself or any contained groups have to
2594: have their offsets adjusted. That one of the jobs of this function. Before it
2595: is called, the partially compiled regex must be temporarily terminated with
2596: OP_END.
2597:
2598: This function has been extended with the possibility of forward references for
2599: recursions and subroutine calls. It must also check the list of such references
2600: for the group we are dealing with. If it finds that one of the recursions in
2601: the current group is on this list, it adjusts the offset in the list, not the
2602: value in the reference (which is a group number).
2603:
2604: Arguments:
2605: group points to the start of the group
2606: adjust the amount by which the group is to be moved
2607: utf8 TRUE in UTF-8 mode
2608: cd contains pointers to tables etc.
2609: save_hwm the hwm forward reference pointer at the start of the group
2610:
2611: Returns: nothing
2612: */
2613:
2614: static void
2615: adjust_recurse(uschar *group, int adjust, BOOL utf8, compile_data *cd,
2616: uschar *save_hwm)
2617: {
2618: uschar *ptr = group;
2619:
2620: while ((ptr = (uschar *)find_recurse(ptr, utf8)) != NULL)
2621: {
2622: int offset;
2623: uschar *hc;
2624:
2625: /* See if this recursion is on the forward reference list. If so, adjust the
2626: reference. */
2627:
2628: for (hc = save_hwm; hc < cd->hwm; hc += LINK_SIZE)
2629: {
2630: offset = GET(hc, 0);
2631: if (cd->start_code + offset == ptr + 1)
2632: {
2633: PUT(hc, 0, offset + adjust);
2634: break;
2635: }
2636: }
2637:
2638: /* Otherwise, adjust the recursion offset if it's after the start of this
2639: group. */
2640:
2641: if (hc >= cd->hwm)
2642: {
2643: offset = GET(ptr, 1);
2644: if (cd->start_code + offset >= group) PUT(ptr, 1, offset + adjust);
2645: }
2646:
2647: ptr += 1 + LINK_SIZE;
2648: }
2649: }
2650:
2651:
2652:
2653: /*************************************************
2654: * Insert an automatic callout point *
2655: *************************************************/
2656:
2657: /* This function is called when the PCRE_AUTO_CALLOUT option is set, to insert
2658: callout points before each pattern item.
2659:
2660: Arguments:
2661: code current code pointer
2662: ptr current pattern pointer
2663: cd pointers to tables etc
2664:
2665: Returns: new code pointer
2666: */
2667:
2668: static uschar *
2669: auto_callout(uschar *code, const uschar *ptr, compile_data *cd)
2670: {
2671: *code++ = OP_CALLOUT;
2672: *code++ = 255;
2673: PUT(code, 0, (int)(ptr - cd->start_pattern)); /* Pattern offset */
2674: PUT(code, LINK_SIZE, 0); /* Default length */
2675: return code + 2*LINK_SIZE;
2676: }
2677:
2678:
2679:
2680: /*************************************************
2681: * Complete a callout item *
2682: *************************************************/
2683:
2684: /* A callout item contains the length of the next item in the pattern, which
2685: we can't fill in till after we have reached the relevant point. This is used
2686: for both automatic and manual callouts.
2687:
2688: Arguments:
2689: previous_callout points to previous callout item
2690: ptr current pattern pointer
2691: cd pointers to tables etc
2692:
2693: Returns: nothing
2694: */
2695:
2696: static void
2697: complete_callout(uschar *previous_callout, const uschar *ptr, compile_data *cd)
2698: {
2699: int length = (int)(ptr - cd->start_pattern - GET(previous_callout, 2));
2700: PUT(previous_callout, 2 + LINK_SIZE, length);
2701: }
2702:
2703:
2704:
2705: #ifdef SUPPORT_UCP
2706: /*************************************************
2707: * Get othercase range *
2708: *************************************************/
2709:
2710: /* This function is passed the start and end of a class range, in UTF-8 mode
2711: with UCP support. It searches up the characters, looking for internal ranges of
2712: characters in the "other" case. Each call returns the next one, updating the
2713: start address.
2714:
2715: Arguments:
2716: cptr points to starting character value; updated
2717: d end value
2718: ocptr where to put start of othercase range
2719: odptr where to put end of othercase range
2720:
2721: Yield: TRUE when range returned; FALSE when no more
2722: */
2723:
2724: static BOOL
2725: get_othercase_range(unsigned int *cptr, unsigned int d, unsigned int *ocptr,
2726: unsigned int *odptr)
2727: {
2728: unsigned int c, othercase, next;
2729:
2730: for (c = *cptr; c <= d; c++)
2731: { if ((othercase = UCD_OTHERCASE(c)) != c) break; }
2732:
2733: if (c > d) return FALSE;
2734:
2735: *ocptr = othercase;
2736: next = othercase + 1;
2737:
2738: for (++c; c <= d; c++)
2739: {
2740: if (UCD_OTHERCASE(c) != next) break;
2741: next++;
2742: }
2743:
2744: *odptr = next - 1;
2745: *cptr = c;
2746:
2747: return TRUE;
2748: }
2749:
2750:
2751:
2752: /*************************************************
2753: * Check a character and a property *
2754: *************************************************/
2755:
2756: /* This function is called by check_auto_possessive() when a property item
2757: is adjacent to a fixed character.
2758:
2759: Arguments:
2760: c the character
2761: ptype the property type
2762: pdata the data for the type
2763: negated TRUE if it's a negated property (\P or \p{^)
2764:
2765: Returns: TRUE if auto-possessifying is OK
2766: */
2767:
2768: static BOOL
2769: check_char_prop(int c, int ptype, int pdata, BOOL negated)
2770: {
2771: const ucd_record *prop = GET_UCD(c);
2772: switch(ptype)
2773: {
2774: case PT_LAMP:
2775: return (prop->chartype == ucp_Lu ||
2776: prop->chartype == ucp_Ll ||
2777: prop->chartype == ucp_Lt) == negated;
2778:
2779: case PT_GC:
2780: return (pdata == _pcre_ucp_gentype[prop->chartype]) == negated;
2781:
2782: case PT_PC:
2783: return (pdata == prop->chartype) == negated;
2784:
2785: case PT_SC:
2786: return (pdata == prop->script) == negated;
2787:
2788: /* These are specials */
2789:
2790: case PT_ALNUM:
2791: return (_pcre_ucp_gentype[prop->chartype] == ucp_L ||
2792: _pcre_ucp_gentype[prop->chartype] == ucp_N) == negated;
2793:
2794: case PT_SPACE: /* Perl space */
2795: return (_pcre_ucp_gentype[prop->chartype] == ucp_Z ||
2796: c == CHAR_HT || c == CHAR_NL || c == CHAR_FF || c == CHAR_CR)
2797: == negated;
2798:
2799: case PT_PXSPACE: /* POSIX space */
2800: return (_pcre_ucp_gentype[prop->chartype] == ucp_Z ||
2801: c == CHAR_HT || c == CHAR_NL || c == CHAR_VT ||
2802: c == CHAR_FF || c == CHAR_CR)
2803: == negated;
2804:
2805: case PT_WORD:
2806: return (_pcre_ucp_gentype[prop->chartype] == ucp_L ||
2807: _pcre_ucp_gentype[prop->chartype] == ucp_N ||
2808: c == CHAR_UNDERSCORE) == negated;
2809: }
2810: return FALSE;
2811: }
2812: #endif /* SUPPORT_UCP */
2813:
2814:
2815:
2816: /*************************************************
2817: * Check if auto-possessifying is possible *
2818: *************************************************/
2819:
2820: /* This function is called for unlimited repeats of certain items, to see
2821: whether the next thing could possibly match the repeated item. If not, it makes
2822: sense to automatically possessify the repeated item.
2823:
2824: Arguments:
2825: previous pointer to the repeated opcode
2826: utf8 TRUE in UTF-8 mode
2827: ptr next character in pattern
2828: options options bits
2829: cd contains pointers to tables etc.
2830:
2831: Returns: TRUE if possessifying is wanted
2832: */
2833:
2834: static BOOL
2835: check_auto_possessive(const uschar *previous, BOOL utf8, const uschar *ptr,
2836: int options, compile_data *cd)
2837: {
2838: int c, next;
2839: int op_code = *previous++;
2840:
2841: /* Skip whitespace and comments in extended mode */
2842:
2843: if ((options & PCRE_EXTENDED) != 0)
2844: {
2845: for (;;)
2846: {
2847: while ((cd->ctypes[*ptr] & ctype_space) != 0) ptr++;
2848: if (*ptr == CHAR_NUMBER_SIGN)
2849: {
2850: ptr++;
2851: while (*ptr != 0)
2852: {
2853: if (IS_NEWLINE(ptr)) { ptr += cd->nllen; break; }
2854: ptr++;
2855: #ifdef SUPPORT_UTF8
2856: if (utf8) while ((*ptr & 0xc0) == 0x80) ptr++;
2857: #endif
2858: }
2859: }
2860: else break;
2861: }
2862: }
2863:
2864: /* If the next item is one that we can handle, get its value. A non-negative
2865: value is a character, a negative value is an escape value. */
2866:
2867: if (*ptr == CHAR_BACKSLASH)
2868: {
2869: int temperrorcode = 0;
2870: next = check_escape(&ptr, &temperrorcode, cd->bracount, options, FALSE);
2871: if (temperrorcode != 0) return FALSE;
2872: ptr++; /* Point after the escape sequence */
2873: }
2874:
2875: else if ((cd->ctypes[*ptr] & ctype_meta) == 0)
2876: {
2877: #ifdef SUPPORT_UTF8
2878: if (utf8) { GETCHARINC(next, ptr); } else
2879: #endif
2880: next = *ptr++;
2881: }
2882:
2883: else return FALSE;
2884:
2885: /* Skip whitespace and comments in extended mode */
2886:
2887: if ((options & PCRE_EXTENDED) != 0)
2888: {
2889: for (;;)
2890: {
2891: while ((cd->ctypes[*ptr] & ctype_space) != 0) ptr++;
2892: if (*ptr == CHAR_NUMBER_SIGN)
2893: {
2894: ptr++;
2895: while (*ptr != 0)
2896: {
2897: if (IS_NEWLINE(ptr)) { ptr += cd->nllen; break; }
2898: ptr++;
2899: #ifdef SUPPORT_UTF8
2900: if (utf8) while ((*ptr & 0xc0) == 0x80) ptr++;
2901: #endif
2902: }
2903: }
2904: else break;
2905: }
2906: }
2907:
2908: /* If the next thing is itself optional, we have to give up. */
2909:
2910: if (*ptr == CHAR_ASTERISK || *ptr == CHAR_QUESTION_MARK ||
2911: strncmp((char *)ptr, STR_LEFT_CURLY_BRACKET STR_0 STR_COMMA, 3) == 0)
2912: return FALSE;
2913:
2914: /* Now compare the next item with the previous opcode. First, handle cases when
2915: the next item is a character. */
2916:
2917: if (next >= 0) switch(op_code)
2918: {
2919: case OP_CHAR:
2920: #ifdef SUPPORT_UTF8
2921: GETCHARTEST(c, previous);
2922: #else
2923: c = *previous;
2924: #endif
2925: return c != next;
2926:
2927: /* For CHARI (caseless character) we must check the other case. If we have
2928: Unicode property support, we can use it to test the other case of
2929: high-valued characters. */
2930:
2931: case OP_CHARI:
2932: #ifdef SUPPORT_UTF8
2933: GETCHARTEST(c, previous);
2934: #else
2935: c = *previous;
2936: #endif
2937: if (c == next) return FALSE;
2938: #ifdef SUPPORT_UTF8
2939: if (utf8)
2940: {
2941: unsigned int othercase;
2942: if (next < 128) othercase = cd->fcc[next]; else
2943: #ifdef SUPPORT_UCP
2944: othercase = UCD_OTHERCASE((unsigned int)next);
2945: #else
2946: othercase = NOTACHAR;
2947: #endif
2948: return (unsigned int)c != othercase;
2949: }
2950: else
2951: #endif /* SUPPORT_UTF8 */
2952: return (c != cd->fcc[next]); /* Non-UTF-8 mode */
2953:
2954: /* For OP_NOT and OP_NOTI, the data is always a single-byte character. These
2955: opcodes are not used for multi-byte characters, because they are coded using
2956: an XCLASS instead. */
2957:
2958: case OP_NOT:
2959: return (c = *previous) == next;
2960:
2961: case OP_NOTI:
2962: if ((c = *previous) == next) return TRUE;
2963: #ifdef SUPPORT_UTF8
2964: if (utf8)
2965: {
2966: unsigned int othercase;
2967: if (next < 128) othercase = cd->fcc[next]; else
2968: #ifdef SUPPORT_UCP
2969: othercase = UCD_OTHERCASE(next);
2970: #else
2971: othercase = NOTACHAR;
2972: #endif
2973: return (unsigned int)c == othercase;
2974: }
2975: else
2976: #endif /* SUPPORT_UTF8 */
2977: return (c == cd->fcc[next]); /* Non-UTF-8 mode */
2978:
2979: /* Note that OP_DIGIT etc. are generated only when PCRE_UCP is *not* set.
2980: When it is set, \d etc. are converted into OP_(NOT_)PROP codes. */
2981:
2982: case OP_DIGIT:
2983: return next > 127 || (cd->ctypes[next] & ctype_digit) == 0;
2984:
2985: case OP_NOT_DIGIT:
2986: return next <= 127 && (cd->ctypes[next] & ctype_digit) != 0;
2987:
2988: case OP_WHITESPACE:
2989: return next > 127 || (cd->ctypes[next] & ctype_space) == 0;
2990:
2991: case OP_NOT_WHITESPACE:
2992: return next <= 127 && (cd->ctypes[next] & ctype_space) != 0;
2993:
2994: case OP_WORDCHAR:
2995: return next > 127 || (cd->ctypes[next] & ctype_word) == 0;
2996:
2997: case OP_NOT_WORDCHAR:
2998: return next <= 127 && (cd->ctypes[next] & ctype_word) != 0;
2999:
3000: case OP_HSPACE:
3001: case OP_NOT_HSPACE:
3002: switch(next)
3003: {
3004: case 0x09:
3005: case 0x20:
3006: case 0xa0:
3007: case 0x1680:
3008: case 0x180e:
3009: case 0x2000:
3010: case 0x2001:
3011: case 0x2002:
3012: case 0x2003:
3013: case 0x2004:
3014: case 0x2005:
3015: case 0x2006:
3016: case 0x2007:
3017: case 0x2008:
3018: case 0x2009:
3019: case 0x200A:
3020: case 0x202f:
3021: case 0x205f:
3022: case 0x3000:
3023: return op_code == OP_NOT_HSPACE;
3024: default:
3025: return op_code != OP_NOT_HSPACE;
3026: }
3027:
3028: case OP_ANYNL:
3029: case OP_VSPACE:
3030: case OP_NOT_VSPACE:
3031: switch(next)
3032: {
3033: case 0x0a:
3034: case 0x0b:
3035: case 0x0c:
3036: case 0x0d:
3037: case 0x85:
3038: case 0x2028:
3039: case 0x2029:
3040: return op_code == OP_NOT_VSPACE;
3041: default:
3042: return op_code != OP_NOT_VSPACE;
3043: }
3044:
3045: #ifdef SUPPORT_UCP
3046: case OP_PROP:
3047: return check_char_prop(next, previous[0], previous[1], FALSE);
3048:
3049: case OP_NOTPROP:
3050: return check_char_prop(next, previous[0], previous[1], TRUE);
3051: #endif
3052:
3053: default:
3054: return FALSE;
3055: }
3056:
3057:
3058: /* Handle the case when the next item is \d, \s, etc. Note that when PCRE_UCP
3059: is set, \d turns into ESC_du rather than ESC_d, etc., so ESC_d etc. are
3060: generated only when PCRE_UCP is *not* set, that is, when only ASCII
3061: characteristics are recognized. Similarly, the opcodes OP_DIGIT etc. are
3062: replaced by OP_PROP codes when PCRE_UCP is set. */
3063:
3064: switch(op_code)
3065: {
3066: case OP_CHAR:
3067: case OP_CHARI:
3068: #ifdef SUPPORT_UTF8
3069: GETCHARTEST(c, previous);
3070: #else
3071: c = *previous;
3072: #endif
3073: switch(-next)
3074: {
3075: case ESC_d:
3076: return c > 127 || (cd->ctypes[c] & ctype_digit) == 0;
3077:
3078: case ESC_D:
3079: return c <= 127 && (cd->ctypes[c] & ctype_digit) != 0;
3080:
3081: case ESC_s:
3082: return c > 127 || (cd->ctypes[c] & ctype_space) == 0;
3083:
3084: case ESC_S:
3085: return c <= 127 && (cd->ctypes[c] & ctype_space) != 0;
3086:
3087: case ESC_w:
3088: return c > 127 || (cd->ctypes[c] & ctype_word) == 0;
3089:
3090: case ESC_W:
3091: return c <= 127 && (cd->ctypes[c] & ctype_word) != 0;
3092:
3093: case ESC_h:
3094: case ESC_H:
3095: switch(c)
3096: {
3097: case 0x09:
3098: case 0x20:
3099: case 0xa0:
3100: case 0x1680:
3101: case 0x180e:
3102: case 0x2000:
3103: case 0x2001:
3104: case 0x2002:
3105: case 0x2003:
3106: case 0x2004:
3107: case 0x2005:
3108: case 0x2006:
3109: case 0x2007:
3110: case 0x2008:
3111: case 0x2009:
3112: case 0x200A:
3113: case 0x202f:
3114: case 0x205f:
3115: case 0x3000:
3116: return -next != ESC_h;
3117: default:
3118: return -next == ESC_h;
3119: }
3120:
3121: case ESC_v:
3122: case ESC_V:
3123: switch(c)
3124: {
3125: case 0x0a:
3126: case 0x0b:
3127: case 0x0c:
3128: case 0x0d:
3129: case 0x85:
3130: case 0x2028:
3131: case 0x2029:
3132: return -next != ESC_v;
3133: default:
3134: return -next == ESC_v;
3135: }
3136:
3137: /* When PCRE_UCP is set, these values get generated for \d etc. Find
3138: their substitutions and process them. The result will always be either
3139: -ESC_p or -ESC_P. Then fall through to process those values. */
3140:
3141: #ifdef SUPPORT_UCP
3142: case ESC_du:
3143: case ESC_DU:
3144: case ESC_wu:
3145: case ESC_WU:
3146: case ESC_su:
3147: case ESC_SU:
3148: {
3149: int temperrorcode = 0;
3150: ptr = substitutes[-next - ESC_DU];
3151: next = check_escape(&ptr, &temperrorcode, 0, options, FALSE);
3152: if (temperrorcode != 0) return FALSE;
3153: ptr++; /* For compatibility */
3154: }
3155: /* Fall through */
3156:
3157: case ESC_p:
3158: case ESC_P:
3159: {
3160: int ptype, pdata, errorcodeptr;
3161: BOOL negated;
3162:
3163: ptr--; /* Make ptr point at the p or P */
3164: ptype = get_ucp(&ptr, &negated, &pdata, &errorcodeptr);
3165: if (ptype < 0) return FALSE;
3166: ptr++; /* Point past the final curly ket */
3167:
3168: /* If the property item is optional, we have to give up. (When generated
3169: from \d etc by PCRE_UCP, this test will have been applied much earlier,
3170: to the original \d etc. At this point, ptr will point to a zero byte. */
3171:
3172: if (*ptr == CHAR_ASTERISK || *ptr == CHAR_QUESTION_MARK ||
3173: strncmp((char *)ptr, STR_LEFT_CURLY_BRACKET STR_0 STR_COMMA, 3) == 0)
3174: return FALSE;
3175:
3176: /* Do the property check. */
3177:
3178: return check_char_prop(c, ptype, pdata, (next == -ESC_P) != negated);
3179: }
3180: #endif
3181:
3182: default:
3183: return FALSE;
3184: }
3185:
3186: /* In principle, support for Unicode properties should be integrated here as
3187: well. It means re-organizing the above code so as to get hold of the property
3188: values before switching on the op-code. However, I wonder how many patterns
3189: combine ASCII \d etc with Unicode properties? (Note that if PCRE_UCP is set,
3190: these op-codes are never generated.) */
3191:
3192: case OP_DIGIT:
3193: return next == -ESC_D || next == -ESC_s || next == -ESC_W ||
3194: next == -ESC_h || next == -ESC_v || next == -ESC_R;
3195:
3196: case OP_NOT_DIGIT:
3197: return next == -ESC_d;
3198:
3199: case OP_WHITESPACE:
3200: return next == -ESC_S || next == -ESC_d || next == -ESC_w || next == -ESC_R;
3201:
3202: case OP_NOT_WHITESPACE:
3203: return next == -ESC_s || next == -ESC_h || next == -ESC_v;
3204:
3205: case OP_HSPACE:
3206: return next == -ESC_S || next == -ESC_H || next == -ESC_d ||
3207: next == -ESC_w || next == -ESC_v || next == -ESC_R;
3208:
3209: case OP_NOT_HSPACE:
3210: return next == -ESC_h;
3211:
3212: /* Can't have \S in here because VT matches \S (Perl anomaly) */
3213: case OP_ANYNL:
3214: case OP_VSPACE:
3215: return next == -ESC_V || next == -ESC_d || next == -ESC_w;
3216:
3217: case OP_NOT_VSPACE:
3218: return next == -ESC_v || next == -ESC_R;
3219:
3220: case OP_WORDCHAR:
3221: return next == -ESC_W || next == -ESC_s || next == -ESC_h ||
3222: next == -ESC_v || next == -ESC_R;
3223:
3224: case OP_NOT_WORDCHAR:
3225: return next == -ESC_w || next == -ESC_d;
3226:
3227: default:
3228: return FALSE;
3229: }
3230:
3231: /* Control does not reach here */
3232: }
3233:
3234:
3235:
3236: /*************************************************
3237: * Compile one branch *
3238: *************************************************/
3239:
3240: /* Scan the pattern, compiling it into the a vector. If the options are
3241: changed during the branch, the pointer is used to change the external options
3242: bits. This function is used during the pre-compile phase when we are trying
3243: to find out the amount of memory needed, as well as during the real compile
3244: phase. The value of lengthptr distinguishes the two phases.
3245:
3246: Arguments:
3247: optionsptr pointer to the option bits
3248: codeptr points to the pointer to the current code point
3249: ptrptr points to the current pattern pointer
3250: errorcodeptr points to error code variable
3251: firstbyteptr set to initial literal character, or < 0 (REQ_UNSET, REQ_NONE)
3252: reqbyteptr set to the last literal character required, else < 0
3253: bcptr points to current branch chain
3254: cond_depth conditional nesting depth
3255: cd contains pointers to tables etc.
3256: lengthptr NULL during the real compile phase
3257: points to length accumulator during pre-compile phase
3258:
3259: Returns: TRUE on success
3260: FALSE, with *errorcodeptr set non-zero on error
3261: */
3262:
3263: static BOOL
3264: compile_branch(int *optionsptr, uschar **codeptr, const uschar **ptrptr,
3265: int *errorcodeptr, int *firstbyteptr, int *reqbyteptr, branch_chain *bcptr,
3266: int cond_depth, compile_data *cd, int *lengthptr)
3267: {
3268: int repeat_type, op_type;
3269: int repeat_min = 0, repeat_max = 0; /* To please picky compilers */
3270: int bravalue = 0;
3271: int greedy_default, greedy_non_default;
3272: int firstbyte, reqbyte;
3273: int zeroreqbyte, zerofirstbyte;
3274: int req_caseopt, reqvary, tempreqvary;
3275: int options = *optionsptr; /* May change dynamically */
3276: int after_manual_callout = 0;
3277: int length_prevgroup = 0;
3278: register int c;
3279: register uschar *code = *codeptr;
3280: uschar *last_code = code;
3281: uschar *orig_code = code;
3282: uschar *tempcode;
3283: BOOL inescq = FALSE;
3284: BOOL groupsetfirstbyte = FALSE;
3285: const uschar *ptr = *ptrptr;
3286: const uschar *tempptr;
3287: const uschar *nestptr = NULL;
3288: uschar *previous = NULL;
3289: uschar *previous_callout = NULL;
3290: uschar *save_hwm = NULL;
3291: uschar classbits[32];
3292:
3293: /* We can fish out the UTF-8 setting once and for all into a BOOL, but we
3294: must not do this for other options (e.g. PCRE_EXTENDED) because they may change
3295: dynamically as we process the pattern. */
3296:
3297: #ifdef SUPPORT_UTF8
3298: BOOL class_utf8;
3299: BOOL utf8 = (options & PCRE_UTF8) != 0;
3300: uschar *class_utf8data;
3301: uschar *class_utf8data_base;
3302: uschar utf8_char[6];
3303: #else
3304: BOOL utf8 = FALSE;
3305: #endif
3306:
3307: #ifdef PCRE_DEBUG
3308: if (lengthptr != NULL) DPRINTF((">> start branch\n"));
3309: #endif
3310:
3311: /* Set up the default and non-default settings for greediness */
3312:
3313: greedy_default = ((options & PCRE_UNGREEDY) != 0);
3314: greedy_non_default = greedy_default ^ 1;
3315:
3316: /* Initialize no first byte, no required byte. REQ_UNSET means "no char
3317: matching encountered yet". It gets changed to REQ_NONE if we hit something that
3318: matches a non-fixed char first char; reqbyte just remains unset if we never
3319: find one.
3320:
3321: When we hit a repeat whose minimum is zero, we may have to adjust these values
3322: to take the zero repeat into account. This is implemented by setting them to
3323: zerofirstbyte and zeroreqbyte when such a repeat is encountered. The individual
3324: item types that can be repeated set these backoff variables appropriately. */
3325:
3326: firstbyte = reqbyte = zerofirstbyte = zeroreqbyte = REQ_UNSET;
3327:
3328: /* The variable req_caseopt contains either the REQ_CASELESS value or zero,
3329: according to the current setting of the caseless flag. REQ_CASELESS is a bit
3330: value > 255. It is added into the firstbyte or reqbyte variables to record the
3331: case status of the value. This is used only for ASCII characters. */
3332:
3333: req_caseopt = ((options & PCRE_CASELESS) != 0)? REQ_CASELESS : 0;
3334:
3335: /* Switch on next character until the end of the branch */
3336:
3337: for (;; ptr++)
3338: {
3339: BOOL negate_class;
3340: BOOL should_flip_negation;
3341: BOOL possessive_quantifier;
3342: BOOL is_quantifier;
3343: BOOL is_recurse;
3344: BOOL reset_bracount;
3345: int class_charcount;
3346: int class_lastchar;
3347: int newoptions;
3348: int recno;
3349: int refsign;
3350: int skipbytes;
3351: int subreqbyte;
3352: int subfirstbyte;
3353: int terminator;
3354: int mclength;
3355: int tempbracount;
3356: uschar mcbuffer[8];
3357:
3358: /* Get next byte in the pattern */
3359:
3360: c = *ptr;
3361:
3362: /* If we are at the end of a nested substitution, revert to the outer level
3363: string. Nesting only happens one level deep. */
3364:
3365: if (c == 0 && nestptr != NULL)
3366: {
3367: ptr = nestptr;
3368: nestptr = NULL;
3369: c = *ptr;
3370: }
3371:
3372: /* If we are in the pre-compile phase, accumulate the length used for the
3373: previous cycle of this loop. */
3374:
3375: if (lengthptr != NULL)
3376: {
3377: #ifdef PCRE_DEBUG
3378: if (code > cd->hwm) cd->hwm = code; /* High water info */
3379: #endif
3380: if (code > cd->start_workspace + cd->workspace_size -
3381: WORK_SIZE_SAFETY_MARGIN) /* Check for overrun */
3382: {
3383: *errorcodeptr = ERR52;
3384: goto FAILED;
3385: }
3386:
3387: /* There is at least one situation where code goes backwards: this is the
3388: case of a zero quantifier after a class (e.g. [ab]{0}). At compile time,
3389: the class is simply eliminated. However, it is created first, so we have to
3390: allow memory for it. Therefore, don't ever reduce the length at this point.
3391: */
3392:
3393: if (code < last_code) code = last_code;
3394:
3395: /* Paranoid check for integer overflow */
3396:
3397: if (OFLOW_MAX - *lengthptr < code - last_code)
3398: {
3399: *errorcodeptr = ERR20;
3400: goto FAILED;
3401: }
3402:
3403: *lengthptr += (int)(code - last_code);
3404: DPRINTF(("length=%d added %d c=%c\n", *lengthptr, (int)(code - last_code),
3405: c));
3406:
3407: /* If "previous" is set and it is not at the start of the work space, move
3408: it back to there, in order to avoid filling up the work space. Otherwise,
3409: if "previous" is NULL, reset the current code pointer to the start. */
3410:
3411: if (previous != NULL)
3412: {
3413: if (previous > orig_code)
3414: {
3415: memmove(orig_code, previous, code - previous);
3416: code -= previous - orig_code;
3417: previous = orig_code;
3418: }
3419: }
3420: else code = orig_code;
3421:
3422: /* Remember where this code item starts so we can pick up the length
3423: next time round. */
3424:
3425: last_code = code;
3426: }
3427:
3428: /* In the real compile phase, just check the workspace used by the forward
3429: reference list. */
3430:
3431: else if (cd->hwm > cd->start_workspace + cd->workspace_size -
3432: WORK_SIZE_SAFETY_MARGIN)
3433: {
3434: *errorcodeptr = ERR52;
3435: goto FAILED;
3436: }
3437:
3438: /* If in \Q...\E, check for the end; if not, we have a literal */
3439:
3440: if (inescq && c != 0)
3441: {
3442: if (c == CHAR_BACKSLASH && ptr[1] == CHAR_E)
3443: {
3444: inescq = FALSE;
3445: ptr++;
3446: continue;
3447: }
3448: else
3449: {
3450: if (previous_callout != NULL)
3451: {
3452: if (lengthptr == NULL) /* Don't attempt in pre-compile phase */
3453: complete_callout(previous_callout, ptr, cd);
3454: previous_callout = NULL;
3455: }
3456: if ((options & PCRE_AUTO_CALLOUT) != 0)
3457: {
3458: previous_callout = code;
3459: code = auto_callout(code, ptr, cd);
3460: }
3461: goto NORMAL_CHAR;
3462: }
3463: }
3464:
3465: /* Fill in length of a previous callout, except when the next thing is
3466: a quantifier. */
3467:
3468: is_quantifier =
3469: c == CHAR_ASTERISK || c == CHAR_PLUS || c == CHAR_QUESTION_MARK ||
3470: (c == CHAR_LEFT_CURLY_BRACKET && is_counted_repeat(ptr+1));
3471:
3472: if (!is_quantifier && previous_callout != NULL &&
3473: after_manual_callout-- <= 0)
3474: {
3475: if (lengthptr == NULL) /* Don't attempt in pre-compile phase */
3476: complete_callout(previous_callout, ptr, cd);
3477: previous_callout = NULL;
3478: }
3479:
3480: /* In extended mode, skip white space and comments. */
3481:
3482: if ((options & PCRE_EXTENDED) != 0)
3483: {
3484: if ((cd->ctypes[c] & ctype_space) != 0) continue;
3485: if (c == CHAR_NUMBER_SIGN)
3486: {
3487: ptr++;
3488: while (*ptr != 0)
3489: {
3490: if (IS_NEWLINE(ptr)) { ptr += cd->nllen - 1; break; }
3491: ptr++;
3492: #ifdef SUPPORT_UTF8
3493: if (utf8) while ((*ptr & 0xc0) == 0x80) ptr++;
3494: #endif
3495: }
3496: if (*ptr != 0) continue;
3497:
3498: /* Else fall through to handle end of string */
3499: c = 0;
3500: }
3501: }
3502:
3503: /* No auto callout for quantifiers. */
3504:
3505: if ((options & PCRE_AUTO_CALLOUT) != 0 && !is_quantifier)
3506: {
3507: previous_callout = code;
3508: code = auto_callout(code, ptr, cd);
3509: }
3510:
3511: switch(c)
3512: {
3513: /* ===================================================================*/
3514: case 0: /* The branch terminates at string end */
3515: case CHAR_VERTICAL_LINE: /* or | or ) */
3516: case CHAR_RIGHT_PARENTHESIS:
3517: *firstbyteptr = firstbyte;
3518: *reqbyteptr = reqbyte;
3519: *codeptr = code;
3520: *ptrptr = ptr;
3521: if (lengthptr != NULL)
3522: {
3523: if (OFLOW_MAX - *lengthptr < code - last_code)
3524: {
3525: *errorcodeptr = ERR20;
3526: goto FAILED;
3527: }
3528: *lengthptr += (int)(code - last_code); /* To include callout length */
3529: DPRINTF((">> end branch\n"));
3530: }
3531: return TRUE;
3532:
3533:
3534: /* ===================================================================*/
3535: /* Handle single-character metacharacters. In multiline mode, ^ disables
3536: the setting of any following char as a first character. */
3537:
3538: case CHAR_CIRCUMFLEX_ACCENT:
3539: previous = NULL;
3540: if ((options & PCRE_MULTILINE) != 0)
3541: {
3542: if (firstbyte == REQ_UNSET) firstbyte = REQ_NONE;
3543: *code++ = OP_CIRCM;
3544: }
3545: else *code++ = OP_CIRC;
3546: break;
3547:
3548: case CHAR_DOLLAR_SIGN:
3549: previous = NULL;
3550: *code++ = ((options & PCRE_MULTILINE) != 0)? OP_DOLLM : OP_DOLL;
3551: break;
3552:
3553: /* There can never be a first char if '.' is first, whatever happens about
3554: repeats. The value of reqbyte doesn't change either. */
3555:
3556: case CHAR_DOT:
3557: if (firstbyte == REQ_UNSET) firstbyte = REQ_NONE;
3558: zerofirstbyte = firstbyte;
3559: zeroreqbyte = reqbyte;
3560: previous = code;
3561: *code++ = ((options & PCRE_DOTALL) != 0)? OP_ALLANY: OP_ANY;
3562: break;
3563:
3564:
3565: /* ===================================================================*/
3566: /* Character classes. If the included characters are all < 256, we build a
3567: 32-byte bitmap of the permitted characters, except in the special case
3568: where there is only one such character. For negated classes, we build the
3569: map as usual, then invert it at the end. However, we use a different opcode
3570: so that data characters > 255 can be handled correctly.
3571:
3572: If the class contains characters outside the 0-255 range, a different
3573: opcode is compiled. It may optionally have a bit map for characters < 256,
3574: but those above are are explicitly listed afterwards. A flag byte tells
3575: whether the bitmap is present, and whether this is a negated class or not.
3576:
3577: In JavaScript compatibility mode, an isolated ']' causes an error. In
3578: default (Perl) mode, it is treated as a data character. */
3579:
3580: case CHAR_RIGHT_SQUARE_BRACKET:
3581: if ((cd->external_options & PCRE_JAVASCRIPT_COMPAT) != 0)
3582: {
3583: *errorcodeptr = ERR64;
3584: goto FAILED;
3585: }
3586: goto NORMAL_CHAR;
3587:
3588: case CHAR_LEFT_SQUARE_BRACKET:
3589: previous = code;
3590:
3591: /* PCRE supports POSIX class stuff inside a class. Perl gives an error if
3592: they are encountered at the top level, so we'll do that too. */
3593:
3594: if ((ptr[1] == CHAR_COLON || ptr[1] == CHAR_DOT ||
3595: ptr[1] == CHAR_EQUALS_SIGN) &&
3596: check_posix_syntax(ptr, &tempptr))
3597: {
3598: *errorcodeptr = (ptr[1] == CHAR_COLON)? ERR13 : ERR31;
3599: goto FAILED;
3600: }
3601:
3602: /* If the first character is '^', set the negation flag and skip it. Also,
3603: if the first few characters (either before or after ^) are \Q\E or \E we
3604: skip them too. This makes for compatibility with Perl. */
3605:
3606: negate_class = FALSE;
3607: for (;;)
3608: {
3609: c = *(++ptr);
3610: if (c == CHAR_BACKSLASH)
3611: {
3612: if (ptr[1] == CHAR_E)
3613: ptr++;
3614: else if (strncmp((const char *)ptr+1,
3615: STR_Q STR_BACKSLASH STR_E, 3) == 0)
3616: ptr += 3;
3617: else
3618: break;
3619: }
3620: else if (!negate_class && c == CHAR_CIRCUMFLEX_ACCENT)
3621: negate_class = TRUE;
3622: else break;
3623: }
3624:
3625: /* Empty classes are allowed in JavaScript compatibility mode. Otherwise,
3626: an initial ']' is taken as a data character -- the code below handles
3627: that. In JS mode, [] must always fail, so generate OP_FAIL, whereas
3628: [^] must match any character, so generate OP_ALLANY. */
3629:
3630: if (c == CHAR_RIGHT_SQUARE_BRACKET &&
3631: (cd->external_options & PCRE_JAVASCRIPT_COMPAT) != 0)
3632: {
3633: *code++ = negate_class? OP_ALLANY : OP_FAIL;
3634: if (firstbyte == REQ_UNSET) firstbyte = REQ_NONE;
3635: zerofirstbyte = firstbyte;
3636: break;
3637: }
3638:
3639: /* If a class contains a negative special such as \S, we need to flip the
3640: negation flag at the end, so that support for characters > 255 works
3641: correctly (they are all included in the class). */
3642:
3643: should_flip_negation = FALSE;
3644:
3645: /* Keep a count of chars with values < 256 so that we can optimize the case
3646: of just a single character (as long as it's < 256). However, For higher
3647: valued UTF-8 characters, we don't yet do any optimization. */
3648:
3649: class_charcount = 0;
3650: class_lastchar = -1;
3651:
3652: /* Initialize the 32-char bit map to all zeros. We build the map in a
3653: temporary bit of memory, in case the class contains only 1 character (less
3654: than 256), because in that case the compiled code doesn't use the bit map.
3655: */
3656:
3657: memset(classbits, 0, 32 * sizeof(uschar));
3658:
3659: #ifdef SUPPORT_UTF8
3660: class_utf8 = FALSE; /* No chars >= 256 */
3661: class_utf8data = code + LINK_SIZE + 2; /* For UTF-8 items */
3662: class_utf8data_base = class_utf8data; /* For resetting in pass 1 */
3663: #endif
3664:
3665: /* Process characters until ] is reached. By writing this as a "do" it
3666: means that an initial ] is taken as a data character. At the start of the
3667: loop, c contains the first byte of the character. */
3668:
3669: if (c != 0) do
3670: {
3671: const uschar *oldptr;
3672:
3673: #ifdef SUPPORT_UTF8
3674: if (utf8 && c > 127)
3675: { /* Braces are required because the */
3676: GETCHARLEN(c, ptr, ptr); /* macro generates multiple statements */
3677: }
3678:
3679: /* In the pre-compile phase, accumulate the length of any UTF-8 extra
3680: data and reset the pointer. This is so that very large classes that
3681: contain a zillion UTF-8 characters no longer overwrite the work space
3682: (which is on the stack). */
3683:
3684: if (lengthptr != NULL)
3685: {
3686: *lengthptr += (int)(class_utf8data - class_utf8data_base);
3687: class_utf8data = class_utf8data_base;
3688: }
3689:
3690: #endif
3691:
3692: /* Inside \Q...\E everything is literal except \E */
3693:
3694: if (inescq)
3695: {
3696: if (c == CHAR_BACKSLASH && ptr[1] == CHAR_E) /* If we are at \E */
3697: {
3698: inescq = FALSE; /* Reset literal state */
3699: ptr++; /* Skip the 'E' */
3700: continue; /* Carry on with next */
3701: }
3702: goto CHECK_RANGE; /* Could be range if \E follows */
3703: }
3704:
3705: /* Handle POSIX class names. Perl allows a negation extension of the
3706: form [:^name:]. A square bracket that doesn't match the syntax is
3707: treated as a literal. We also recognize the POSIX constructions
3708: [.ch.] and [=ch=] ("collating elements") and fault them, as Perl
3709: 5.6 and 5.8 do. */
3710:
3711: if (c == CHAR_LEFT_SQUARE_BRACKET &&
3712: (ptr[1] == CHAR_COLON || ptr[1] == CHAR_DOT ||
3713: ptr[1] == CHAR_EQUALS_SIGN) && check_posix_syntax(ptr, &tempptr))
3714: {
3715: BOOL local_negate = FALSE;
3716: int posix_class, taboffset, tabopt;
3717: register const uschar *cbits = cd->cbits;
3718: uschar pbits[32];
3719:
3720: if (ptr[1] != CHAR_COLON)
3721: {
3722: *errorcodeptr = ERR31;
3723: goto FAILED;
3724: }
3725:
3726: ptr += 2;
3727: if (*ptr == CHAR_CIRCUMFLEX_ACCENT)
3728: {
3729: local_negate = TRUE;
3730: should_flip_negation = TRUE; /* Note negative special */
3731: ptr++;
3732: }
3733:
3734: posix_class = check_posix_name(ptr, (int)(tempptr - ptr));
3735: if (posix_class < 0)
3736: {
3737: *errorcodeptr = ERR30;
3738: goto FAILED;
3739: }
3740:
3741: /* If matching is caseless, upper and lower are converted to
3742: alpha. This relies on the fact that the class table starts with
3743: alpha, lower, upper as the first 3 entries. */
3744:
3745: if ((options & PCRE_CASELESS) != 0 && posix_class <= 2)
3746: posix_class = 0;
3747:
3748: /* When PCRE_UCP is set, some of the POSIX classes are converted to
3749: different escape sequences that use Unicode properties. */
3750:
3751: #ifdef SUPPORT_UCP
3752: if ((options & PCRE_UCP) != 0)
3753: {
3754: int pc = posix_class + ((local_negate)? POSIX_SUBSIZE/2 : 0);
3755: if (posix_substitutes[pc] != NULL)
3756: {
3757: nestptr = tempptr + 1;
3758: ptr = posix_substitutes[pc] - 1;
3759: continue;
3760: }
3761: }
3762: #endif
3763: /* In the non-UCP case, we build the bit map for the POSIX class in a
3764: chunk of local store because we may be adding and subtracting from it,
3765: and we don't want to subtract bits that may be in the main map already.
3766: At the end we or the result into the bit map that is being built. */
3767:
3768: posix_class *= 3;
3769:
3770: /* Copy in the first table (always present) */
3771:
3772: memcpy(pbits, cbits + posix_class_maps[posix_class],
3773: 32 * sizeof(uschar));
3774:
3775: /* If there is a second table, add or remove it as required. */
3776:
3777: taboffset = posix_class_maps[posix_class + 1];
3778: tabopt = posix_class_maps[posix_class + 2];
3779:
3780: if (taboffset >= 0)
3781: {
3782: if (tabopt >= 0)
3783: for (c = 0; c < 32; c++) pbits[c] |= cbits[c + taboffset];
3784: else
3785: for (c = 0; c < 32; c++) pbits[c] &= ~cbits[c + taboffset];
3786: }
3787:
3788: /* Not see if we need to remove any special characters. An option
3789: value of 1 removes vertical space and 2 removes underscore. */
3790:
3791: if (tabopt < 0) tabopt = -tabopt;
3792: if (tabopt == 1) pbits[1] &= ~0x3c;
3793: else if (tabopt == 2) pbits[11] &= 0x7f;
3794:
3795: /* Add the POSIX table or its complement into the main table that is
3796: being built and we are done. */
3797:
3798: if (local_negate)
3799: for (c = 0; c < 32; c++) classbits[c] |= ~pbits[c];
3800: else
3801: for (c = 0; c < 32; c++) classbits[c] |= pbits[c];
3802:
3803: ptr = tempptr + 1;
3804: class_charcount = 10; /* Set > 1; assumes more than 1 per class */
3805: continue; /* End of POSIX syntax handling */
3806: }
3807:
3808: /* Backslash may introduce a single character, or it may introduce one
3809: of the specials, which just set a flag. The sequence \b is a special
3810: case. Inside a class (and only there) it is treated as backspace. We
3811: assume that other escapes have more than one character in them, so set
3812: class_charcount bigger than one. Unrecognized escapes fall through and
3813: are either treated as literal characters (by default), or are faulted if
3814: PCRE_EXTRA is set. */
3815:
3816: if (c == CHAR_BACKSLASH)
3817: {
3818: c = check_escape(&ptr, errorcodeptr, cd->bracount, options, TRUE);
3819: if (*errorcodeptr != 0) goto FAILED;
3820:
3821: if (-c == ESC_b) c = CHAR_BS; /* \b is backspace in a class */
3822: else if (-c == ESC_N) /* \N is not supported in a class */
3823: {
3824: *errorcodeptr = ERR71;
3825: goto FAILED;
3826: }
3827: else if (-c == ESC_Q) /* Handle start of quoted string */
3828: {
3829: if (ptr[1] == CHAR_BACKSLASH && ptr[2] == CHAR_E)
3830: {
3831: ptr += 2; /* avoid empty string */
3832: }
3833: else inescq = TRUE;
3834: continue;
3835: }
3836: else if (-c == ESC_E) continue; /* Ignore orphan \E */
3837:
3838: if (c < 0)
3839: {
3840: register const uschar *cbits = cd->cbits;
3841: class_charcount += 2; /* Greater than 1 is what matters */
3842:
3843: switch (-c)
3844: {
3845: #ifdef SUPPORT_UCP
3846: case ESC_du: /* These are the values given for \d etc */
3847: case ESC_DU: /* when PCRE_UCP is set. We replace the */
3848: case ESC_wu: /* escape sequence with an appropriate \p */
3849: case ESC_WU: /* or \P to test Unicode properties instead */
3850: case ESC_su: /* of the default ASCII testing. */
3851: case ESC_SU:
3852: nestptr = ptr;
3853: ptr = substitutes[-c - ESC_DU] - 1; /* Just before substitute */
3854: class_charcount -= 2; /* Undo! */
3855: continue;
3856: #endif
3857: case ESC_d:
3858: for (c = 0; c < 32; c++) classbits[c] |= cbits[c+cbit_digit];
3859: continue;
3860:
3861: case ESC_D:
3862: should_flip_negation = TRUE;
3863: for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_digit];
3864: continue;
3865:
3866: case ESC_w:
3867: for (c = 0; c < 32; c++) classbits[c] |= cbits[c+cbit_word];
3868: continue;
3869:
3870: case ESC_W:
3871: should_flip_negation = TRUE;
3872: for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_word];
3873: continue;
3874:
3875: /* Perl 5.004 onwards omits VT from \s, but we must preserve it
3876: if it was previously set by something earlier in the character
3877: class. */
3878:
3879: case ESC_s:
3880: classbits[0] |= cbits[cbit_space];
3881: classbits[1] |= cbits[cbit_space+1] & ~0x08;
3882: for (c = 2; c < 32; c++) classbits[c] |= cbits[c+cbit_space];
3883: continue;
3884:
3885: case ESC_S:
3886: should_flip_negation = TRUE;
3887: for (c = 0; c < 32; c++) classbits[c] |= ~cbits[c+cbit_space];
3888: classbits[1] |= 0x08; /* Perl 5.004 onwards omits VT from \s */
3889: continue;
3890:
3891: case ESC_h:
3892: SETBIT(classbits, 0x09); /* VT */
3893: SETBIT(classbits, 0x20); /* SPACE */
3894: SETBIT(classbits, 0xa0); /* NSBP */
3895: #ifdef SUPPORT_UTF8
3896: if (utf8)
3897: {
3898: class_utf8 = TRUE;
3899: *class_utf8data++ = XCL_SINGLE;
3900: class_utf8data += _pcre_ord2utf8(0x1680, class_utf8data);
3901: *class_utf8data++ = XCL_SINGLE;
3902: class_utf8data += _pcre_ord2utf8(0x180e, class_utf8data);
3903: *class_utf8data++ = XCL_RANGE;
3904: class_utf8data += _pcre_ord2utf8(0x2000, class_utf8data);
3905: class_utf8data += _pcre_ord2utf8(0x200A, class_utf8data);
3906: *class_utf8data++ = XCL_SINGLE;
3907: class_utf8data += _pcre_ord2utf8(0x202f, class_utf8data);
3908: *class_utf8data++ = XCL_SINGLE;
3909: class_utf8data += _pcre_ord2utf8(0x205f, class_utf8data);
3910: *class_utf8data++ = XCL_SINGLE;
3911: class_utf8data += _pcre_ord2utf8(0x3000, class_utf8data);
3912: }
3913: #endif
3914: continue;
3915:
3916: case ESC_H:
3917: for (c = 0; c < 32; c++)
3918: {
3919: int x = 0xff;
3920: switch (c)
3921: {
3922: case 0x09/8: x ^= 1 << (0x09%8); break;
3923: case 0x20/8: x ^= 1 << (0x20%8); break;
3924: case 0xa0/8: x ^= 1 << (0xa0%8); break;
3925: default: break;
3926: }
3927: classbits[c] |= x;
3928: }
3929:
3930: #ifdef SUPPORT_UTF8
3931: if (utf8)
3932: {
3933: class_utf8 = TRUE;
3934: *class_utf8data++ = XCL_RANGE;
3935: class_utf8data += _pcre_ord2utf8(0x0100, class_utf8data);
3936: class_utf8data += _pcre_ord2utf8(0x167f, class_utf8data);
3937: *class_utf8data++ = XCL_RANGE;
3938: class_utf8data += _pcre_ord2utf8(0x1681, class_utf8data);
3939: class_utf8data += _pcre_ord2utf8(0x180d, class_utf8data);
3940: *class_utf8data++ = XCL_RANGE;
3941: class_utf8data += _pcre_ord2utf8(0x180f, class_utf8data);
3942: class_utf8data += _pcre_ord2utf8(0x1fff, class_utf8data);
3943: *class_utf8data++ = XCL_RANGE;
3944: class_utf8data += _pcre_ord2utf8(0x200B, class_utf8data);
3945: class_utf8data += _pcre_ord2utf8(0x202e, class_utf8data);
3946: *class_utf8data++ = XCL_RANGE;
3947: class_utf8data += _pcre_ord2utf8(0x2030, class_utf8data);
3948: class_utf8data += _pcre_ord2utf8(0x205e, class_utf8data);
3949: *class_utf8data++ = XCL_RANGE;
3950: class_utf8data += _pcre_ord2utf8(0x2060, class_utf8data);
3951: class_utf8data += _pcre_ord2utf8(0x2fff, class_utf8data);
3952: *class_utf8data++ = XCL_RANGE;
3953: class_utf8data += _pcre_ord2utf8(0x3001, class_utf8data);
3954: class_utf8data += _pcre_ord2utf8(0x7fffffff, class_utf8data);
3955: }
3956: #endif
3957: continue;
3958:
3959: case ESC_v:
3960: SETBIT(classbits, 0x0a); /* LF */
3961: SETBIT(classbits, 0x0b); /* VT */
3962: SETBIT(classbits, 0x0c); /* FF */
3963: SETBIT(classbits, 0x0d); /* CR */
3964: SETBIT(classbits, 0x85); /* NEL */
3965: #ifdef SUPPORT_UTF8
3966: if (utf8)
3967: {
3968: class_utf8 = TRUE;
3969: *class_utf8data++ = XCL_RANGE;
3970: class_utf8data += _pcre_ord2utf8(0x2028, class_utf8data);
3971: class_utf8data += _pcre_ord2utf8(0x2029, class_utf8data);
3972: }
3973: #endif
3974: continue;
3975:
3976: case ESC_V:
3977: for (c = 0; c < 32; c++)
3978: {
3979: int x = 0xff;
3980: switch (c)
3981: {
3982: case 0x0a/8: x ^= 1 << (0x0a%8);
3983: x ^= 1 << (0x0b%8);
3984: x ^= 1 << (0x0c%8);
3985: x ^= 1 << (0x0d%8);
3986: break;
3987: case 0x85/8: x ^= 1 << (0x85%8); break;
3988: default: break;
3989: }
3990: classbits[c] |= x;
3991: }
3992:
3993: #ifdef SUPPORT_UTF8
3994: if (utf8)
3995: {
3996: class_utf8 = TRUE;
3997: *class_utf8data++ = XCL_RANGE;
3998: class_utf8data += _pcre_ord2utf8(0x0100, class_utf8data);
3999: class_utf8data += _pcre_ord2utf8(0x2027, class_utf8data);
4000: *class_utf8data++ = XCL_RANGE;
4001: class_utf8data += _pcre_ord2utf8(0x2029, class_utf8data);
4002: class_utf8data += _pcre_ord2utf8(0x7fffffff, class_utf8data);
4003: }
4004: #endif
4005: continue;
4006:
4007: #ifdef SUPPORT_UCP
4008: case ESC_p:
4009: case ESC_P:
4010: {
4011: BOOL negated;
4012: int pdata;
4013: int ptype = get_ucp(&ptr, &negated, &pdata, errorcodeptr);
4014: if (ptype < 0) goto FAILED;
4015: class_utf8 = TRUE;
4016: *class_utf8data++ = ((-c == ESC_p) != negated)?
4017: XCL_PROP : XCL_NOTPROP;
4018: *class_utf8data++ = ptype;
4019: *class_utf8data++ = pdata;
4020: class_charcount -= 2; /* Not a < 256 character */
4021: continue;
4022: }
4023: #endif
4024: /* Unrecognized escapes are faulted if PCRE is running in its
4025: strict mode. By default, for compatibility with Perl, they are
4026: treated as literals. */
4027:
4028: default:
4029: if ((options & PCRE_EXTRA) != 0)
4030: {
4031: *errorcodeptr = ERR7;
4032: goto FAILED;
4033: }
4034: class_charcount -= 2; /* Undo the default count from above */
4035: c = *ptr; /* Get the final character and fall through */
4036: break;
4037: }
4038: }
4039:
4040: /* Fall through if we have a single character (c >= 0). This may be
4041: greater than 256 in UTF-8 mode. */
4042:
4043: } /* End of backslash handling */
4044:
4045: /* A single character may be followed by '-' to form a range. However,
4046: Perl does not permit ']' to be the end of the range. A '-' character
4047: at the end is treated as a literal. Perl ignores orphaned \E sequences
4048: entirely. The code for handling \Q and \E is messy. */
4049:
4050: CHECK_RANGE:
4051: while (ptr[1] == CHAR_BACKSLASH && ptr[2] == CHAR_E)
4052: {
4053: inescq = FALSE;
4054: ptr += 2;
4055: }
4056:
4057: oldptr = ptr;
4058:
4059: /* Remember \r or \n */
4060:
4061: if (c == CHAR_CR || c == CHAR_NL) cd->external_flags |= PCRE_HASCRORLF;
4062:
4063: /* Check for range */
4064:
4065: if (!inescq && ptr[1] == CHAR_MINUS)
4066: {
4067: int d;
4068: ptr += 2;
4069: while (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_E) ptr += 2;
4070:
4071: /* If we hit \Q (not followed by \E) at this point, go into escaped
4072: mode. */
4073:
4074: while (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_Q)
4075: {
4076: ptr += 2;
4077: if (*ptr == CHAR_BACKSLASH && ptr[1] == CHAR_E)
4078: { ptr += 2; continue; }
4079: inescq = TRUE;
4080: break;
4081: }
4082:
4083: if (*ptr == 0 || (!inescq && *ptr == CHAR_RIGHT_SQUARE_BRACKET))
4084: {
4085: ptr = oldptr;
4086: goto LONE_SINGLE_CHARACTER;
4087: }
4088:
4089: #ifdef SUPPORT_UTF8
4090: if (utf8)
4091: { /* Braces are required because the */
4092: GETCHARLEN(d, ptr, ptr); /* macro generates multiple statements */
4093: }
4094: else
4095: #endif
4096: d = *ptr; /* Not UTF-8 mode */
4097:
4098: /* The second part of a range can be a single-character escape, but
4099: not any of the other escapes. Perl 5.6 treats a hyphen as a literal
4100: in such circumstances. */
4101:
4102: if (!inescq && d == CHAR_BACKSLASH)
4103: {
4104: d = check_escape(&ptr, errorcodeptr, cd->bracount, options, TRUE);
4105: if (*errorcodeptr != 0) goto FAILED;
4106:
4107: /* \b is backspace; any other special means the '-' was literal */
4108:
4109: if (d < 0)
4110: {
4111: if (d == -ESC_b) d = CHAR_BS; else
4112: {
4113: ptr = oldptr;
4114: goto LONE_SINGLE_CHARACTER; /* A few lines below */
4115: }
4116: }
4117: }
4118:
4119: /* Check that the two values are in the correct order. Optimize
4120: one-character ranges */
4121:
4122: if (d < c)
4123: {
4124: *errorcodeptr = ERR8;
4125: goto FAILED;
4126: }
4127:
4128: if (d == c) goto LONE_SINGLE_CHARACTER; /* A few lines below */
4129:
4130: /* Remember \r or \n */
4131:
4132: if (d == CHAR_CR || d == CHAR_NL) cd->external_flags |= PCRE_HASCRORLF;
4133:
4134: /* In UTF-8 mode, if the upper limit is > 255, or > 127 for caseless
4135: matching, we have to use an XCLASS with extra data items. Caseless
4136: matching for characters > 127 is available only if UCP support is
4137: available. */
4138:
4139: #ifdef SUPPORT_UTF8
4140: if (utf8 && (d > 255 || ((options & PCRE_CASELESS) != 0 && d > 127)))
4141: {
4142: class_utf8 = TRUE;
4143:
4144: /* With UCP support, we can find the other case equivalents of
4145: the relevant characters. There may be several ranges. Optimize how
4146: they fit with the basic range. */
4147:
4148: #ifdef SUPPORT_UCP
4149: if ((options & PCRE_CASELESS) != 0)
4150: {
4151: unsigned int occ, ocd;
4152: unsigned int cc = c;
4153: unsigned int origd = d;
4154: while (get_othercase_range(&cc, origd, &occ, &ocd))
4155: {
4156: if (occ >= (unsigned int)c &&
4157: ocd <= (unsigned int)d)
4158: continue; /* Skip embedded ranges */
4159:
4160: if (occ < (unsigned int)c &&
4161: ocd >= (unsigned int)c - 1) /* Extend the basic range */
4162: { /* if there is overlap, */
4163: c = occ; /* noting that if occ < c */
4164: continue; /* we can't have ocd > d */
4165: } /* because a subrange is */
4166: if (ocd > (unsigned int)d &&
4167: occ <= (unsigned int)d + 1) /* always shorter than */
4168: { /* the basic range. */
4169: d = ocd;
4170: continue;
4171: }
4172:
4173: if (occ == ocd)
4174: {
4175: *class_utf8data++ = XCL_SINGLE;
4176: }
4177: else
4178: {
4179: *class_utf8data++ = XCL_RANGE;
4180: class_utf8data += _pcre_ord2utf8(occ, class_utf8data);
4181: }
4182: class_utf8data += _pcre_ord2utf8(ocd, class_utf8data);
4183: }
4184: }
4185: #endif /* SUPPORT_UCP */
4186:
4187: /* Now record the original range, possibly modified for UCP caseless
4188: overlapping ranges. */
4189:
4190: *class_utf8data++ = XCL_RANGE;
4191: class_utf8data += _pcre_ord2utf8(c, class_utf8data);
4192: class_utf8data += _pcre_ord2utf8(d, class_utf8data);
4193:
4194: /* With UCP support, we are done. Without UCP support, there is no
4195: caseless matching for UTF-8 characters > 127; we can use the bit map
4196: for the smaller ones. */
4197:
4198: #ifdef SUPPORT_UCP
4199: continue; /* With next character in the class */
4200: #else
4201: if ((options & PCRE_CASELESS) == 0 || c > 127) continue;
4202:
4203: /* Adjust upper limit and fall through to set up the map */
4204:
4205: d = 127;
4206:
4207: #endif /* SUPPORT_UCP */
4208: }
4209: #endif /* SUPPORT_UTF8 */
4210:
4211: /* We use the bit map for all cases when not in UTF-8 mode; else
4212: ranges that lie entirely within 0-127 when there is UCP support; else
4213: for partial ranges without UCP support. */
4214:
4215: class_charcount += d - c + 1;
4216: class_lastchar = d;
4217:
4218: /* We can save a bit of time by skipping this in the pre-compile. */
4219:
4220: if (lengthptr == NULL) for (; c <= d; c++)
4221: {
4222: classbits[c/8] |= (1 << (c&7));
4223: if ((options & PCRE_CASELESS) != 0)
4224: {
4225: int uc = cd->fcc[c]; /* flip case */
4226: classbits[uc/8] |= (1 << (uc&7));
4227: }
4228: }
4229:
4230: continue; /* Go get the next char in the class */
4231: }
4232:
4233: /* Handle a lone single character - we can get here for a normal
4234: non-escape char, or after \ that introduces a single character or for an
4235: apparent range that isn't. */
4236:
4237: LONE_SINGLE_CHARACTER:
4238:
4239: /* Handle a character that cannot go in the bit map */
4240:
4241: #ifdef SUPPORT_UTF8
4242: if (utf8 && (c > 255 || ((options & PCRE_CASELESS) != 0 && c > 127)))
4243: {
4244: class_utf8 = TRUE;
4245: *class_utf8data++ = XCL_SINGLE;
4246: class_utf8data += _pcre_ord2utf8(c, class_utf8data);
4247:
4248: #ifdef SUPPORT_UCP
4249: if ((options & PCRE_CASELESS) != 0)
4250: {
4251: unsigned int othercase;
4252: if ((othercase = UCD_OTHERCASE(c)) != c)
4253: {
4254: *class_utf8data++ = XCL_SINGLE;
4255: class_utf8data += _pcre_ord2utf8(othercase, class_utf8data);
4256: }
4257: }
4258: #endif /* SUPPORT_UCP */
4259:
4260: }
4261: else
4262: #endif /* SUPPORT_UTF8 */
4263:
4264: /* Handle a single-byte character */
4265: {
4266: classbits[c/8] |= (1 << (c&7));
4267: if ((options & PCRE_CASELESS) != 0)
4268: {
4269: c = cd->fcc[c]; /* flip case */
4270: classbits[c/8] |= (1 << (c&7));
4271: }
4272: class_charcount++;
4273: class_lastchar = c;
4274: }
4275: }
4276:
4277: /* Loop until ']' reached. This "while" is the end of the "do" far above.
4278: If we are at the end of an internal nested string, revert to the outer
4279: string. */
4280:
4281: while (((c = *(++ptr)) != 0 ||
4282: (nestptr != NULL &&
4283: (ptr = nestptr, nestptr = NULL, c = *(++ptr)) != 0)) &&
4284: (c != CHAR_RIGHT_SQUARE_BRACKET || inescq));
4285:
4286: /* Check for missing terminating ']' */
4287:
4288: if (c == 0)
4289: {
4290: *errorcodeptr = ERR6;
4291: goto FAILED;
4292: }
4293:
4294: /* If class_charcount is 1, we saw precisely one character whose value is
4295: less than 256. As long as there were no characters >= 128 and there was no
4296: use of \p or \P, in other words, no use of any XCLASS features, we can
4297: optimize.
4298:
4299: In UTF-8 mode, we can optimize the negative case only if there were no
4300: characters >= 128 because OP_NOT and the related opcodes like OP_NOTSTAR
4301: operate on single-bytes characters only. This is an historical hangover.
4302: Maybe one day we can tidy these opcodes to handle multi-byte characters.
4303:
4304: The optimization throws away the bit map. We turn the item into a
4305: 1-character OP_CHAR[I] if it's positive, or OP_NOT[I] if it's negative.
4306: Note that OP_NOT[I] does not support multibyte characters. In the positive
4307: case, it can cause firstbyte to be set. Otherwise, there can be no first
4308: char if this item is first, whatever repeat count may follow. In the case
4309: of reqbyte, save the previous value for reinstating. */
4310:
4311: #ifdef SUPPORT_UTF8
4312: if (class_charcount == 1 && !class_utf8 &&
4313: (!utf8 || !negate_class || class_lastchar < 128))
4314: #else
4315: if (class_charcount == 1)
4316: #endif
4317: {
4318: zeroreqbyte = reqbyte;
4319:
4320: /* The OP_NOT[I] opcodes work on one-byte characters only. */
4321:
4322: if (negate_class)
4323: {
4324: if (firstbyte == REQ_UNSET) firstbyte = REQ_NONE;
4325: zerofirstbyte = firstbyte;
4326: *code++ = ((options & PCRE_CASELESS) != 0)? OP_NOTI: OP_NOT;
4327: *code++ = class_lastchar;
4328: break;
4329: }
4330:
4331: /* For a single, positive character, get the value into mcbuffer, and
4332: then we can handle this with the normal one-character code. */
4333:
4334: #ifdef SUPPORT_UTF8
4335: if (utf8 && class_lastchar > 127)
4336: mclength = _pcre_ord2utf8(class_lastchar, mcbuffer);
4337: else
4338: #endif
4339: {
4340: mcbuffer[0] = class_lastchar;
4341: mclength = 1;
4342: }
4343: goto ONE_CHAR;
4344: } /* End of 1-char optimization */
4345:
4346: /* The general case - not the one-char optimization. If this is the first
4347: thing in the branch, there can be no first char setting, whatever the
4348: repeat count. Any reqbyte setting must remain unchanged after any kind of
4349: repeat. */
4350:
4351: if (firstbyte == REQ_UNSET) firstbyte = REQ_NONE;
4352: zerofirstbyte = firstbyte;
4353: zeroreqbyte = reqbyte;
4354:
4355: /* If there are characters with values > 255, we have to compile an
4356: extended class, with its own opcode, unless there was a negated special
4357: such as \S in the class, and PCRE_UCP is not set, because in that case all
4358: characters > 255 are in the class, so any that were explicitly given as
4359: well can be ignored. If (when there are explicit characters > 255 that must
4360: be listed) there are no characters < 256, we can omit the bitmap in the
4361: actual compiled code. */
4362:
4363: #ifdef SUPPORT_UTF8
4364: if (class_utf8 && (!should_flip_negation || (options & PCRE_UCP) != 0))
4365: {
4366: *class_utf8data++ = XCL_END; /* Marks the end of extra data */
4367: *code++ = OP_XCLASS;
4368: code += LINK_SIZE;
4369: *code = negate_class? XCL_NOT : 0;
4370:
4371: /* If the map is required, move up the extra data to make room for it;
4372: otherwise just move the code pointer to the end of the extra data. */
4373:
4374: if (class_charcount > 0)
4375: {
4376: *code++ |= XCL_MAP;
4377: memmove(code + 32, code, class_utf8data - code);
4378: memcpy(code, classbits, 32);
4379: code = class_utf8data + 32;
4380: }
4381: else code = class_utf8data;
4382:
4383: /* Now fill in the complete length of the item */
4384:
4385: PUT(previous, 1, (int)(code - previous));
4386: break; /* End of class handling */
4387: }
4388: #endif
4389:
4390: /* If there are no characters > 255, or they are all to be included or
4391: excluded, set the opcode to OP_CLASS or OP_NCLASS, depending on whether the
4392: whole class was negated and whether there were negative specials such as \S
4393: (non-UCP) in the class. Then copy the 32-byte map into the code vector,
4394: negating it if necessary. */
4395:
4396: *code++ = (negate_class == should_flip_negation) ? OP_CLASS : OP_NCLASS;
4397: if (negate_class)
4398: {
4399: if (lengthptr == NULL) /* Save time in the pre-compile phase */
4400: for (c = 0; c < 32; c++) code[c] = ~classbits[c];
4401: }
4402: else
4403: {
4404: memcpy(code, classbits, 32);
4405: }
4406: code += 32;
4407: break;
4408:
4409:
4410: /* ===================================================================*/
4411: /* Various kinds of repeat; '{' is not necessarily a quantifier, but this
4412: has been tested above. */
4413:
4414: case CHAR_LEFT_CURLY_BRACKET:
4415: if (!is_quantifier) goto NORMAL_CHAR;
4416: ptr = read_repeat_counts(ptr+1, &repeat_min, &repeat_max, errorcodeptr);
4417: if (*errorcodeptr != 0) goto FAILED;
4418: goto REPEAT;
4419:
4420: case CHAR_ASTERISK:
4421: repeat_min = 0;
4422: repeat_max = -1;
4423: goto REPEAT;
4424:
4425: case CHAR_PLUS:
4426: repeat_min = 1;
4427: repeat_max = -1;
4428: goto REPEAT;
4429:
4430: case CHAR_QUESTION_MARK:
4431: repeat_min = 0;
4432: repeat_max = 1;
4433:
4434: REPEAT:
4435: if (previous == NULL)
4436: {
4437: *errorcodeptr = ERR9;
4438: goto FAILED;
4439: }
4440:
4441: if (repeat_min == 0)
4442: {
4443: firstbyte = zerofirstbyte; /* Adjust for zero repeat */
4444: reqbyte = zeroreqbyte; /* Ditto */
4445: }
4446:
4447: /* Remember whether this is a variable length repeat */
4448:
4449: reqvary = (repeat_min == repeat_max)? 0 : REQ_VARY;
4450:
4451: op_type = 0; /* Default single-char op codes */
4452: possessive_quantifier = FALSE; /* Default not possessive quantifier */
4453:
4454: /* Save start of previous item, in case we have to move it up in order to
4455: insert something before it. */
4456:
4457: tempcode = previous;
4458:
4459: /* If the next character is '+', we have a possessive quantifier. This
4460: implies greediness, whatever the setting of the PCRE_UNGREEDY option.
4461: If the next character is '?' this is a minimizing repeat, by default,
4462: but if PCRE_UNGREEDY is set, it works the other way round. We change the
4463: repeat type to the non-default. */
4464:
4465: if (ptr[1] == CHAR_PLUS)
4466: {
4467: repeat_type = 0; /* Force greedy */
4468: possessive_quantifier = TRUE;
4469: ptr++;
4470: }
4471: else if (ptr[1] == CHAR_QUESTION_MARK)
4472: {
4473: repeat_type = greedy_non_default;
4474: ptr++;
4475: }
4476: else repeat_type = greedy_default;
4477:
4478: /* If previous was a recursion call, wrap it in atomic brackets so that
4479: previous becomes the atomic group. All recursions were so wrapped in the
4480: past, but it no longer happens for non-repeated recursions. In fact, the
4481: repeated ones could be re-implemented independently so as not to need this,
4482: but for the moment we rely on the code for repeating groups. */
4483:
4484: if (*previous == OP_RECURSE)
4485: {
4486: memmove(previous + 1 + LINK_SIZE, previous, 1 + LINK_SIZE);
4487: *previous = OP_ONCE;
4488: PUT(previous, 1, 2 + 2*LINK_SIZE);
4489: previous[2 + 2*LINK_SIZE] = OP_KET;
4490: PUT(previous, 3 + 2*LINK_SIZE, 2 + 2*LINK_SIZE);
4491: code += 2 + 2 * LINK_SIZE;
4492: length_prevgroup = 3 + 3*LINK_SIZE;
4493:
4494: /* When actually compiling, we need to check whether this was a forward
4495: reference, and if so, adjust the offset. */
4496:
4497: if (lengthptr == NULL && cd->hwm >= cd->start_workspace + LINK_SIZE)
4498: {
4499: int offset = GET(cd->hwm, -LINK_SIZE);
4500: if (offset == previous + 1 - cd->start_code)
4501: PUT(cd->hwm, -LINK_SIZE, offset + 1 + LINK_SIZE);
4502: }
4503: }
4504:
4505: /* Now handle repetition for the different types of item. */
4506:
4507: /* If previous was a character match, abolish the item and generate a
4508: repeat item instead. If a char item has a minumum of more than one, ensure
4509: that it is set in reqbyte - it might not be if a sequence such as x{3} is
4510: the first thing in a branch because the x will have gone into firstbyte
4511: instead. */
4512:
4513: if (*previous == OP_CHAR || *previous == OP_CHARI)
4514: {
4515: op_type = (*previous == OP_CHAR)? 0 : OP_STARI - OP_STAR;
4516:
4517: /* Deal with UTF-8 characters that take up more than one byte. It's
4518: easier to write this out separately than try to macrify it. Use c to
4519: hold the length of the character in bytes, plus 0x80 to flag that it's a
4520: length rather than a small character. */
4521:
4522: #ifdef SUPPORT_UTF8
4523: if (utf8 && (code[-1] & 0x80) != 0)
4524: {
4525: uschar *lastchar = code - 1;
4526: while((*lastchar & 0xc0) == 0x80) lastchar--;
4527: c = (int)(code - lastchar); /* Length of UTF-8 character */
4528: memcpy(utf8_char, lastchar, c); /* Save the char */
4529: c |= 0x80; /* Flag c as a length */
4530: }
4531: else
4532: #endif
4533:
4534: /* Handle the case of a single byte - either with no UTF8 support, or
4535: with UTF-8 disabled, or for a UTF-8 character < 128. */
4536:
4537: {
4538: c = code[-1];
4539: if (repeat_min > 1) reqbyte = c | req_caseopt | cd->req_varyopt;
4540: }
4541:
4542: /* If the repetition is unlimited, it pays to see if the next thing on
4543: the line is something that cannot possibly match this character. If so,
4544: automatically possessifying this item gains some performance in the case
4545: where the match fails. */
4546:
4547: if (!possessive_quantifier &&
4548: repeat_max < 0 &&
4549: check_auto_possessive(previous, utf8, ptr + 1, options, cd))
4550: {
4551: repeat_type = 0; /* Force greedy */
4552: possessive_quantifier = TRUE;
4553: }
4554:
4555: goto OUTPUT_SINGLE_REPEAT; /* Code shared with single character types */
4556: }
4557:
4558: /* If previous was a single negated character ([^a] or similar), we use
4559: one of the special opcodes, replacing it. The code is shared with single-
4560: character repeats by setting opt_type to add a suitable offset into
4561: repeat_type. We can also test for auto-possessification. OP_NOT and OP_NOTI
4562: are currently used only for single-byte chars. */
4563:
4564: else if (*previous == OP_NOT || *previous == OP_NOTI)
4565: {
4566: op_type = ((*previous == OP_NOT)? OP_NOTSTAR : OP_NOTSTARI) - OP_STAR;
4567: c = previous[1];
4568: if (!possessive_quantifier &&
4569: repeat_max < 0 &&
4570: check_auto_possessive(previous, utf8, ptr + 1, options, cd))
4571: {
4572: repeat_type = 0; /* Force greedy */
4573: possessive_quantifier = TRUE;
4574: }
4575: goto OUTPUT_SINGLE_REPEAT;
4576: }
4577:
4578: /* If previous was a character type match (\d or similar), abolish it and
4579: create a suitable repeat item. The code is shared with single-character
4580: repeats by setting op_type to add a suitable offset into repeat_type. Note
4581: the the Unicode property types will be present only when SUPPORT_UCP is
4582: defined, but we don't wrap the little bits of code here because it just
4583: makes it horribly messy. */
4584:
4585: else if (*previous < OP_EODN)
4586: {
4587: uschar *oldcode;
4588: int prop_type, prop_value;
4589: op_type = OP_TYPESTAR - OP_STAR; /* Use type opcodes */
4590: c = *previous;
4591:
4592: if (!possessive_quantifier &&
4593: repeat_max < 0 &&
4594: check_auto_possessive(previous, utf8, ptr + 1, options, cd))
4595: {
4596: repeat_type = 0; /* Force greedy */
4597: possessive_quantifier = TRUE;
4598: }
4599:
4600: OUTPUT_SINGLE_REPEAT:
4601: if (*previous == OP_PROP || *previous == OP_NOTPROP)
4602: {
4603: prop_type = previous[1];
4604: prop_value = previous[2];
4605: }
4606: else prop_type = prop_value = -1;
4607:
4608: oldcode = code;
4609: code = previous; /* Usually overwrite previous item */
4610:
4611: /* If the maximum is zero then the minimum must also be zero; Perl allows
4612: this case, so we do too - by simply omitting the item altogether. */
4613:
4614: if (repeat_max == 0) goto END_REPEAT;
4615:
4616: /*--------------------------------------------------------------------*/
4617: /* This code is obsolete from release 8.00; the restriction was finally
4618: removed: */
4619:
4620: /* All real repeats make it impossible to handle partial matching (maybe
4621: one day we will be able to remove this restriction). */
4622:
4623: /* if (repeat_max != 1) cd->external_flags |= PCRE_NOPARTIAL; */
4624: /*--------------------------------------------------------------------*/
4625:
4626: /* Combine the op_type with the repeat_type */
4627:
4628: repeat_type += op_type;
4629:
4630: /* A minimum of zero is handled either as the special case * or ?, or as
4631: an UPTO, with the maximum given. */
4632:
4633: if (repeat_min == 0)
4634: {
4635: if (repeat_max == -1) *code++ = OP_STAR + repeat_type;
4636: else if (repeat_max == 1) *code++ = OP_QUERY + repeat_type;
4637: else
4638: {
4639: *code++ = OP_UPTO + repeat_type;
4640: PUT2INC(code, 0, repeat_max);
4641: }
4642: }
4643:
4644: /* A repeat minimum of 1 is optimized into some special cases. If the
4645: maximum is unlimited, we use OP_PLUS. Otherwise, the original item is
4646: left in place and, if the maximum is greater than 1, we use OP_UPTO with
4647: one less than the maximum. */
4648:
4649: else if (repeat_min == 1)
4650: {
4651: if (repeat_max == -1)
4652: *code++ = OP_PLUS + repeat_type;
4653: else
4654: {
4655: code = oldcode; /* leave previous item in place */
4656: if (repeat_max == 1) goto END_REPEAT;
4657: *code++ = OP_UPTO + repeat_type;
4658: PUT2INC(code, 0, repeat_max - 1);
4659: }
4660: }
4661:
4662: /* The case {n,n} is just an EXACT, while the general case {n,m} is
4663: handled as an EXACT followed by an UPTO. */
4664:
4665: else
4666: {
4667: *code++ = OP_EXACT + op_type; /* NB EXACT doesn't have repeat_type */
4668: PUT2INC(code, 0, repeat_min);
4669:
4670: /* If the maximum is unlimited, insert an OP_STAR. Before doing so,
4671: we have to insert the character for the previous code. For a repeated
4672: Unicode property match, there are two extra bytes that define the
4673: required property. In UTF-8 mode, long characters have their length in
4674: c, with the 0x80 bit as a flag. */
4675:
4676: if (repeat_max < 0)
4677: {
4678: #ifdef SUPPORT_UTF8
4679: if (utf8 && c >= 128)
4680: {
4681: memcpy(code, utf8_char, c & 7);
4682: code += c & 7;
4683: }
4684: else
4685: #endif
4686: {
4687: *code++ = c;
4688: if (prop_type >= 0)
4689: {
4690: *code++ = prop_type;
4691: *code++ = prop_value;
4692: }
4693: }
4694: *code++ = OP_STAR + repeat_type;
4695: }
4696:
4697: /* Else insert an UPTO if the max is greater than the min, again
4698: preceded by the character, for the previously inserted code. If the
4699: UPTO is just for 1 instance, we can use QUERY instead. */
4700:
4701: else if (repeat_max != repeat_min)
4702: {
4703: #ifdef SUPPORT_UTF8
4704: if (utf8 && c >= 128)
4705: {
4706: memcpy(code, utf8_char, c & 7);
4707: code += c & 7;
4708: }
4709: else
4710: #endif
4711: *code++ = c;
4712: if (prop_type >= 0)
4713: {
4714: *code++ = prop_type;
4715: *code++ = prop_value;
4716: }
4717: repeat_max -= repeat_min;
4718:
4719: if (repeat_max == 1)
4720: {
4721: *code++ = OP_QUERY + repeat_type;
4722: }
4723: else
4724: {
4725: *code++ = OP_UPTO + repeat_type;
4726: PUT2INC(code, 0, repeat_max);
4727: }
4728: }
4729: }
4730:
4731: /* The character or character type itself comes last in all cases. */
4732:
4733: #ifdef SUPPORT_UTF8
4734: if (utf8 && c >= 128)
4735: {
4736: memcpy(code, utf8_char, c & 7);
4737: code += c & 7;
4738: }
4739: else
4740: #endif
4741: *code++ = c;
4742:
4743: /* For a repeated Unicode property match, there are two extra bytes that
4744: define the required property. */
4745:
4746: #ifdef SUPPORT_UCP
4747: if (prop_type >= 0)
4748: {
4749: *code++ = prop_type;
4750: *code++ = prop_value;
4751: }
4752: #endif
4753: }
4754:
4755: /* If previous was a character class or a back reference, we put the repeat
4756: stuff after it, but just skip the item if the repeat was {0,0}. */
4757:
4758: else if (*previous == OP_CLASS ||
4759: *previous == OP_NCLASS ||
4760: #ifdef SUPPORT_UTF8
4761: *previous == OP_XCLASS ||
4762: #endif
4763: *previous == OP_REF ||
4764: *previous == OP_REFI)
4765: {
4766: if (repeat_max == 0)
4767: {
4768: code = previous;
4769: goto END_REPEAT;
4770: }
4771:
4772: /*--------------------------------------------------------------------*/
4773: /* This code is obsolete from release 8.00; the restriction was finally
4774: removed: */
4775:
4776: /* All real repeats make it impossible to handle partial matching (maybe
4777: one day we will be able to remove this restriction). */
4778:
4779: /* if (repeat_max != 1) cd->external_flags |= PCRE_NOPARTIAL; */
4780: /*--------------------------------------------------------------------*/
4781:
4782: if (repeat_min == 0 && repeat_max == -1)
4783: *code++ = OP_CRSTAR + repeat_type;
4784: else if (repeat_min == 1 && repeat_max == -1)
4785: *code++ = OP_CRPLUS + repeat_type;
4786: else if (repeat_min == 0 && repeat_max == 1)
4787: *code++ = OP_CRQUERY + repeat_type;
4788: else
4789: {
4790: *code++ = OP_CRRANGE + repeat_type;
4791: PUT2INC(code, 0, repeat_min);
4792: if (repeat_max == -1) repeat_max = 0; /* 2-byte encoding for max */
4793: PUT2INC(code, 0, repeat_max);
4794: }
4795: }
4796:
4797: /* If previous was a bracket group, we may have to replicate it in certain
4798: cases. Note that at this point we can encounter only the "basic" bracket
4799: opcodes such as BRA and CBRA, as this is the place where they get converted
4800: into the more special varieties such as BRAPOS and SBRA. A test for >=
4801: OP_ASSERT and <= OP_COND includes ASSERT, ASSERT_NOT, ASSERTBACK,
4802: ASSERTBACK_NOT, ONCE, BRA, CBRA, and COND. Originally, PCRE did not allow
4803: repetition of assertions, but now it does, for Perl compatibility. */
4804:
4805: else if (*previous >= OP_ASSERT && *previous <= OP_COND)
4806: {
4807: register int i;
4808: int len = (int)(code - previous);
4809: uschar *bralink = NULL;
4810: uschar *brazeroptr = NULL;
4811:
4812: /* Repeating a DEFINE group is pointless, but Perl allows the syntax, so
4813: we just ignore the repeat. */
4814:
4815: if (*previous == OP_COND && previous[LINK_SIZE+1] == OP_DEF)
4816: goto END_REPEAT;
4817:
4818: /* There is no sense in actually repeating assertions. The only potential
4819: use of repetition is in cases when the assertion is optional. Therefore,
4820: if the minimum is greater than zero, just ignore the repeat. If the
4821: maximum is not not zero or one, set it to 1. */
4822:
4823: if (*previous < OP_ONCE) /* Assertion */
4824: {
4825: if (repeat_min > 0) goto END_REPEAT;
4826: if (repeat_max < 0 || repeat_max > 1) repeat_max = 1;
4827: }
4828:
4829: /* The case of a zero minimum is special because of the need to stick
4830: OP_BRAZERO in front of it, and because the group appears once in the
4831: data, whereas in other cases it appears the minimum number of times. For
4832: this reason, it is simplest to treat this case separately, as otherwise
4833: the code gets far too messy. There are several special subcases when the
4834: minimum is zero. */
4835:
4836: if (repeat_min == 0)
4837: {
4838: /* If the maximum is also zero, we used to just omit the group from the
4839: output altogether, like this:
4840:
4841: ** if (repeat_max == 0)
4842: ** {
4843: ** code = previous;
4844: ** goto END_REPEAT;
4845: ** }
4846:
4847: However, that fails when a group or a subgroup within it is referenced
4848: as a subroutine from elsewhere in the pattern, so now we stick in
4849: OP_SKIPZERO in front of it so that it is skipped on execution. As we
4850: don't have a list of which groups are referenced, we cannot do this
4851: selectively.
4852:
4853: If the maximum is 1 or unlimited, we just have to stick in the BRAZERO
4854: and do no more at this point. However, we do need to adjust any
4855: OP_RECURSE calls inside the group that refer to the group itself or any
4856: internal or forward referenced group, because the offset is from the
4857: start of the whole regex. Temporarily terminate the pattern while doing
4858: this. */
4859:
4860: if (repeat_max <= 1) /* Covers 0, 1, and unlimited */
4861: {
4862: *code = OP_END;
4863: adjust_recurse(previous, 1, utf8, cd, save_hwm);
4864: memmove(previous+1, previous, len);
4865: code++;
4866: if (repeat_max == 0)
4867: {
4868: *previous++ = OP_SKIPZERO;
4869: goto END_REPEAT;
4870: }
4871: brazeroptr = previous; /* Save for possessive optimizing */
4872: *previous++ = OP_BRAZERO + repeat_type;
4873: }
4874:
4875: /* If the maximum is greater than 1 and limited, we have to replicate
4876: in a nested fashion, sticking OP_BRAZERO before each set of brackets.
4877: The first one has to be handled carefully because it's the original
4878: copy, which has to be moved up. The remainder can be handled by code
4879: that is common with the non-zero minimum case below. We have to
4880: adjust the value or repeat_max, since one less copy is required. Once
4881: again, we may have to adjust any OP_RECURSE calls inside the group. */
4882:
4883: else
4884: {
4885: int offset;
4886: *code = OP_END;
4887: adjust_recurse(previous, 2 + LINK_SIZE, utf8, cd, save_hwm);
4888: memmove(previous + 2 + LINK_SIZE, previous, len);
4889: code += 2 + LINK_SIZE;
4890: *previous++ = OP_BRAZERO + repeat_type;
4891: *previous++ = OP_BRA;
4892:
4893: /* We chain together the bracket offset fields that have to be
4894: filled in later when the ends of the brackets are reached. */
4895:
4896: offset = (bralink == NULL)? 0 : (int)(previous - bralink);
4897: bralink = previous;
4898: PUTINC(previous, 0, offset);
4899: }
4900:
4901: repeat_max--;
4902: }
4903:
4904: /* If the minimum is greater than zero, replicate the group as many
4905: times as necessary, and adjust the maximum to the number of subsequent
4906: copies that we need. If we set a first char from the group, and didn't
4907: set a required char, copy the latter from the former. If there are any
4908: forward reference subroutine calls in the group, there will be entries on
4909: the workspace list; replicate these with an appropriate increment. */
4910:
4911: else
4912: {
4913: if (repeat_min > 1)
4914: {
4915: /* In the pre-compile phase, we don't actually do the replication. We
4916: just adjust the length as if we had. Do some paranoid checks for
4917: potential integer overflow. The INT64_OR_DOUBLE type is a 64-bit
4918: integer type when available, otherwise double. */
4919:
4920: if (lengthptr != NULL)
4921: {
4922: int delta = (repeat_min - 1)*length_prevgroup;
4923: if ((INT64_OR_DOUBLE)(repeat_min - 1)*
4924: (INT64_OR_DOUBLE)length_prevgroup >
4925: (INT64_OR_DOUBLE)INT_MAX ||
4926: OFLOW_MAX - *lengthptr < delta)
4927: {
4928: *errorcodeptr = ERR20;
4929: goto FAILED;
4930: }
4931: *lengthptr += delta;
4932: }
4933:
4934: /* This is compiling for real. If there is a set first byte for
4935: the group, and we have not yet set a "required byte", set it. Make
4936: sure there is enough workspace for copying forward references before
4937: doing the copy. */
4938:
4939: else
4940: {
4941: if (groupsetfirstbyte && reqbyte < 0) reqbyte = firstbyte;
4942:
4943: for (i = 1; i < repeat_min; i++)
4944: {
4945: uschar *hc;
4946: uschar *this_hwm = cd->hwm;
4947: memcpy(code, previous, len);
4948:
4949: while (cd->hwm > cd->start_workspace + cd->workspace_size -
4950: WORK_SIZE_SAFETY_MARGIN - (this_hwm - save_hwm))
4951: {
4952: int save_offset = save_hwm - cd->start_workspace;
4953: int this_offset = this_hwm - cd->start_workspace;
4954: *errorcodeptr = expand_workspace(cd);
4955: if (*errorcodeptr != 0) goto FAILED;
4956: save_hwm = (uschar *)cd->start_workspace + save_offset;
4957: this_hwm = (uschar *)cd->start_workspace + this_offset;
4958: }
4959:
4960: for (hc = save_hwm; hc < this_hwm; hc += LINK_SIZE)
4961: {
4962: PUT(cd->hwm, 0, GET(hc, 0) + len);
4963: cd->hwm += LINK_SIZE;
4964: }
4965: save_hwm = this_hwm;
4966: code += len;
4967: }
4968: }
4969: }
4970:
4971: if (repeat_max > 0) repeat_max -= repeat_min;
4972: }
4973:
4974: /* This code is common to both the zero and non-zero minimum cases. If
4975: the maximum is limited, it replicates the group in a nested fashion,
4976: remembering the bracket starts on a stack. In the case of a zero minimum,
4977: the first one was set up above. In all cases the repeat_max now specifies
4978: the number of additional copies needed. Again, we must remember to
4979: replicate entries on the forward reference list. */
4980:
4981: if (repeat_max >= 0)
4982: {
4983: /* In the pre-compile phase, we don't actually do the replication. We
4984: just adjust the length as if we had. For each repetition we must add 1
4985: to the length for BRAZERO and for all but the last repetition we must
4986: add 2 + 2*LINKSIZE to allow for the nesting that occurs. Do some
4987: paranoid checks to avoid integer overflow. The INT64_OR_DOUBLE type is
4988: a 64-bit integer type when available, otherwise double. */
4989:
4990: if (lengthptr != NULL && repeat_max > 0)
4991: {
4992: int delta = repeat_max * (length_prevgroup + 1 + 2 + 2*LINK_SIZE) -
4993: 2 - 2*LINK_SIZE; /* Last one doesn't nest */
4994: if ((INT64_OR_DOUBLE)repeat_max *
4995: (INT64_OR_DOUBLE)(length_prevgroup + 1 + 2 + 2*LINK_SIZE)
4996: > (INT64_OR_DOUBLE)INT_MAX ||
4997: OFLOW_MAX - *lengthptr < delta)
4998: {
4999: *errorcodeptr = ERR20;
5000: goto FAILED;
5001: }
5002: *lengthptr += delta;
5003: }
5004:
5005: /* This is compiling for real */
5006:
5007: else for (i = repeat_max - 1; i >= 0; i--)
5008: {
5009: uschar *hc;
5010: uschar *this_hwm = cd->hwm;
5011:
5012: *code++ = OP_BRAZERO + repeat_type;
5013:
5014: /* All but the final copy start a new nesting, maintaining the
5015: chain of brackets outstanding. */
5016:
5017: if (i != 0)
5018: {
5019: int offset;
5020: *code++ = OP_BRA;
5021: offset = (bralink == NULL)? 0 : (int)(code - bralink);
5022: bralink = code;
5023: PUTINC(code, 0, offset);
5024: }
5025:
5026: memcpy(code, previous, len);
5027:
5028: /* Ensure there is enough workspace for forward references before
5029: copying them. */
5030:
5031: while (cd->hwm > cd->start_workspace + cd->workspace_size -
5032: WORK_SIZE_SAFETY_MARGIN - (this_hwm - save_hwm))
5033: {
5034: int save_offset = save_hwm - cd->start_workspace;
5035: int this_offset = this_hwm - cd->start_workspace;
5036: *errorcodeptr = expand_workspace(cd);
5037: if (*errorcodeptr != 0) goto FAILED;
5038: save_hwm = (uschar *)cd->start_workspace + save_offset;
5039: this_hwm = (uschar *)cd->start_workspace + this_offset;
5040: }
5041:
5042: for (hc = save_hwm; hc < this_hwm; hc += LINK_SIZE)
5043: {
5044: PUT(cd->hwm, 0, GET(hc, 0) + len + ((i != 0)? 2+LINK_SIZE : 1));
5045: cd->hwm += LINK_SIZE;
5046: }
5047: save_hwm = this_hwm;
5048: code += len;
5049: }
5050:
5051: /* Now chain through the pending brackets, and fill in their length
5052: fields (which are holding the chain links pro tem). */
5053:
5054: while (bralink != NULL)
5055: {
5056: int oldlinkoffset;
5057: int offset = (int)(code - bralink + 1);
5058: uschar *bra = code - offset;
5059: oldlinkoffset = GET(bra, 1);
5060: bralink = (oldlinkoffset == 0)? NULL : bralink - oldlinkoffset;
5061: *code++ = OP_KET;
5062: PUTINC(code, 0, offset);
5063: PUT(bra, 1, offset);
5064: }
5065: }
5066:
5067: /* If the maximum is unlimited, set a repeater in the final copy. For
5068: ONCE brackets, that's all we need to do. However, possessively repeated
5069: ONCE brackets can be converted into non-capturing brackets, as the
5070: behaviour of (?:xx)++ is the same as (?>xx)++ and this saves having to
5071: deal with possessive ONCEs specially.
5072:
5073: Otherwise, when we are doing the actual compile phase, check to see
5074: whether this group is one that could match an empty string. If so,
5075: convert the initial operator to the S form (e.g. OP_BRA -> OP_SBRA) so
5076: that runtime checking can be done. [This check is also applied to ONCE
5077: groups at runtime, but in a different way.]
5078:
5079: Then, if the quantifier was possessive and the bracket is not a
5080: conditional, we convert the BRA code to the POS form, and the KET code to
5081: KETRPOS. (It turns out to be convenient at runtime to detect this kind of
5082: subpattern at both the start and at the end.) The use of special opcodes
5083: makes it possible to reduce greatly the stack usage in pcre_exec(). If
5084: the group is preceded by OP_BRAZERO, convert this to OP_BRAPOSZERO.
5085:
5086: Then, if the minimum number of matches is 1 or 0, cancel the possessive
5087: flag so that the default action below, of wrapping everything inside
5088: atomic brackets, does not happen. When the minimum is greater than 1,
5089: there will be earlier copies of the group, and so we still have to wrap
5090: the whole thing. */
5091:
5092: else
5093: {
5094: uschar *ketcode = code - 1 - LINK_SIZE;
5095: uschar *bracode = ketcode - GET(ketcode, 1);
5096:
5097: /* Convert possessive ONCE brackets to non-capturing */
5098:
5099: if ((*bracode == OP_ONCE || *bracode == OP_ONCE_NC) &&
5100: possessive_quantifier) *bracode = OP_BRA;
5101:
5102: /* For non-possessive ONCE brackets, all we need to do is to
5103: set the KET. */
5104:
5105: if (*bracode == OP_ONCE || *bracode == OP_ONCE_NC)
5106: *ketcode = OP_KETRMAX + repeat_type;
5107:
5108: /* Handle non-ONCE brackets and possessive ONCEs (which have been
5109: converted to non-capturing above). */
5110:
5111: else
5112: {
5113: /* In the compile phase, check for empty string matching. */
5114:
5115: if (lengthptr == NULL)
5116: {
5117: uschar *scode = bracode;
5118: do
5119: {
5120: if (could_be_empty_branch(scode, ketcode, utf8, cd))
5121: {
5122: *bracode += OP_SBRA - OP_BRA;
5123: break;
5124: }
5125: scode += GET(scode, 1);
5126: }
5127: while (*scode == OP_ALT);
5128: }
5129:
5130: /* Handle possessive quantifiers. */
5131:
5132: if (possessive_quantifier)
5133: {
5134: /* For COND brackets, we wrap the whole thing in a possessively
5135: repeated non-capturing bracket, because we have not invented POS
5136: versions of the COND opcodes. Because we are moving code along, we
5137: must ensure that any pending recursive references are updated. */
5138:
5139: if (*bracode == OP_COND || *bracode == OP_SCOND)
5140: {
5141: int nlen = (int)(code - bracode);
5142: *code = OP_END;
5143: adjust_recurse(bracode, 1 + LINK_SIZE, utf8, cd, save_hwm);
5144: memmove(bracode + 1+LINK_SIZE, bracode, nlen);
5145: code += 1 + LINK_SIZE;
5146: nlen += 1 + LINK_SIZE;
5147: *bracode = OP_BRAPOS;
5148: *code++ = OP_KETRPOS;
5149: PUTINC(code, 0, nlen);
5150: PUT(bracode, 1, nlen);
5151: }
5152:
5153: /* For non-COND brackets, we modify the BRA code and use KETRPOS. */
5154:
5155: else
5156: {
5157: *bracode += 1; /* Switch to xxxPOS opcodes */
5158: *ketcode = OP_KETRPOS;
5159: }
5160:
5161: /* If the minimum is zero, mark it as possessive, then unset the
5162: possessive flag when the minimum is 0 or 1. */
5163:
5164: if (brazeroptr != NULL) *brazeroptr = OP_BRAPOSZERO;
5165: if (repeat_min < 2) possessive_quantifier = FALSE;
5166: }
5167:
5168: /* Non-possessive quantifier */
5169:
5170: else *ketcode = OP_KETRMAX + repeat_type;
5171: }
5172: }
5173: }
5174:
5175: /* If previous is OP_FAIL, it was generated by an empty class [] in
5176: JavaScript mode. The other ways in which OP_FAIL can be generated, that is
5177: by (*FAIL) or (?!) set previous to NULL, which gives a "nothing to repeat"
5178: error above. We can just ignore the repeat in JS case. */
5179:
5180: else if (*previous == OP_FAIL) goto END_REPEAT;
5181:
5182: /* Else there's some kind of shambles */
5183:
5184: else
5185: {
5186: *errorcodeptr = ERR11;
5187: goto FAILED;
5188: }
5189:
5190: /* If the character following a repeat is '+', or if certain optimization
5191: tests above succeeded, possessive_quantifier is TRUE. For some opcodes,
5192: there are special alternative opcodes for this case. For anything else, we
5193: wrap the entire repeated item inside OP_ONCE brackets. Logically, the '+'
5194: notation is just syntactic sugar, taken from Sun's Java package, but the
5195: special opcodes can optimize it.
5196:
5197: Some (but not all) possessively repeated subpatterns have already been
5198: completely handled in the code just above. For them, possessive_quantifier
5199: is always FALSE at this stage.
5200:
5201: Note that the repeated item starts at tempcode, not at previous, which
5202: might be the first part of a string whose (former) last char we repeated.
5203:
5204: Possessifying an 'exact' quantifier has no effect, so we can ignore it. But
5205: an 'upto' may follow. We skip over an 'exact' item, and then test the
5206: length of what remains before proceeding. */
5207:
5208: if (possessive_quantifier)
5209: {
5210: int len;
5211:
5212: if (*tempcode == OP_TYPEEXACT)
5213: tempcode += _pcre_OP_lengths[*tempcode] +
5214: ((tempcode[3] == OP_PROP || tempcode[3] == OP_NOTPROP)? 2 : 0);
5215:
5216: else if (*tempcode == OP_EXACT || *tempcode == OP_NOTEXACT)
5217: {
5218: tempcode += _pcre_OP_lengths[*tempcode];
5219: #ifdef SUPPORT_UTF8
5220: if (utf8 && tempcode[-1] >= 0xc0)
5221: tempcode += _pcre_utf8_table4[tempcode[-1] & 0x3f];
5222: #endif
5223: }
5224:
5225: len = (int)(code - tempcode);
5226: if (len > 0) switch (*tempcode)
5227: {
5228: case OP_STAR: *tempcode = OP_POSSTAR; break;
5229: case OP_PLUS: *tempcode = OP_POSPLUS; break;
5230: case OP_QUERY: *tempcode = OP_POSQUERY; break;
5231: case OP_UPTO: *tempcode = OP_POSUPTO; break;
5232:
5233: case OP_STARI: *tempcode = OP_POSSTARI; break;
5234: case OP_PLUSI: *tempcode = OP_POSPLUSI; break;
5235: case OP_QUERYI: *tempcode = OP_POSQUERYI; break;
5236: case OP_UPTOI: *tempcode = OP_POSUPTOI; break;
5237:
5238: case OP_NOTSTAR: *tempcode = OP_NOTPOSSTAR; break;
5239: case OP_NOTPLUS: *tempcode = OP_NOTPOSPLUS; break;
5240: case OP_NOTQUERY: *tempcode = OP_NOTPOSQUERY; break;
5241: case OP_NOTUPTO: *tempcode = OP_NOTPOSUPTO; break;
5242:
5243: case OP_NOTSTARI: *tempcode = OP_NOTPOSSTARI; break;
5244: case OP_NOTPLUSI: *tempcode = OP_NOTPOSPLUSI; break;
5245: case OP_NOTQUERYI: *tempcode = OP_NOTPOSQUERYI; break;
5246: case OP_NOTUPTOI: *tempcode = OP_NOTPOSUPTOI; break;
5247:
5248: case OP_TYPESTAR: *tempcode = OP_TYPEPOSSTAR; break;
5249: case OP_TYPEPLUS: *tempcode = OP_TYPEPOSPLUS; break;
5250: case OP_TYPEQUERY: *tempcode = OP_TYPEPOSQUERY; break;
5251: case OP_TYPEUPTO: *tempcode = OP_TYPEPOSUPTO; break;
5252:
5253: /* Because we are moving code along, we must ensure that any
5254: pending recursive references are updated. */
5255:
5256: default:
5257: *code = OP_END;
5258: adjust_recurse(tempcode, 1 + LINK_SIZE, utf8, cd, save_hwm);
5259: memmove(tempcode + 1+LINK_SIZE, tempcode, len);
5260: code += 1 + LINK_SIZE;
5261: len += 1 + LINK_SIZE;
5262: tempcode[0] = OP_ONCE;
5263: *code++ = OP_KET;
5264: PUTINC(code, 0, len);
5265: PUT(tempcode, 1, len);
5266: break;
5267: }
5268: }
5269:
5270: /* In all case we no longer have a previous item. We also set the
5271: "follows varying string" flag for subsequently encountered reqbytes if
5272: it isn't already set and we have just passed a varying length item. */
5273:
5274: END_REPEAT:
5275: previous = NULL;
5276: cd->req_varyopt |= reqvary;
5277: break;
5278:
5279:
5280: /* ===================================================================*/
5281: /* Start of nested parenthesized sub-expression, or comment or lookahead or
5282: lookbehind or option setting or condition or all the other extended
5283: parenthesis forms. */
5284:
5285: case CHAR_LEFT_PARENTHESIS:
5286: newoptions = options;
5287: skipbytes = 0;
5288: bravalue = OP_CBRA;
5289: save_hwm = cd->hwm;
5290: reset_bracount = FALSE;
5291:
5292: /* First deal with various "verbs" that can be introduced by '*'. */
5293:
5294: if (*(++ptr) == CHAR_ASTERISK &&
5295: ((cd->ctypes[ptr[1]] & ctype_letter) != 0 || ptr[1] == ':'))
5296: {
5297: int i, namelen;
5298: int arglen = 0;
5299: const char *vn = verbnames;
5300: const uschar *name = ptr + 1;
5301: const uschar *arg = NULL;
5302: previous = NULL;
5303: while ((cd->ctypes[*++ptr] & ctype_letter) != 0) {};
5304: namelen = (int)(ptr - name);
5305:
5306: /* It appears that Perl allows any characters whatsoever, other than
5307: a closing parenthesis, to appear in arguments, so we no longer insist on
5308: letters, digits, and underscores. */
5309:
5310: if (*ptr == CHAR_COLON)
5311: {
5312: arg = ++ptr;
5313: while (*ptr != 0 && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++;
5314: arglen = (int)(ptr - arg);
5315: }
5316:
5317: if (*ptr != CHAR_RIGHT_PARENTHESIS)
5318: {
5319: *errorcodeptr = ERR60;
5320: goto FAILED;
5321: }
5322:
5323: /* Scan the table of verb names */
5324:
5325: for (i = 0; i < verbcount; i++)
5326: {
5327: if (namelen == verbs[i].len &&
5328: strncmp((char *)name, vn, namelen) == 0)
5329: {
5330: /* Check for open captures before ACCEPT and convert it to
5331: ASSERT_ACCEPT if in an assertion. */
5332:
5333: if (verbs[i].op == OP_ACCEPT)
5334: {
5335: open_capitem *oc;
5336: if (arglen != 0)
5337: {
5338: *errorcodeptr = ERR59;
5339: goto FAILED;
5340: }
5341: cd->had_accept = TRUE;
5342: for (oc = cd->open_caps; oc != NULL; oc = oc->next)
5343: {
5344: *code++ = OP_CLOSE;
5345: PUT2INC(code, 0, oc->number);
5346: }
5347: *code++ = (cd->assert_depth > 0)? OP_ASSERT_ACCEPT : OP_ACCEPT;
5348:
5349: /* Do not set firstbyte after *ACCEPT */
5350: if (firstbyte == REQ_UNSET) firstbyte = REQ_NONE;
5351: }
5352:
5353: /* Handle other cases with/without an argument */
5354:
5355: else if (arglen == 0)
5356: {
5357: if (verbs[i].op < 0) /* Argument is mandatory */
5358: {
5359: *errorcodeptr = ERR66;
5360: goto FAILED;
5361: }
5362: *code = verbs[i].op;
5363: if (*code++ == OP_THEN) cd->external_flags |= PCRE_HASTHEN;
5364: }
5365:
5366: else
5367: {
5368: if (verbs[i].op_arg < 0) /* Argument is forbidden */
5369: {
5370: *errorcodeptr = ERR59;
5371: goto FAILED;
5372: }
5373: *code = verbs[i].op_arg;
5374: if (*code++ == OP_THEN_ARG) cd->external_flags |= PCRE_HASTHEN;
5375: *code++ = arglen;
5376: memcpy(code, arg, arglen);
5377: code += arglen;
5378: *code++ = 0;
5379: }
5380:
5381: break; /* Found verb, exit loop */
5382: }
5383:
5384: vn += verbs[i].len + 1;
5385: }
5386:
5387: if (i < verbcount) continue; /* Successfully handled a verb */
5388: *errorcodeptr = ERR60; /* Verb not recognized */
5389: goto FAILED;
5390: }
5391:
5392: /* Deal with the extended parentheses; all are introduced by '?', and the
5393: appearance of any of them means that this is not a capturing group. */
5394:
5395: else if (*ptr == CHAR_QUESTION_MARK)
5396: {
5397: int i, set, unset, namelen;
5398: int *optset;
5399: const uschar *name;
5400: uschar *slot;
5401:
5402: switch (*(++ptr))
5403: {
5404: case CHAR_NUMBER_SIGN: /* Comment; skip to ket */
5405: ptr++;
5406: while (*ptr != 0 && *ptr != CHAR_RIGHT_PARENTHESIS) ptr++;
5407: if (*ptr == 0)
5408: {
5409: *errorcodeptr = ERR18;
5410: goto FAILED;
5411: }
5412: continue;
5413:
5414:
5415: /* ------------------------------------------------------------ */
5416: case CHAR_VERTICAL_LINE: /* Reset capture count for each branch */
5417: reset_bracount = TRUE;
5418: /* Fall through */
5419:
5420: /* ------------------------------------------------------------ */
5421: case CHAR_COLON: /* Non-capturing bracket */
5422: bravalue = OP_BRA;
5423: ptr++;
5424: break;
5425:
5426:
5427: /* ------------------------------------------------------------ */
5428: case CHAR_LEFT_PARENTHESIS:
5429: bravalue = OP_COND; /* Conditional group */
5430:
5431: /* A condition can be an assertion, a number (referring to a numbered
5432: group), a name (referring to a named group), or 'R', referring to
5433: recursion. R<digits> and R&name are also permitted for recursion tests.
5434:
5435: There are several syntaxes for testing a named group: (?(name)) is used
5436: by Python; Perl 5.10 onwards uses (?(<name>) or (?('name')).
5437:
5438: There are two unfortunate ambiguities, caused by history. (a) 'R' can
5439: be the recursive thing or the name 'R' (and similarly for 'R' followed
5440: by digits), and (b) a number could be a name that consists of digits.
5441: In both cases, we look for a name first; if not found, we try the other
5442: cases. */
5443:
5444: /* For conditions that are assertions, check the syntax, and then exit
5445: the switch. This will take control down to where bracketed groups,
5446: including assertions, are processed. */
5447:
5448: if (ptr[1] == CHAR_QUESTION_MARK && (ptr[2] == CHAR_EQUALS_SIGN ||
5449: ptr[2] == CHAR_EXCLAMATION_MARK || ptr[2] == CHAR_LESS_THAN_SIGN))
5450: break;
5451:
5452: /* Most other conditions use OP_CREF (a couple change to OP_RREF
5453: below), and all need to skip 3 bytes at the start of the group. */
5454:
5455: code[1+LINK_SIZE] = OP_CREF;
5456: skipbytes = 3;
5457: refsign = -1;
5458:
5459: /* Check for a test for recursion in a named group. */
5460:
5461: if (ptr[1] == CHAR_R && ptr[2] == CHAR_AMPERSAND)
5462: {
5463: terminator = -1;
5464: ptr += 2;
5465: code[1+LINK_SIZE] = OP_RREF; /* Change the type of test */
5466: }
5467:
5468: /* Check for a test for a named group's having been set, using the Perl
5469: syntax (?(<name>) or (?('name') */
5470:
5471: else if (ptr[1] == CHAR_LESS_THAN_SIGN)
5472: {
5473: terminator = CHAR_GREATER_THAN_SIGN;
5474: ptr++;
5475: }
5476: else if (ptr[1] == CHAR_APOSTROPHE)
5477: {
5478: terminator = CHAR_APOSTROPHE;
5479: ptr++;
5480: }
5481: else
5482: {
5483: terminator = 0;
5484: if (ptr[1] == CHAR_MINUS || ptr[1] == CHAR_PLUS) refsign = *(++ptr);
5485: }
5486:
5487: /* We now expect to read a name; any thing else is an error */
5488:
5489: if ((cd->ctypes[ptr[1]] & ctype_word) == 0)
5490: {
5491: ptr += 1; /* To get the right offset */
5492: *errorcodeptr = ERR28;
5493: goto FAILED;
5494: }
5495:
5496: /* Read the name, but also get it as a number if it's all digits */
5497:
5498: recno = 0;
5499: name = ++ptr;
5500: while ((cd->ctypes[*ptr] & ctype_word) != 0)
5501: {
5502: if (recno >= 0)
5503: recno = ((digitab[*ptr] & ctype_digit) != 0)?
5504: recno * 10 + *ptr - CHAR_0 : -1;
5505: ptr++;
5506: }
5507: namelen = (int)(ptr - name);
5508:
5509: if ((terminator > 0 && *ptr++ != terminator) ||
5510: *ptr++ != CHAR_RIGHT_PARENTHESIS)
5511: {
5512: ptr--; /* Error offset */
5513: *errorcodeptr = ERR26;
5514: goto FAILED;
5515: }
5516:
5517: /* Do no further checking in the pre-compile phase. */
5518:
5519: if (lengthptr != NULL) break;
5520:
5521: /* In the real compile we do the work of looking for the actual
5522: reference. If the string started with "+" or "-" we require the rest to
5523: be digits, in which case recno will be set. */
5524:
5525: if (refsign > 0)
5526: {
5527: if (recno <= 0)
5528: {
5529: *errorcodeptr = ERR58;
5530: goto FAILED;
5531: }
5532: recno = (refsign == CHAR_MINUS)?
5533: cd->bracount - recno + 1 : recno +cd->bracount;
5534: if (recno <= 0 || recno > cd->final_bracount)
5535: {
5536: *errorcodeptr = ERR15;
5537: goto FAILED;
5538: }
5539: PUT2(code, 2+LINK_SIZE, recno);
5540: break;
5541: }
5542:
5543: /* Otherwise (did not start with "+" or "-"), start by looking for the
5544: name. If we find a name, add one to the opcode to change OP_CREF or
5545: OP_RREF into OP_NCREF or OP_NRREF. These behave exactly the same,
5546: except they record that the reference was originally to a name. The
5547: information is used to check duplicate names. */
5548:
5549: slot = cd->name_table;
5550: for (i = 0; i < cd->names_found; i++)
5551: {
5552: if (strncmp((char *)name, (char *)slot+2, namelen) == 0) break;
5553: slot += cd->name_entry_size;
5554: }
5555:
5556: /* Found a previous named subpattern */
5557:
5558: if (i < cd->names_found)
5559: {
5560: recno = GET2(slot, 0);
5561: PUT2(code, 2+LINK_SIZE, recno);
5562: code[1+LINK_SIZE]++;
5563: }
5564:
5565: /* Search the pattern for a forward reference */
5566:
5567: else if ((i = find_parens(cd, name, namelen,
5568: (options & PCRE_EXTENDED) != 0, utf8)) > 0)
5569: {
5570: PUT2(code, 2+LINK_SIZE, i);
5571: code[1+LINK_SIZE]++;
5572: }
5573:
5574: /* If terminator == 0 it means that the name followed directly after
5575: the opening parenthesis [e.g. (?(abc)...] and in this case there are
5576: some further alternatives to try. For the cases where terminator != 0
5577: [things like (?(<name>... or (?('name')... or (?(R&name)... ] we have
5578: now checked all the possibilities, so give an error. */
5579:
5580: else if (terminator != 0)
5581: {
5582: *errorcodeptr = ERR15;
5583: goto FAILED;
5584: }
5585:
5586: /* Check for (?(R) for recursion. Allow digits after R to specify a
5587: specific group number. */
5588:
5589: else if (*name == CHAR_R)
5590: {
5591: recno = 0;
5592: for (i = 1; i < namelen; i++)
5593: {
5594: if ((digitab[name[i]] & ctype_digit) == 0)
5595: {
5596: *errorcodeptr = ERR15;
5597: goto FAILED;
5598: }
5599: recno = recno * 10 + name[i] - CHAR_0;
5600: }
5601: if (recno == 0) recno = RREF_ANY;
5602: code[1+LINK_SIZE] = OP_RREF; /* Change test type */
5603: PUT2(code, 2+LINK_SIZE, recno);
5604: }
5605:
5606: /* Similarly, check for the (?(DEFINE) "condition", which is always
5607: false. */
5608:
5609: else if (namelen == 6 && strncmp((char *)name, STRING_DEFINE, 6) == 0)
5610: {
5611: code[1+LINK_SIZE] = OP_DEF;
5612: skipbytes = 1;
5613: }
5614:
5615: /* Check for the "name" actually being a subpattern number. We are
5616: in the second pass here, so final_bracount is set. */
5617:
5618: else if (recno > 0 && recno <= cd->final_bracount)
5619: {
5620: PUT2(code, 2+LINK_SIZE, recno);
5621: }
5622:
5623: /* Either an unidentified subpattern, or a reference to (?(0) */
5624:
5625: else
5626: {
5627: *errorcodeptr = (recno == 0)? ERR35: ERR15;
5628: goto FAILED;
5629: }
5630: break;
5631:
5632:
5633: /* ------------------------------------------------------------ */
5634: case CHAR_EQUALS_SIGN: /* Positive lookahead */
5635: bravalue = OP_ASSERT;
5636: cd->assert_depth += 1;
5637: ptr++;
5638: break;
5639:
5640:
5641: /* ------------------------------------------------------------ */
5642: case CHAR_EXCLAMATION_MARK: /* Negative lookahead */
5643: ptr++;
5644: if (*ptr == CHAR_RIGHT_PARENTHESIS) /* Optimize (?!) */
5645: {
5646: *code++ = OP_FAIL;
5647: previous = NULL;
5648: continue;
5649: }
5650: bravalue = OP_ASSERT_NOT;
5651: cd->assert_depth += 1;
5652: break;
5653:
5654:
5655: /* ------------------------------------------------------------ */
5656: case CHAR_LESS_THAN_SIGN: /* Lookbehind or named define */
5657: switch (ptr[1])
5658: {
5659: case CHAR_EQUALS_SIGN: /* Positive lookbehind */
5660: bravalue = OP_ASSERTBACK;
5661: cd->assert_depth += 1;
5662: ptr += 2;
5663: break;
5664:
5665: case CHAR_EXCLAMATION_MARK: /* Negative lookbehind */
5666: bravalue = OP_ASSERTBACK_NOT;
5667: cd->assert_depth += 1;
5668: ptr += 2;
5669: break;
5670:
5671: default: /* Could be name define, else bad */
5672: if ((cd->ctypes[ptr[1]] & ctype_word) != 0) goto DEFINE_NAME;
5673: ptr++; /* Correct offset for error */
5674: *errorcodeptr = ERR24;
5675: goto FAILED;
5676: }
5677: break;
5678:
5679:
5680: /* ------------------------------------------------------------ */
5681: case CHAR_GREATER_THAN_SIGN: /* One-time brackets */
5682: bravalue = OP_ONCE;
5683: ptr++;
5684: break;
5685:
5686:
5687: /* ------------------------------------------------------------ */
5688: case CHAR_C: /* Callout - may be followed by digits; */
5689: previous_callout = code; /* Save for later completion */
5690: after_manual_callout = 1; /* Skip one item before completing */
5691: *code++ = OP_CALLOUT;
5692: {
5693: int n = 0;
5694: while ((digitab[*(++ptr)] & ctype_digit) != 0)
5695: n = n * 10 + *ptr - CHAR_0;
5696: if (*ptr != CHAR_RIGHT_PARENTHESIS)
5697: {
5698: *errorcodeptr = ERR39;
5699: goto FAILED;
5700: }
5701: if (n > 255)
5702: {
5703: *errorcodeptr = ERR38;
5704: goto FAILED;
5705: }
5706: *code++ = n;
5707: PUT(code, 0, (int)(ptr - cd->start_pattern + 1)); /* Pattern offset */
5708: PUT(code, LINK_SIZE, 0); /* Default length */
5709: code += 2 * LINK_SIZE;
5710: }
5711: previous = NULL;
5712: continue;
5713:
5714:
5715: /* ------------------------------------------------------------ */
5716: case CHAR_P: /* Python-style named subpattern handling */
5717: if (*(++ptr) == CHAR_EQUALS_SIGN ||
5718: *ptr == CHAR_GREATER_THAN_SIGN) /* Reference or recursion */
5719: {
5720: is_recurse = *ptr == CHAR_GREATER_THAN_SIGN;
5721: terminator = CHAR_RIGHT_PARENTHESIS;
5722: goto NAMED_REF_OR_RECURSE;
5723: }
5724: else if (*ptr != CHAR_LESS_THAN_SIGN) /* Test for Python-style defn */
5725: {
5726: *errorcodeptr = ERR41;
5727: goto FAILED;
5728: }
5729: /* Fall through to handle (?P< as (?< is handled */
5730:
5731:
5732: /* ------------------------------------------------------------ */
5733: DEFINE_NAME: /* Come here from (?< handling */
5734: case CHAR_APOSTROPHE:
5735: {
5736: terminator = (*ptr == CHAR_LESS_THAN_SIGN)?
5737: CHAR_GREATER_THAN_SIGN : CHAR_APOSTROPHE;
5738: name = ++ptr;
5739:
5740: while ((cd->ctypes[*ptr] & ctype_word) != 0) ptr++;
5741: namelen = (int)(ptr - name);
5742:
5743: /* In the pre-compile phase, just do a syntax check. */
5744:
5745: if (lengthptr != NULL)
5746: {
5747: if (*ptr != terminator)
5748: {
5749: *errorcodeptr = ERR42;
5750: goto FAILED;
5751: }
5752: if (cd->names_found >= MAX_NAME_COUNT)
5753: {
5754: *errorcodeptr = ERR49;
5755: goto FAILED;
5756: }
5757: if (namelen + 3 > cd->name_entry_size)
5758: {
5759: cd->name_entry_size = namelen + 3;
5760: if (namelen > MAX_NAME_SIZE)
5761: {
5762: *errorcodeptr = ERR48;
5763: goto FAILED;
5764: }
5765: }
5766: }
5767:
5768: /* In the real compile, create the entry in the table, maintaining
5769: alphabetical order. Duplicate names for different numbers are
5770: permitted only if PCRE_DUPNAMES is set. Duplicate names for the same
5771: number are always OK. (An existing number can be re-used if (?|
5772: appears in the pattern.) In either event, a duplicate name results in
5773: a duplicate entry in the table, even if the number is the same. This
5774: is because the number of names, and hence the table size, is computed
5775: in the pre-compile, and it affects various numbers and pointers which
5776: would all have to be modified, and the compiled code moved down, if
5777: duplicates with the same number were omitted from the table. This
5778: doesn't seem worth the hassle. However, *different* names for the
5779: same number are not permitted. */
5780:
5781: else
5782: {
5783: BOOL dupname = FALSE;
5784: slot = cd->name_table;
5785:
5786: for (i = 0; i < cd->names_found; i++)
5787: {
5788: int crc = memcmp(name, slot+2, namelen);
5789: if (crc == 0)
5790: {
5791: if (slot[2+namelen] == 0)
5792: {
5793: if (GET2(slot, 0) != cd->bracount + 1 &&
5794: (options & PCRE_DUPNAMES) == 0)
5795: {
5796: *errorcodeptr = ERR43;
5797: goto FAILED;
5798: }
5799: else dupname = TRUE;
5800: }
5801: else crc = -1; /* Current name is a substring */
5802: }
5803:
5804: /* Make space in the table and break the loop for an earlier
5805: name. For a duplicate or later name, carry on. We do this for
5806: duplicates so that in the simple case (when ?(| is not used) they
5807: are in order of their numbers. */
5808:
5809: if (crc < 0)
5810: {
5811: memmove(slot + cd->name_entry_size, slot,
5812: (cd->names_found - i) * cd->name_entry_size);
5813: break;
5814: }
5815:
5816: /* Continue the loop for a later or duplicate name */
5817:
5818: slot += cd->name_entry_size;
5819: }
5820:
5821: /* For non-duplicate names, check for a duplicate number before
5822: adding the new name. */
5823:
5824: if (!dupname)
5825: {
5826: uschar *cslot = cd->name_table;
5827: for (i = 0; i < cd->names_found; i++)
5828: {
5829: if (cslot != slot)
5830: {
5831: if (GET2(cslot, 0) == cd->bracount + 1)
5832: {
5833: *errorcodeptr = ERR65;
5834: goto FAILED;
5835: }
5836: }
5837: else i--;
5838: cslot += cd->name_entry_size;
5839: }
5840: }
5841:
5842: PUT2(slot, 0, cd->bracount + 1);
5843: memcpy(slot + 2, name, namelen);
5844: slot[2+namelen] = 0;
5845: }
5846: }
5847:
5848: /* In both pre-compile and compile, count the number of names we've
5849: encountered. */
5850:
5851: cd->names_found++;
5852: ptr++; /* Move past > or ' */
5853: goto NUMBERED_GROUP;
5854:
5855:
5856: /* ------------------------------------------------------------ */
5857: case CHAR_AMPERSAND: /* Perl recursion/subroutine syntax */
5858: terminator = CHAR_RIGHT_PARENTHESIS;
5859: is_recurse = TRUE;
5860: /* Fall through */
5861:
5862: /* We come here from the Python syntax above that handles both
5863: references (?P=name) and recursion (?P>name), as well as falling
5864: through from the Perl recursion syntax (?&name). We also come here from
5865: the Perl \k<name> or \k'name' back reference syntax and the \k{name}
5866: .NET syntax, and the Oniguruma \g<...> and \g'...' subroutine syntax. */
5867:
5868: NAMED_REF_OR_RECURSE:
5869: name = ++ptr;
5870: while ((cd->ctypes[*ptr] & ctype_word) != 0) ptr++;
5871: namelen = (int)(ptr - name);
5872:
5873: /* In the pre-compile phase, do a syntax check. We used to just set
5874: a dummy reference number, because it was not used in the first pass.
5875: However, with the change of recursive back references to be atomic,
5876: we have to look for the number so that this state can be identified, as
5877: otherwise the incorrect length is computed. If it's not a backwards
5878: reference, the dummy number will do. */
5879:
5880: if (lengthptr != NULL)
5881: {
5882: const uschar *temp;
5883:
5884: if (namelen == 0)
5885: {
5886: *errorcodeptr = ERR62;
5887: goto FAILED;
5888: }
5889: if (*ptr != terminator)
5890: {
5891: *errorcodeptr = ERR42;
5892: goto FAILED;
5893: }
5894: if (namelen > MAX_NAME_SIZE)
5895: {
5896: *errorcodeptr = ERR48;
5897: goto FAILED;
5898: }
5899:
5900: /* The name table does not exist in the first pass, so we cannot
5901: do a simple search as in the code below. Instead, we have to scan the
5902: pattern to find the number. It is important that we scan it only as
5903: far as we have got because the syntax of named subpatterns has not
5904: been checked for the rest of the pattern, and find_parens() assumes
5905: correct syntax. In any case, it's a waste of resources to scan
5906: further. We stop the scan at the current point by temporarily
5907: adjusting the value of cd->endpattern. */
5908:
5909: temp = cd->end_pattern;
5910: cd->end_pattern = ptr;
5911: recno = find_parens(cd, name, namelen,
5912: (options & PCRE_EXTENDED) != 0, utf8);
5913: cd->end_pattern = temp;
5914: if (recno < 0) recno = 0; /* Forward ref; set dummy number */
5915: }
5916:
5917: /* In the real compile, seek the name in the table. We check the name
5918: first, and then check that we have reached the end of the name in the
5919: table. That way, if the name that is longer than any in the table,
5920: the comparison will fail without reading beyond the table entry. */
5921:
5922: else
5923: {
5924: slot = cd->name_table;
5925: for (i = 0; i < cd->names_found; i++)
5926: {
5927: if (strncmp((char *)name, (char *)slot+2, namelen) == 0 &&
5928: slot[2+namelen] == 0)
5929: break;
5930: slot += cd->name_entry_size;
5931: }
5932:
5933: if (i < cd->names_found) /* Back reference */
5934: {
5935: recno = GET2(slot, 0);
5936: }
5937: else if ((recno = /* Forward back reference */
5938: find_parens(cd, name, namelen,
5939: (options & PCRE_EXTENDED) != 0, utf8)) <= 0)
5940: {
5941: *errorcodeptr = ERR15;
5942: goto FAILED;
5943: }
5944: }
5945:
5946: /* In both phases, we can now go to the code than handles numerical
5947: recursion or backreferences. */
5948:
5949: if (is_recurse) goto HANDLE_RECURSION;
5950: else goto HANDLE_REFERENCE;
5951:
5952:
5953: /* ------------------------------------------------------------ */
5954: case CHAR_R: /* Recursion */
5955: ptr++; /* Same as (?0) */
5956: /* Fall through */
5957:
5958:
5959: /* ------------------------------------------------------------ */
5960: case CHAR_MINUS: case CHAR_PLUS: /* Recursion or subroutine */
5961: case CHAR_0: case CHAR_1: case CHAR_2: case CHAR_3: case CHAR_4:
5962: case CHAR_5: case CHAR_6: case CHAR_7: case CHAR_8: case CHAR_9:
5963: {
5964: const uschar *called;
5965: terminator = CHAR_RIGHT_PARENTHESIS;
5966:
5967: /* Come here from the \g<...> and \g'...' code (Oniguruma
5968: compatibility). However, the syntax has been checked to ensure that
5969: the ... are a (signed) number, so that neither ERR63 nor ERR29 will
5970: be called on this path, nor with the jump to OTHER_CHAR_AFTER_QUERY
5971: ever be taken. */
5972:
5973: HANDLE_NUMERICAL_RECURSION:
5974:
5975: if ((refsign = *ptr) == CHAR_PLUS)
5976: {
5977: ptr++;
5978: if ((digitab[*ptr] & ctype_digit) == 0)
5979: {
5980: *errorcodeptr = ERR63;
5981: goto FAILED;
5982: }
5983: }
5984: else if (refsign == CHAR_MINUS)
5985: {
5986: if ((digitab[ptr[1]] & ctype_digit) == 0)
5987: goto OTHER_CHAR_AFTER_QUERY;
5988: ptr++;
5989: }
5990:
5991: recno = 0;
5992: while((digitab[*ptr] & ctype_digit) != 0)
5993: recno = recno * 10 + *ptr++ - CHAR_0;
5994:
5995: if (*ptr != terminator)
5996: {
5997: *errorcodeptr = ERR29;
5998: goto FAILED;
5999: }
6000:
6001: if (refsign == CHAR_MINUS)
6002: {
6003: if (recno == 0)
6004: {
6005: *errorcodeptr = ERR58;
6006: goto FAILED;
6007: }
6008: recno = cd->bracount - recno + 1;
6009: if (recno <= 0)
6010: {
6011: *errorcodeptr = ERR15;
6012: goto FAILED;
6013: }
6014: }
6015: else if (refsign == CHAR_PLUS)
6016: {
6017: if (recno == 0)
6018: {
6019: *errorcodeptr = ERR58;
6020: goto FAILED;
6021: }
6022: recno += cd->bracount;
6023: }
6024:
6025: /* Come here from code above that handles a named recursion */
6026:
6027: HANDLE_RECURSION:
6028:
6029: previous = code;
6030: called = cd->start_code;
6031:
6032: /* When we are actually compiling, find the bracket that is being
6033: referenced. Temporarily end the regex in case it doesn't exist before
6034: this point. If we end up with a forward reference, first check that
6035: the bracket does occur later so we can give the error (and position)
6036: now. Then remember this forward reference in the workspace so it can
6037: be filled in at the end. */
6038:
6039: if (lengthptr == NULL)
6040: {
6041: *code = OP_END;
6042: if (recno != 0)
6043: called = _pcre_find_bracket(cd->start_code, utf8, recno);
6044:
6045: /* Forward reference */
6046:
6047: if (called == NULL)
6048: {
6049: if (find_parens(cd, NULL, recno,
6050: (options & PCRE_EXTENDED) != 0, utf8) < 0)
6051: {
6052: *errorcodeptr = ERR15;
6053: goto FAILED;
6054: }
6055:
6056: /* Fudge the value of "called" so that when it is inserted as an
6057: offset below, what it actually inserted is the reference number
6058: of the group. Then remember the forward reference. */
6059:
6060: called = cd->start_code + recno;
6061: if (cd->hwm >= cd->start_workspace + cd->workspace_size -
6062: WORK_SIZE_SAFETY_MARGIN)
6063: {
6064: *errorcodeptr = expand_workspace(cd);
6065: if (*errorcodeptr != 0) goto FAILED;
6066: }
6067: PUTINC(cd->hwm, 0, (int)(code + 1 - cd->start_code));
6068: }
6069:
6070: /* If not a forward reference, and the subpattern is still open,
6071: this is a recursive call. We check to see if this is a left
6072: recursion that could loop for ever, and diagnose that case. We
6073: must not, however, do this check if we are in a conditional
6074: subpattern because the condition might be testing for recursion in
6075: a pattern such as /(?(R)a+|(?R)b)/, which is perfectly valid.
6076: Forever loops are also detected at runtime, so those that occur in
6077: conditional subpatterns will be picked up then. */
6078:
6079: else if (GET(called, 1) == 0 && cond_depth <= 0 &&
6080: could_be_empty(called, code, bcptr, utf8, cd))
6081: {
6082: *errorcodeptr = ERR40;
6083: goto FAILED;
6084: }
6085: }
6086:
6087: /* Insert the recursion/subroutine item. It does not have a set first
6088: byte (relevant if it is repeated, because it will then be wrapped
6089: with ONCE brackets). */
6090:
6091: *code = OP_RECURSE;
6092: PUT(code, 1, (int)(called - cd->start_code));
6093: code += 1 + LINK_SIZE;
6094: groupsetfirstbyte = FALSE;
6095: }
6096:
6097: /* Can't determine a first byte now */
6098:
6099: if (firstbyte == REQ_UNSET) firstbyte = REQ_NONE;
6100: continue;
6101:
6102:
6103: /* ------------------------------------------------------------ */
6104: default: /* Other characters: check option setting */
6105: OTHER_CHAR_AFTER_QUERY:
6106: set = unset = 0;
6107: optset = &set;
6108:
6109: while (*ptr != CHAR_RIGHT_PARENTHESIS && *ptr != CHAR_COLON)
6110: {
6111: switch (*ptr++)
6112: {
6113: case CHAR_MINUS: optset = &unset; break;
6114:
6115: case CHAR_J: /* Record that it changed in the external options */
6116: *optset |= PCRE_DUPNAMES;
6117: cd->external_flags |= PCRE_JCHANGED;
6118: break;
6119:
6120: case CHAR_i: *optset |= PCRE_CASELESS; break;
6121: case CHAR_m: *optset |= PCRE_MULTILINE; break;
6122: case CHAR_s: *optset |= PCRE_DOTALL; break;
6123: case CHAR_x: *optset |= PCRE_EXTENDED; break;
6124: case CHAR_U: *optset |= PCRE_UNGREEDY; break;
6125: case CHAR_X: *optset |= PCRE_EXTRA; break;
6126:
6127: default: *errorcodeptr = ERR12;
6128: ptr--; /* Correct the offset */
6129: goto FAILED;
6130: }
6131: }
6132:
6133: /* Set up the changed option bits, but don't change anything yet. */
6134:
6135: newoptions = (options | set) & (~unset);
6136:
6137: /* If the options ended with ')' this is not the start of a nested
6138: group with option changes, so the options change at this level. If this
6139: item is right at the start of the pattern, the options can be
6140: abstracted and made external in the pre-compile phase, and ignored in
6141: the compile phase. This can be helpful when matching -- for instance in
6142: caseless checking of required bytes.
6143:
6144: If the code pointer is not (cd->start_code + 1 + LINK_SIZE), we are
6145: definitely *not* at the start of the pattern because something has been
6146: compiled. In the pre-compile phase, however, the code pointer can have
6147: that value after the start, because it gets reset as code is discarded
6148: during the pre-compile. However, this can happen only at top level - if
6149: we are within parentheses, the starting BRA will still be present. At
6150: any parenthesis level, the length value can be used to test if anything
6151: has been compiled at that level. Thus, a test for both these conditions
6152: is necessary to ensure we correctly detect the start of the pattern in
6153: both phases.
6154:
6155: If we are not at the pattern start, reset the greedy defaults and the
6156: case value for firstbyte and reqbyte. */
6157:
6158: if (*ptr == CHAR_RIGHT_PARENTHESIS)
6159: {
6160: if (code == cd->start_code + 1 + LINK_SIZE &&
6161: (lengthptr == NULL || *lengthptr == 2 + 2*LINK_SIZE))
6162: {
6163: cd->external_options = newoptions;
6164: }
6165: else
6166: {
6167: greedy_default = ((newoptions & PCRE_UNGREEDY) != 0);
6168: greedy_non_default = greedy_default ^ 1;
6169: req_caseopt = ((newoptions & PCRE_CASELESS) != 0)? REQ_CASELESS : 0;
6170: }
6171:
6172: /* Change options at this level, and pass them back for use
6173: in subsequent branches. */
6174:
6175: *optionsptr = options = newoptions;
6176: previous = NULL; /* This item can't be repeated */
6177: continue; /* It is complete */
6178: }
6179:
6180: /* If the options ended with ':' we are heading into a nested group
6181: with possible change of options. Such groups are non-capturing and are
6182: not assertions of any kind. All we need to do is skip over the ':';
6183: the newoptions value is handled below. */
6184:
6185: bravalue = OP_BRA;
6186: ptr++;
6187: } /* End of switch for character following (? */
6188: } /* End of (? handling */
6189:
6190: /* Opening parenthesis not followed by '*' or '?'. If PCRE_NO_AUTO_CAPTURE
6191: is set, all unadorned brackets become non-capturing and behave like (?:...)
6192: brackets. */
6193:
6194: else if ((options & PCRE_NO_AUTO_CAPTURE) != 0)
6195: {
6196: bravalue = OP_BRA;
6197: }
6198:
6199: /* Else we have a capturing group. */
6200:
6201: else
6202: {
6203: NUMBERED_GROUP:
6204: cd->bracount += 1;
6205: PUT2(code, 1+LINK_SIZE, cd->bracount);
6206: skipbytes = 2;
6207: }
6208:
6209: /* Process nested bracketed regex. Assertions used not to be repeatable,
6210: but this was changed for Perl compatibility, so all kinds can now be
6211: repeated. We copy code into a non-register variable (tempcode) in order to
6212: be able to pass its address because some compilers complain otherwise. */
6213:
6214: previous = code; /* For handling repetition */
6215: *code = bravalue;
6216: tempcode = code;
6217: tempreqvary = cd->req_varyopt; /* Save value before bracket */
6218: tempbracount = cd->bracount; /* Save value before bracket */
6219: length_prevgroup = 0; /* Initialize for pre-compile phase */
6220:
6221: if (!compile_regex(
6222: newoptions, /* The complete new option state */
6223: &tempcode, /* Where to put code (updated) */
6224: &ptr, /* Input pointer (updated) */
6225: errorcodeptr, /* Where to put an error message */
6226: (bravalue == OP_ASSERTBACK ||
6227: bravalue == OP_ASSERTBACK_NOT), /* TRUE if back assert */
6228: reset_bracount, /* True if (?| group */
6229: skipbytes, /* Skip over bracket number */
6230: cond_depth +
6231: ((bravalue == OP_COND)?1:0), /* Depth of condition subpatterns */
6232: &subfirstbyte, /* For possible first char */
6233: &subreqbyte, /* For possible last char */
6234: bcptr, /* Current branch chain */
6235: cd, /* Tables block */
6236: (lengthptr == NULL)? NULL : /* Actual compile phase */
6237: &length_prevgroup /* Pre-compile phase */
6238: ))
6239: goto FAILED;
6240:
6241: /* If this was an atomic group and there are no capturing groups within it,
6242: generate OP_ONCE_NC instead of OP_ONCE. */
6243:
6244: if (bravalue == OP_ONCE && cd->bracount <= tempbracount)
6245: *code = OP_ONCE_NC;
6246:
6247: if (bravalue >= OP_ASSERT && bravalue <= OP_ASSERTBACK_NOT)
6248: cd->assert_depth -= 1;
6249:
6250: /* At the end of compiling, code is still pointing to the start of the
6251: group, while tempcode has been updated to point past the end of the group.
6252: The pattern pointer (ptr) is on the bracket.
6253:
6254: If this is a conditional bracket, check that there are no more than
6255: two branches in the group, or just one if it's a DEFINE group. We do this
6256: in the real compile phase, not in the pre-pass, where the whole group may
6257: not be available. */
6258:
6259: if (bravalue == OP_COND && lengthptr == NULL)
6260: {
6261: uschar *tc = code;
6262: int condcount = 0;
6263:
6264: do {
6265: condcount++;
6266: tc += GET(tc,1);
6267: }
6268: while (*tc != OP_KET);
6269:
6270: /* A DEFINE group is never obeyed inline (the "condition" is always
6271: false). It must have only one branch. */
6272:
6273: if (code[LINK_SIZE+1] == OP_DEF)
6274: {
6275: if (condcount > 1)
6276: {
6277: *errorcodeptr = ERR54;
6278: goto FAILED;
6279: }
6280: bravalue = OP_DEF; /* Just a flag to suppress char handling below */
6281: }
6282:
6283: /* A "normal" conditional group. If there is just one branch, we must not
6284: make use of its firstbyte or reqbyte, because this is equivalent to an
6285: empty second branch. */
6286:
6287: else
6288: {
6289: if (condcount > 2)
6290: {
6291: *errorcodeptr = ERR27;
6292: goto FAILED;
6293: }
6294: if (condcount == 1) subfirstbyte = subreqbyte = REQ_NONE;
6295: }
6296: }
6297:
6298: /* Error if hit end of pattern */
6299:
6300: if (*ptr != CHAR_RIGHT_PARENTHESIS)
6301: {
6302: *errorcodeptr = ERR14;
6303: goto FAILED;
6304: }
6305:
6306: /* In the pre-compile phase, update the length by the length of the group,
6307: less the brackets at either end. Then reduce the compiled code to just a
6308: set of non-capturing brackets so that it doesn't use much memory if it is
6309: duplicated by a quantifier.*/
6310:
6311: if (lengthptr != NULL)
6312: {
6313: if (OFLOW_MAX - *lengthptr < length_prevgroup - 2 - 2*LINK_SIZE)
6314: {
6315: *errorcodeptr = ERR20;
6316: goto FAILED;
6317: }
6318: *lengthptr += length_prevgroup - 2 - 2*LINK_SIZE;
6319: code++; /* This already contains bravalue */
6320: PUTINC(code, 0, 1 + LINK_SIZE);
6321: *code++ = OP_KET;
6322: PUTINC(code, 0, 1 + LINK_SIZE);
6323: break; /* No need to waste time with special character handling */
6324: }
6325:
6326: /* Otherwise update the main code pointer to the end of the group. */
6327:
6328: code = tempcode;
6329:
6330: /* For a DEFINE group, required and first character settings are not
6331: relevant. */
6332:
6333: if (bravalue == OP_DEF) break;
6334:
6335: /* Handle updating of the required and first characters for other types of
6336: group. Update for normal brackets of all kinds, and conditions with two
6337: branches (see code above). If the bracket is followed by a quantifier with
6338: zero repeat, we have to back off. Hence the definition of zeroreqbyte and
6339: zerofirstbyte outside the main loop so that they can be accessed for the
6340: back off. */
6341:
6342: zeroreqbyte = reqbyte;
6343: zerofirstbyte = firstbyte;
6344: groupsetfirstbyte = FALSE;
6345:
6346: if (bravalue >= OP_ONCE)
6347: {
6348: /* If we have not yet set a firstbyte in this branch, take it from the
6349: subpattern, remembering that it was set here so that a repeat of more
6350: than one can replicate it as reqbyte if necessary. If the subpattern has
6351: no firstbyte, set "none" for the whole branch. In both cases, a zero
6352: repeat forces firstbyte to "none". */
6353:
6354: if (firstbyte == REQ_UNSET)
6355: {
6356: if (subfirstbyte >= 0)
6357: {
6358: firstbyte = subfirstbyte;
6359: groupsetfirstbyte = TRUE;
6360: }
6361: else firstbyte = REQ_NONE;
6362: zerofirstbyte = REQ_NONE;
6363: }
6364:
6365: /* If firstbyte was previously set, convert the subpattern's firstbyte
6366: into reqbyte if there wasn't one, using the vary flag that was in
6367: existence beforehand. */
6368:
6369: else if (subfirstbyte >= 0 && subreqbyte < 0)
6370: subreqbyte = subfirstbyte | tempreqvary;
6371:
6372: /* If the subpattern set a required byte (or set a first byte that isn't
6373: really the first byte - see above), set it. */
6374:
6375: if (subreqbyte >= 0) reqbyte = subreqbyte;
6376: }
6377:
6378: /* For a forward assertion, we take the reqbyte, if set. This can be
6379: helpful if the pattern that follows the assertion doesn't set a different
6380: char. For example, it's useful for /(?=abcde).+/. We can't set firstbyte
6381: for an assertion, however because it leads to incorrect effect for patterns
6382: such as /(?=a)a.+/ when the "real" "a" would then become a reqbyte instead
6383: of a firstbyte. This is overcome by a scan at the end if there's no
6384: firstbyte, looking for an asserted first char. */
6385:
6386: else if (bravalue == OP_ASSERT && subreqbyte >= 0) reqbyte = subreqbyte;
6387: break; /* End of processing '(' */
6388:
6389:
6390: /* ===================================================================*/
6391: /* Handle metasequences introduced by \. For ones like \d, the ESC_ values
6392: are arranged to be the negation of the corresponding OP_values in the
6393: default case when PCRE_UCP is not set. For the back references, the values
6394: are ESC_REF plus the reference number. Only back references and those types
6395: that consume a character may be repeated. We can test for values between
6396: ESC_b and ESC_Z for the latter; this may have to change if any new ones are
6397: ever created. */
6398:
6399: case CHAR_BACKSLASH:
6400: tempptr = ptr;
6401: c = check_escape(&ptr, errorcodeptr, cd->bracount, options, FALSE);
6402: if (*errorcodeptr != 0) goto FAILED;
6403:
6404: if (c < 0)
6405: {
6406: if (-c == ESC_Q) /* Handle start of quoted string */
6407: {
6408: if (ptr[1] == CHAR_BACKSLASH && ptr[2] == CHAR_E)
6409: ptr += 2; /* avoid empty string */
6410: else inescq = TRUE;
6411: continue;
6412: }
6413:
6414: if (-c == ESC_E) continue; /* Perl ignores an orphan \E */
6415:
6416: /* For metasequences that actually match a character, we disable the
6417: setting of a first character if it hasn't already been set. */
6418:
6419: if (firstbyte == REQ_UNSET && -c > ESC_b && -c < ESC_Z)
6420: firstbyte = REQ_NONE;
6421:
6422: /* Set values to reset to if this is followed by a zero repeat. */
6423:
6424: zerofirstbyte = firstbyte;
6425: zeroreqbyte = reqbyte;
6426:
6427: /* \g<name> or \g'name' is a subroutine call by name and \g<n> or \g'n'
6428: is a subroutine call by number (Oniguruma syntax). In fact, the value
6429: -ESC_g is returned only for these cases. So we don't need to check for <
6430: or ' if the value is -ESC_g. For the Perl syntax \g{n} the value is
6431: -ESC_REF+n, and for the Perl syntax \g{name} the result is -ESC_k (as
6432: that is a synonym for a named back reference). */
6433:
6434: if (-c == ESC_g)
6435: {
6436: const uschar *p;
6437: save_hwm = cd->hwm; /* Normally this is set when '(' is read */
6438: terminator = (*(++ptr) == CHAR_LESS_THAN_SIGN)?
6439: CHAR_GREATER_THAN_SIGN : CHAR_APOSTROPHE;
6440:
6441: /* These two statements stop the compiler for warning about possibly
6442: unset variables caused by the jump to HANDLE_NUMERICAL_RECURSION. In
6443: fact, because we actually check for a number below, the paths that
6444: would actually be in error are never taken. */
6445:
6446: skipbytes = 0;
6447: reset_bracount = FALSE;
6448:
6449: /* Test for a name */
6450:
6451: if (ptr[1] != CHAR_PLUS && ptr[1] != CHAR_MINUS)
6452: {
6453: BOOL isnumber = TRUE;
6454: for (p = ptr + 1; *p != 0 && *p != terminator; p++)
6455: {
6456: if ((cd->ctypes[*p] & ctype_digit) == 0) isnumber = FALSE;
6457: if ((cd->ctypes[*p] & ctype_word) == 0) break;
6458: }
6459: if (*p != terminator)
6460: {
6461: *errorcodeptr = ERR57;
6462: break;
6463: }
6464: if (isnumber)
6465: {
6466: ptr++;
6467: goto HANDLE_NUMERICAL_RECURSION;
6468: }
6469: is_recurse = TRUE;
6470: goto NAMED_REF_OR_RECURSE;
6471: }
6472:
6473: /* Test a signed number in angle brackets or quotes. */
6474:
6475: p = ptr + 2;
6476: while ((digitab[*p] & ctype_digit) != 0) p++;
6477: if (*p != terminator)
6478: {
6479: *errorcodeptr = ERR57;
6480: break;
6481: }
6482: ptr++;
6483: goto HANDLE_NUMERICAL_RECURSION;
6484: }
6485:
6486: /* \k<name> or \k'name' is a back reference by name (Perl syntax).
6487: We also support \k{name} (.NET syntax). */
6488:
6489: if (-c == ESC_k)
6490: {
6491: if ((ptr[1] != CHAR_LESS_THAN_SIGN &&
6492: ptr[1] != CHAR_APOSTROPHE && ptr[1] != CHAR_LEFT_CURLY_BRACKET))
6493: {
6494: *errorcodeptr = ERR69;
6495: break;
6496: }
6497: is_recurse = FALSE;
6498: terminator = (*(++ptr) == CHAR_LESS_THAN_SIGN)?
6499: CHAR_GREATER_THAN_SIGN : (*ptr == CHAR_APOSTROPHE)?
6500: CHAR_APOSTROPHE : CHAR_RIGHT_CURLY_BRACKET;
6501: goto NAMED_REF_OR_RECURSE;
6502: }
6503:
6504: /* Back references are handled specially; must disable firstbyte if
6505: not set to cope with cases like (?=(\w+))\1: which would otherwise set
6506: ':' later. */
6507:
6508: if (-c >= ESC_REF)
6509: {
6510: open_capitem *oc;
6511: recno = -c - ESC_REF;
6512:
6513: HANDLE_REFERENCE: /* Come here from named backref handling */
6514: if (firstbyte == REQ_UNSET) firstbyte = REQ_NONE;
6515: previous = code;
6516: *code++ = ((options & PCRE_CASELESS) != 0)? OP_REFI : OP_REF;
6517: PUT2INC(code, 0, recno);
6518: cd->backref_map |= (recno < 32)? (1 << recno) : 1;
6519: if (recno > cd->top_backref) cd->top_backref = recno;
6520:
6521: /* Check to see if this back reference is recursive, that it, it
6522: is inside the group that it references. A flag is set so that the
6523: group can be made atomic. */
6524:
6525: for (oc = cd->open_caps; oc != NULL; oc = oc->next)
6526: {
6527: if (oc->number == recno)
6528: {
6529: oc->flag = TRUE;
6530: break;
6531: }
6532: }
6533: }
6534:
6535: /* So are Unicode property matches, if supported. */
6536:
6537: #ifdef SUPPORT_UCP
6538: else if (-c == ESC_P || -c == ESC_p)
6539: {
6540: BOOL negated;
6541: int pdata;
6542: int ptype = get_ucp(&ptr, &negated, &pdata, errorcodeptr);
6543: if (ptype < 0) goto FAILED;
6544: previous = code;
6545: *code++ = ((-c == ESC_p) != negated)? OP_PROP : OP_NOTPROP;
6546: *code++ = ptype;
6547: *code++ = pdata;
6548: }
6549: #else
6550:
6551: /* If Unicode properties are not supported, \X, \P, and \p are not
6552: allowed. */
6553:
6554: else if (-c == ESC_X || -c == ESC_P || -c == ESC_p)
6555: {
6556: *errorcodeptr = ERR45;
6557: goto FAILED;
6558: }
6559: #endif
6560:
6561: /* For the rest (including \X when Unicode properties are supported), we
6562: can obtain the OP value by negating the escape value in the default
6563: situation when PCRE_UCP is not set. When it *is* set, we substitute
6564: Unicode property tests. */
6565:
6566: else
6567: {
6568: #ifdef SUPPORT_UCP
6569: if (-c >= ESC_DU && -c <= ESC_wu)
6570: {
6571: nestptr = ptr + 1; /* Where to resume */
6572: ptr = substitutes[-c - ESC_DU] - 1; /* Just before substitute */
6573: }
6574: else
6575: #endif
6576: /* In non-UTF-8 mode, we turn \C into OP_ALLANY instead of OP_ANYBYTE
6577: so that it works in DFA mode and in lookbehinds. */
6578:
6579: {
6580: previous = (-c > ESC_b && -c < ESC_Z)? code : NULL;
6581: *code++ = (!utf8 && c == -ESC_C)? OP_ALLANY : -c;
6582: }
6583: }
6584: continue;
6585: }
6586:
6587: /* We have a data character whose value is in c. In UTF-8 mode it may have
6588: a value > 127. We set its representation in the length/buffer, and then
6589: handle it as a data character. */
6590:
6591: #ifdef SUPPORT_UTF8
6592: if (utf8 && c > 127)
6593: mclength = _pcre_ord2utf8(c, mcbuffer);
6594: else
6595: #endif
6596:
6597: {
6598: mcbuffer[0] = c;
6599: mclength = 1;
6600: }
6601: goto ONE_CHAR;
6602:
6603:
6604: /* ===================================================================*/
6605: /* Handle a literal character. It is guaranteed not to be whitespace or #
6606: when the extended flag is set. If we are in UTF-8 mode, it may be a
6607: multi-byte literal character. */
6608:
6609: default:
6610: NORMAL_CHAR:
6611: mclength = 1;
6612: mcbuffer[0] = c;
6613:
6614: #ifdef SUPPORT_UTF8
6615: if (utf8 && c >= 0xc0)
6616: {
6617: while ((ptr[1] & 0xc0) == 0x80)
6618: mcbuffer[mclength++] = *(++ptr);
6619: }
6620: #endif
6621:
6622: /* At this point we have the character's bytes in mcbuffer, and the length
6623: in mclength. When not in UTF-8 mode, the length is always 1. */
6624:
6625: ONE_CHAR:
6626: previous = code;
6627: *code++ = ((options & PCRE_CASELESS) != 0)? OP_CHARI : OP_CHAR;
6628: for (c = 0; c < mclength; c++) *code++ = mcbuffer[c];
6629:
6630: /* Remember if \r or \n were seen */
6631:
6632: if (mcbuffer[0] == CHAR_CR || mcbuffer[0] == CHAR_NL)
6633: cd->external_flags |= PCRE_HASCRORLF;
6634:
6635: /* Set the first and required bytes appropriately. If no previous first
6636: byte, set it from this character, but revert to none on a zero repeat.
6637: Otherwise, leave the firstbyte value alone, and don't change it on a zero
6638: repeat. */
6639:
6640: if (firstbyte == REQ_UNSET)
6641: {
6642: zerofirstbyte = REQ_NONE;
6643: zeroreqbyte = reqbyte;
6644:
6645: /* If the character is more than one byte long, we can set firstbyte
6646: only if it is not to be matched caselessly. */
6647:
6648: if (mclength == 1 || req_caseopt == 0)
6649: {
6650: firstbyte = mcbuffer[0] | req_caseopt;
6651: if (mclength != 1) reqbyte = code[-1] | cd->req_varyopt;
6652: }
6653: else firstbyte = reqbyte = REQ_NONE;
6654: }
6655:
6656: /* firstbyte was previously set; we can set reqbyte only if the length is
6657: 1 or the matching is caseful. */
6658:
6659: else
6660: {
6661: zerofirstbyte = firstbyte;
6662: zeroreqbyte = reqbyte;
6663: if (mclength == 1 || req_caseopt == 0)
6664: reqbyte = code[-1] | req_caseopt | cd->req_varyopt;
6665: }
6666:
6667: break; /* End of literal character handling */
6668: }
6669: } /* end of big loop */
6670:
6671:
6672: /* Control never reaches here by falling through, only by a goto for all the
6673: error states. Pass back the position in the pattern so that it can be displayed
6674: to the user for diagnosing the error. */
6675:
6676: FAILED:
6677: *ptrptr = ptr;
6678: return FALSE;
6679: }
6680:
6681:
6682:
6683:
6684: /*************************************************
6685: * Compile sequence of alternatives *
6686: *************************************************/
6687:
6688: /* On entry, ptr is pointing past the bracket character, but on return it
6689: points to the closing bracket, or vertical bar, or end of string. The code
6690: variable is pointing at the byte into which the BRA operator has been stored.
6691: This function is used during the pre-compile phase when we are trying to find
6692: out the amount of memory needed, as well as during the real compile phase. The
6693: value of lengthptr distinguishes the two phases.
6694:
6695: Arguments:
6696: options option bits, including any changes for this subpattern
6697: codeptr -> the address of the current code pointer
6698: ptrptr -> the address of the current pattern pointer
6699: errorcodeptr -> pointer to error code variable
6700: lookbehind TRUE if this is a lookbehind assertion
6701: reset_bracount TRUE to reset the count for each branch
6702: skipbytes skip this many bytes at start (for brackets and OP_COND)
6703: cond_depth depth of nesting for conditional subpatterns
6704: firstbyteptr place to put the first required character, or a negative number
6705: reqbyteptr place to put the last required character, or a negative number
6706: bcptr pointer to the chain of currently open branches
6707: cd points to the data block with tables pointers etc.
6708: lengthptr NULL during the real compile phase
6709: points to length accumulator during pre-compile phase
6710:
6711: Returns: TRUE on success
6712: */
6713:
6714: static BOOL
6715: compile_regex(int options, uschar **codeptr, const uschar **ptrptr,
6716: int *errorcodeptr, BOOL lookbehind, BOOL reset_bracount, int skipbytes,
6717: int cond_depth, int *firstbyteptr, int *reqbyteptr, branch_chain *bcptr,
6718: compile_data *cd, int *lengthptr)
6719: {
6720: const uschar *ptr = *ptrptr;
6721: uschar *code = *codeptr;
6722: uschar *last_branch = code;
6723: uschar *start_bracket = code;
6724: uschar *reverse_count = NULL;
6725: open_capitem capitem;
6726: int capnumber = 0;
6727: int firstbyte, reqbyte;
6728: int branchfirstbyte, branchreqbyte;
6729: int length;
6730: int orig_bracount;
6731: int max_bracount;
6732: branch_chain bc;
6733:
6734: bc.outer = bcptr;
6735: bc.current_branch = code;
6736:
6737: firstbyte = reqbyte = REQ_UNSET;
6738:
6739: /* Accumulate the length for use in the pre-compile phase. Start with the
6740: length of the BRA and KET and any extra bytes that are required at the
6741: beginning. We accumulate in a local variable to save frequent testing of
6742: lenthptr for NULL. We cannot do this by looking at the value of code at the
6743: start and end of each alternative, because compiled items are discarded during
6744: the pre-compile phase so that the work space is not exceeded. */
6745:
6746: length = 2 + 2*LINK_SIZE + skipbytes;
6747:
6748: /* WARNING: If the above line is changed for any reason, you must also change
6749: the code that abstracts option settings at the start of the pattern and makes
6750: them global. It tests the value of length for (2 + 2*LINK_SIZE) in the
6751: pre-compile phase to find out whether anything has yet been compiled or not. */
6752:
6753: /* If this is a capturing subpattern, add to the chain of open capturing items
6754: so that we can detect them if (*ACCEPT) is encountered. This is also used to
6755: detect groups that contain recursive back references to themselves. Note that
6756: only OP_CBRA need be tested here; changing this opcode to one of its variants,
6757: e.g. OP_SCBRAPOS, happens later, after the group has been compiled. */
6758:
6759: if (*code == OP_CBRA)
6760: {
6761: capnumber = GET2(code, 1 + LINK_SIZE);
6762: capitem.number = capnumber;
6763: capitem.next = cd->open_caps;
6764: capitem.flag = FALSE;
6765: cd->open_caps = &capitem;
6766: }
6767:
6768: /* Offset is set zero to mark that this bracket is still open */
6769:
6770: PUT(code, 1, 0);
6771: code += 1 + LINK_SIZE + skipbytes;
6772:
6773: /* Loop for each alternative branch */
6774:
6775: orig_bracount = max_bracount = cd->bracount;
6776: for (;;)
6777: {
6778: /* For a (?| group, reset the capturing bracket count so that each branch
6779: uses the same numbers. */
6780:
6781: if (reset_bracount) cd->bracount = orig_bracount;
6782:
6783: /* Set up dummy OP_REVERSE if lookbehind assertion */
6784:
6785: if (lookbehind)
6786: {
6787: *code++ = OP_REVERSE;
6788: reverse_count = code;
6789: PUTINC(code, 0, 0);
6790: length += 1 + LINK_SIZE;
6791: }
6792:
6793: /* Now compile the branch; in the pre-compile phase its length gets added
6794: into the length. */
6795:
6796: if (!compile_branch(&options, &code, &ptr, errorcodeptr, &branchfirstbyte,
6797: &branchreqbyte, &bc, cond_depth, cd,
6798: (lengthptr == NULL)? NULL : &length))
6799: {
6800: *ptrptr = ptr;
6801: return FALSE;
6802: }
6803:
6804: /* Keep the highest bracket count in case (?| was used and some branch
6805: has fewer than the rest. */
6806:
6807: if (cd->bracount > max_bracount) max_bracount = cd->bracount;
6808:
6809: /* In the real compile phase, there is some post-processing to be done. */
6810:
6811: if (lengthptr == NULL)
6812: {
6813: /* If this is the first branch, the firstbyte and reqbyte values for the
6814: branch become the values for the regex. */
6815:
6816: if (*last_branch != OP_ALT)
6817: {
6818: firstbyte = branchfirstbyte;
6819: reqbyte = branchreqbyte;
6820: }
6821:
6822: /* If this is not the first branch, the first char and reqbyte have to
6823: match the values from all the previous branches, except that if the
6824: previous value for reqbyte didn't have REQ_VARY set, it can still match,
6825: and we set REQ_VARY for the regex. */
6826:
6827: else
6828: {
6829: /* If we previously had a firstbyte, but it doesn't match the new branch,
6830: we have to abandon the firstbyte for the regex, but if there was
6831: previously no reqbyte, it takes on the value of the old firstbyte. */
6832:
6833: if (firstbyte >= 0 && firstbyte != branchfirstbyte)
6834: {
6835: if (reqbyte < 0) reqbyte = firstbyte;
6836: firstbyte = REQ_NONE;
6837: }
6838:
6839: /* If we (now or from before) have no firstbyte, a firstbyte from the
6840: branch becomes a reqbyte if there isn't a branch reqbyte. */
6841:
6842: if (firstbyte < 0 && branchfirstbyte >= 0 && branchreqbyte < 0)
6843: branchreqbyte = branchfirstbyte;
6844:
6845: /* Now ensure that the reqbytes match */
6846:
6847: if ((reqbyte & ~REQ_VARY) != (branchreqbyte & ~REQ_VARY))
6848: reqbyte = REQ_NONE;
6849: else reqbyte |= branchreqbyte; /* To "or" REQ_VARY */
6850: }
6851:
6852: /* If lookbehind, check that this branch matches a fixed-length string, and
6853: put the length into the OP_REVERSE item. Temporarily mark the end of the
6854: branch with OP_END. If the branch contains OP_RECURSE, the result is -3
6855: because there may be forward references that we can't check here. Set a
6856: flag to cause another lookbehind check at the end. Why not do it all at the
6857: end? Because common, erroneous checks are picked up here and the offset of
6858: the problem can be shown. */
6859:
6860: if (lookbehind)
6861: {
6862: int fixed_length;
6863: *code = OP_END;
6864: fixed_length = find_fixedlength(last_branch, (options & PCRE_UTF8) != 0,
6865: FALSE, cd);
6866: DPRINTF(("fixed length = %d\n", fixed_length));
6867: if (fixed_length == -3)
6868: {
6869: cd->check_lookbehind = TRUE;
6870: }
6871: else if (fixed_length < 0)
6872: {
6873: *errorcodeptr = (fixed_length == -2)? ERR36 :
6874: (fixed_length == -4)? ERR70: ERR25;
6875: *ptrptr = ptr;
6876: return FALSE;
6877: }
6878: else { PUT(reverse_count, 0, fixed_length); }
6879: }
6880: }
6881:
6882: /* Reached end of expression, either ')' or end of pattern. In the real
6883: compile phase, go back through the alternative branches and reverse the chain
6884: of offsets, with the field in the BRA item now becoming an offset to the
6885: first alternative. If there are no alternatives, it points to the end of the
6886: group. The length in the terminating ket is always the length of the whole
6887: bracketed item. Return leaving the pointer at the terminating char. */
6888:
6889: if (*ptr != CHAR_VERTICAL_LINE)
6890: {
6891: if (lengthptr == NULL)
6892: {
6893: int branch_length = (int)(code - last_branch);
6894: do
6895: {
6896: int prev_length = GET(last_branch, 1);
6897: PUT(last_branch, 1, branch_length);
6898: branch_length = prev_length;
6899: last_branch -= branch_length;
6900: }
6901: while (branch_length > 0);
6902: }
6903:
6904: /* Fill in the ket */
6905:
6906: *code = OP_KET;
6907: PUT(code, 1, (int)(code - start_bracket));
6908: code += 1 + LINK_SIZE;
6909:
6910: /* If it was a capturing subpattern, check to see if it contained any
6911: recursive back references. If so, we must wrap it in atomic brackets.
6912: In any event, remove the block from the chain. */
6913:
6914: if (capnumber > 0)
6915: {
6916: if (cd->open_caps->flag)
6917: {
6918: memmove(start_bracket + 1 + LINK_SIZE, start_bracket,
6919: code - start_bracket);
6920: *start_bracket = OP_ONCE;
6921: code += 1 + LINK_SIZE;
6922: PUT(start_bracket, 1, (int)(code - start_bracket));
6923: *code = OP_KET;
6924: PUT(code, 1, (int)(code - start_bracket));
6925: code += 1 + LINK_SIZE;
6926: length += 2 + 2*LINK_SIZE;
6927: }
6928: cd->open_caps = cd->open_caps->next;
6929: }
6930:
6931: /* Retain the highest bracket number, in case resetting was used. */
6932:
6933: cd->bracount = max_bracount;
6934:
6935: /* Set values to pass back */
6936:
6937: *codeptr = code;
6938: *ptrptr = ptr;
6939: *firstbyteptr = firstbyte;
6940: *reqbyteptr = reqbyte;
6941: if (lengthptr != NULL)
6942: {
6943: if (OFLOW_MAX - *lengthptr < length)
6944: {
6945: *errorcodeptr = ERR20;
6946: return FALSE;
6947: }
6948: *lengthptr += length;
6949: }
6950: return TRUE;
6951: }
6952:
6953: /* Another branch follows. In the pre-compile phase, we can move the code
6954: pointer back to where it was for the start of the first branch. (That is,
6955: pretend that each branch is the only one.)
6956:
6957: In the real compile phase, insert an ALT node. Its length field points back
6958: to the previous branch while the bracket remains open. At the end the chain
6959: is reversed. It's done like this so that the start of the bracket has a
6960: zero offset until it is closed, making it possible to detect recursion. */
6961:
6962: if (lengthptr != NULL)
6963: {
6964: code = *codeptr + 1 + LINK_SIZE + skipbytes;
6965: length += 1 + LINK_SIZE;
6966: }
6967: else
6968: {
6969: *code = OP_ALT;
6970: PUT(code, 1, (int)(code - last_branch));
6971: bc.current_branch = last_branch = code;
6972: code += 1 + LINK_SIZE;
6973: }
6974:
6975: ptr++;
6976: }
6977: /* Control never reaches here */
6978: }
6979:
6980:
6981:
6982:
6983: /*************************************************
6984: * Check for anchored expression *
6985: *************************************************/
6986:
6987: /* Try to find out if this is an anchored regular expression. Consider each
6988: alternative branch. If they all start with OP_SOD or OP_CIRC, or with a bracket
6989: all of whose alternatives start with OP_SOD or OP_CIRC (recurse ad lib), then
6990: it's anchored. However, if this is a multiline pattern, then only OP_SOD will
6991: be found, because ^ generates OP_CIRCM in that mode.
6992:
6993: We can also consider a regex to be anchored if OP_SOM starts all its branches.
6994: This is the code for \G, which means "match at start of match position, taking
6995: into account the match offset".
6996:
6997: A branch is also implicitly anchored if it starts with .* and DOTALL is set,
6998: because that will try the rest of the pattern at all possible matching points,
6999: so there is no point trying again.... er ....
7000:
7001: .... except when the .* appears inside capturing parentheses, and there is a
7002: subsequent back reference to those parentheses. We haven't enough information
7003: to catch that case precisely.
7004:
7005: At first, the best we could do was to detect when .* was in capturing brackets
7006: and the highest back reference was greater than or equal to that level.
7007: However, by keeping a bitmap of the first 31 back references, we can catch some
7008: of the more common cases more precisely.
7009:
7010: Arguments:
7011: code points to start of expression (the bracket)
7012: bracket_map a bitmap of which brackets we are inside while testing; this
7013: handles up to substring 31; after that we just have to take
7014: the less precise approach
7015: backref_map the back reference bitmap
7016:
7017: Returns: TRUE or FALSE
7018: */
7019:
7020: static BOOL
7021: is_anchored(register const uschar *code, unsigned int bracket_map,
7022: unsigned int backref_map)
7023: {
7024: do {
7025: const uschar *scode = first_significant_code(code + _pcre_OP_lengths[*code],
7026: FALSE);
7027: register int op = *scode;
7028:
7029: /* Non-capturing brackets */
7030:
7031: if (op == OP_BRA || op == OP_BRAPOS ||
7032: op == OP_SBRA || op == OP_SBRAPOS)
7033: {
7034: if (!is_anchored(scode, bracket_map, backref_map)) return FALSE;
7035: }
7036:
7037: /* Capturing brackets */
7038:
7039: else if (op == OP_CBRA || op == OP_CBRAPOS ||
7040: op == OP_SCBRA || op == OP_SCBRAPOS)
7041: {
7042: int n = GET2(scode, 1+LINK_SIZE);
7043: int new_map = bracket_map | ((n < 32)? (1 << n) : 1);
7044: if (!is_anchored(scode, new_map, backref_map)) return FALSE;
7045: }
7046:
7047: /* Other brackets */
7048:
7049: else if (op == OP_ASSERT || op == OP_ONCE || op == OP_ONCE_NC ||
7050: op == OP_COND)
7051: {
7052: if (!is_anchored(scode, bracket_map, backref_map)) return FALSE;
7053: }
7054:
7055: /* .* is not anchored unless DOTALL is set (which generates OP_ALLANY) and
7056: it isn't in brackets that are or may be referenced. */
7057:
7058: else if ((op == OP_TYPESTAR || op == OP_TYPEMINSTAR ||
7059: op == OP_TYPEPOSSTAR))
7060: {
7061: if (scode[1] != OP_ALLANY || (bracket_map & backref_map) != 0)
7062: return FALSE;
7063: }
7064:
7065: /* Check for explicit anchoring */
7066:
7067: else if (op != OP_SOD && op != OP_SOM && op != OP_CIRC) return FALSE;
7068: code += GET(code, 1);
7069: }
7070: while (*code == OP_ALT); /* Loop for each alternative */
7071: return TRUE;
7072: }
7073:
7074:
7075:
7076: /*************************************************
7077: * Check for starting with ^ or .* *
7078: *************************************************/
7079:
7080: /* This is called to find out if every branch starts with ^ or .* so that
7081: "first char" processing can be done to speed things up in multiline
7082: matching and for non-DOTALL patterns that start with .* (which must start at
7083: the beginning or after \n). As in the case of is_anchored() (see above), we
7084: have to take account of back references to capturing brackets that contain .*
7085: because in that case we can't make the assumption.
7086:
7087: Arguments:
7088: code points to start of expression (the bracket)
7089: bracket_map a bitmap of which brackets we are inside while testing; this
7090: handles up to substring 31; after that we just have to take
7091: the less precise approach
7092: backref_map the back reference bitmap
7093:
7094: Returns: TRUE or FALSE
7095: */
7096:
7097: static BOOL
7098: is_startline(const uschar *code, unsigned int bracket_map,
7099: unsigned int backref_map)
7100: {
7101: do {
7102: const uschar *scode = first_significant_code(code + _pcre_OP_lengths[*code],
7103: FALSE);
7104: register int op = *scode;
7105:
7106: /* If we are at the start of a conditional assertion group, *both* the
7107: conditional assertion *and* what follows the condition must satisfy the test
7108: for start of line. Other kinds of condition fail. Note that there may be an
7109: auto-callout at the start of a condition. */
7110:
7111: if (op == OP_COND)
7112: {
7113: scode += 1 + LINK_SIZE;
7114: if (*scode == OP_CALLOUT) scode += _pcre_OP_lengths[OP_CALLOUT];
7115: switch (*scode)
7116: {
7117: case OP_CREF:
7118: case OP_NCREF:
7119: case OP_RREF:
7120: case OP_NRREF:
7121: case OP_DEF:
7122: return FALSE;
7123:
7124: default: /* Assertion */
7125: if (!is_startline(scode, bracket_map, backref_map)) return FALSE;
7126: do scode += GET(scode, 1); while (*scode == OP_ALT);
7127: scode += 1 + LINK_SIZE;
7128: break;
7129: }
7130: scode = first_significant_code(scode, FALSE);
7131: op = *scode;
7132: }
7133:
7134: /* Non-capturing brackets */
7135:
7136: if (op == OP_BRA || op == OP_BRAPOS ||
7137: op == OP_SBRA || op == OP_SBRAPOS)
7138: {
7139: if (!is_startline(scode, bracket_map, backref_map)) return FALSE;
7140: }
7141:
7142: /* Capturing brackets */
7143:
7144: else if (op == OP_CBRA || op == OP_CBRAPOS ||
7145: op == OP_SCBRA || op == OP_SCBRAPOS)
7146: {
7147: int n = GET2(scode, 1+LINK_SIZE);
7148: int new_map = bracket_map | ((n < 32)? (1 << n) : 1);
7149: if (!is_startline(scode, new_map, backref_map)) return FALSE;
7150: }
7151:
7152: /* Other brackets */
7153:
7154: else if (op == OP_ASSERT || op == OP_ONCE || op == OP_ONCE_NC)
7155: {
7156: if (!is_startline(scode, bracket_map, backref_map)) return FALSE;
7157: }
7158:
7159: /* .* means "start at start or after \n" if it isn't in brackets that
7160: may be referenced. */
7161:
7162: else if (op == OP_TYPESTAR || op == OP_TYPEMINSTAR || op == OP_TYPEPOSSTAR)
7163: {
7164: if (scode[1] != OP_ANY || (bracket_map & backref_map) != 0) return FALSE;
7165: }
7166:
7167: /* Check for explicit circumflex */
7168:
7169: else if (op != OP_CIRC && op != OP_CIRCM) return FALSE;
7170:
7171: /* Move on to the next alternative */
7172:
7173: code += GET(code, 1);
7174: }
7175: while (*code == OP_ALT); /* Loop for each alternative */
7176: return TRUE;
7177: }
7178:
7179:
7180:
7181: /*************************************************
7182: * Check for asserted fixed first char *
7183: *************************************************/
7184:
7185: /* During compilation, the "first char" settings from forward assertions are
7186: discarded, because they can cause conflicts with actual literals that follow.
7187: However, if we end up without a first char setting for an unanchored pattern,
7188: it is worth scanning the regex to see if there is an initial asserted first
7189: char. If all branches start with the same asserted char, or with a bracket all
7190: of whose alternatives start with the same asserted char (recurse ad lib), then
7191: we return that char, otherwise -1.
7192:
7193: Arguments:
7194: code points to start of expression (the bracket)
7195: inassert TRUE if in an assertion
7196:
7197: Returns: -1 or the fixed first char
7198: */
7199:
7200: static int
7201: find_firstassertedchar(const uschar *code, BOOL inassert)
7202: {
7203: register int c = -1;
7204: do {
7205: int d;
7206: int xl = (*code == OP_CBRA || *code == OP_SCBRA ||
7207: *code == OP_CBRAPOS || *code == OP_SCBRAPOS)? 2:0;
7208: const uschar *scode = first_significant_code(code + 1+LINK_SIZE + xl, TRUE);
7209: register int op = *scode;
7210:
7211: switch(op)
7212: {
7213: default:
7214: return -1;
7215:
7216: case OP_BRA:
7217: case OP_BRAPOS:
7218: case OP_CBRA:
7219: case OP_SCBRA:
7220: case OP_CBRAPOS:
7221: case OP_SCBRAPOS:
7222: case OP_ASSERT:
7223: case OP_ONCE:
7224: case OP_ONCE_NC:
7225: case OP_COND:
7226: if ((d = find_firstassertedchar(scode, op == OP_ASSERT)) < 0)
7227: return -1;
7228: if (c < 0) c = d; else if (c != d) return -1;
7229: break;
7230:
7231: case OP_EXACT:
7232: scode += 2;
7233: /* Fall through */
7234:
7235: case OP_CHAR:
7236: case OP_PLUS:
7237: case OP_MINPLUS:
7238: case OP_POSPLUS:
7239: if (!inassert) return -1;
7240: if (c < 0) c = scode[1];
7241: else if (c != scode[1]) return -1;
7242: break;
7243:
7244: case OP_EXACTI:
7245: scode += 2;
7246: /* Fall through */
7247:
7248: case OP_CHARI:
7249: case OP_PLUSI:
7250: case OP_MINPLUSI:
7251: case OP_POSPLUSI:
7252: if (!inassert) return -1;
7253: if (c < 0) c = scode[1] | REQ_CASELESS;
7254: else if (c != scode[1]) return -1;
7255: break;
7256: }
7257:
7258: code += GET(code, 1);
7259: }
7260: while (*code == OP_ALT);
7261: return c;
7262: }
7263:
7264:
7265:
7266: /*************************************************
7267: * Compile a Regular Expression *
7268: *************************************************/
7269:
7270: /* This function takes a string and returns a pointer to a block of store
7271: holding a compiled version of the expression. The original API for this
7272: function had no error code return variable; it is retained for backwards
7273: compatibility. The new function is given a new name.
7274:
7275: Arguments:
7276: pattern the regular expression
7277: options various option bits
7278: errorcodeptr pointer to error code variable (pcre_compile2() only)
7279: can be NULL if you don't want a code value
7280: errorptr pointer to pointer to error text
7281: erroroffset ptr offset in pattern where error was detected
7282: tables pointer to character tables or NULL
7283:
7284: Returns: pointer to compiled data block, or NULL on error,
7285: with errorptr and erroroffset set
7286: */
7287:
7288: PCRE_EXP_DEFN pcre * PCRE_CALL_CONVENTION
7289: pcre_compile(const char *pattern, int options, const char **errorptr,
7290: int *erroroffset, const unsigned char *tables)
7291: {
7292: return pcre_compile2(pattern, options, NULL, errorptr, erroroffset, tables);
7293: }
7294:
7295:
7296: PCRE_EXP_DEFN pcre * PCRE_CALL_CONVENTION
7297: pcre_compile2(const char *pattern, int options, int *errorcodeptr,
7298: const char **errorptr, int *erroroffset, const unsigned char *tables)
7299: {
7300: real_pcre *re;
7301: int length = 1; /* For final END opcode */
7302: int firstbyte, reqbyte, newline;
7303: int errorcode = 0;
7304: int skipatstart = 0;
7305: BOOL utf8;
7306: size_t size;
7307: uschar *code;
7308: const uschar *codestart;
7309: const uschar *ptr;
7310: compile_data compile_block;
7311: compile_data *cd = &compile_block;
7312:
7313: /* This space is used for "compiling" into during the first phase, when we are
7314: computing the amount of memory that is needed. Compiled items are thrown away
7315: as soon as possible, so that a fairly large buffer should be sufficient for
7316: this purpose. The same space is used in the second phase for remembering where
7317: to fill in forward references to subpatterns. That may overflow, in which case
7318: new memory is obtained from malloc(). */
7319:
7320: uschar cworkspace[COMPILE_WORK_SIZE];
7321:
7322: /* Set this early so that early errors get offset 0. */
7323:
7324: ptr = (const uschar *)pattern;
7325:
7326: /* We can't pass back an error message if errorptr is NULL; I guess the best we
7327: can do is just return NULL, but we can set a code value if there is a code
7328: pointer. */
7329:
7330: if (errorptr == NULL)
7331: {
7332: if (errorcodeptr != NULL) *errorcodeptr = 99;
7333: return NULL;
7334: }
7335:
7336: *errorptr = NULL;
7337: if (errorcodeptr != NULL) *errorcodeptr = ERR0;
7338:
7339: /* However, we can give a message for this error */
7340:
7341: if (erroroffset == NULL)
7342: {
7343: errorcode = ERR16;
7344: goto PCRE_EARLY_ERROR_RETURN2;
7345: }
7346:
7347: *erroroffset = 0;
7348:
7349: /* Set up pointers to the individual character tables */
7350:
7351: if (tables == NULL) tables = _pcre_default_tables;
7352: cd->lcc = tables + lcc_offset;
7353: cd->fcc = tables + fcc_offset;
7354: cd->cbits = tables + cbits_offset;
7355: cd->ctypes = tables + ctypes_offset;
7356:
7357: /* Check that all undefined public option bits are zero */
7358:
7359: if ((options & ~PUBLIC_COMPILE_OPTIONS) != 0)
7360: {
7361: errorcode = ERR17;
7362: goto PCRE_EARLY_ERROR_RETURN;
7363: }
7364:
7365: /* Check for global one-time settings at the start of the pattern, and remember
7366: the offset for later. */
7367:
7368: while (ptr[skipatstart] == CHAR_LEFT_PARENTHESIS &&
7369: ptr[skipatstart+1] == CHAR_ASTERISK)
7370: {
7371: int newnl = 0;
7372: int newbsr = 0;
7373:
7374: if (strncmp((char *)(ptr+skipatstart+2), STRING_UTF8_RIGHTPAR, 5) == 0)
7375: { skipatstart += 7; options |= PCRE_UTF8; continue; }
7376: else if (strncmp((char *)(ptr+skipatstart+2), STRING_UCP_RIGHTPAR, 4) == 0)
7377: { skipatstart += 6; options |= PCRE_UCP; continue; }
7378: else if (strncmp((char *)(ptr+skipatstart+2), STRING_NO_START_OPT_RIGHTPAR, 13) == 0)
7379: { skipatstart += 15; options |= PCRE_NO_START_OPTIMIZE; continue; }
7380:
7381: if (strncmp((char *)(ptr+skipatstart+2), STRING_CR_RIGHTPAR, 3) == 0)
7382: { skipatstart += 5; newnl = PCRE_NEWLINE_CR; }
7383: else if (strncmp((char *)(ptr+skipatstart+2), STRING_LF_RIGHTPAR, 3) == 0)
7384: { skipatstart += 5; newnl = PCRE_NEWLINE_LF; }
7385: else if (strncmp((char *)(ptr+skipatstart+2), STRING_CRLF_RIGHTPAR, 5) == 0)
7386: { skipatstart += 7; newnl = PCRE_NEWLINE_CR + PCRE_NEWLINE_LF; }
7387: else if (strncmp((char *)(ptr+skipatstart+2), STRING_ANY_RIGHTPAR, 4) == 0)
7388: { skipatstart += 6; newnl = PCRE_NEWLINE_ANY; }
7389: else if (strncmp((char *)(ptr+skipatstart+2), STRING_ANYCRLF_RIGHTPAR, 8) == 0)
7390: { skipatstart += 10; newnl = PCRE_NEWLINE_ANYCRLF; }
7391:
7392: else if (strncmp((char *)(ptr+skipatstart+2), STRING_BSR_ANYCRLF_RIGHTPAR, 12) == 0)
7393: { skipatstart += 14; newbsr = PCRE_BSR_ANYCRLF; }
7394: else if (strncmp((char *)(ptr+skipatstart+2), STRING_BSR_UNICODE_RIGHTPAR, 12) == 0)
7395: { skipatstart += 14; newbsr = PCRE_BSR_UNICODE; }
7396:
7397: if (newnl != 0)
7398: options = (options & ~PCRE_NEWLINE_BITS) | newnl;
7399: else if (newbsr != 0)
7400: options = (options & ~(PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) | newbsr;
7401: else break;
7402: }
7403:
7404: utf8 = (options & PCRE_UTF8) != 0;
7405:
7406: /* Can't support UTF8 unless PCRE has been compiled to include the code. The
7407: return of an error code from _pcre_valid_utf8() is a new feature, introduced in
7408: release 8.13. It is passed back from pcre_[dfa_]exec(), but at the moment is
7409: not used here. */
7410:
7411: #ifdef SUPPORT_UTF8
7412: if (utf8 && (options & PCRE_NO_UTF8_CHECK) == 0 &&
7413: (errorcode = _pcre_valid_utf8((USPTR)pattern, -1, erroroffset)) != 0)
7414: {
7415: errorcode = ERR44;
7416: goto PCRE_EARLY_ERROR_RETURN2;
7417: }
7418: #else
7419: if (utf8)
7420: {
7421: errorcode = ERR32;
7422: goto PCRE_EARLY_ERROR_RETURN;
7423: }
7424: #endif
7425:
7426: /* Can't support UCP unless PCRE has been compiled to include the code. */
7427:
7428: #ifndef SUPPORT_UCP
7429: if ((options & PCRE_UCP) != 0)
7430: {
7431: errorcode = ERR67;
7432: goto PCRE_EARLY_ERROR_RETURN;
7433: }
7434: #endif
7435:
7436: /* Check validity of \R options. */
7437:
7438: if ((options & (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)) ==
7439: (PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE))
7440: {
7441: errorcode = ERR56;
7442: goto PCRE_EARLY_ERROR_RETURN;
7443: }
7444:
7445: /* Handle different types of newline. The three bits give seven cases. The
7446: current code allows for fixed one- or two-byte sequences, plus "any" and
7447: "anycrlf". */
7448:
7449: switch (options & PCRE_NEWLINE_BITS)
7450: {
7451: case 0: newline = NEWLINE; break; /* Build-time default */
7452: case PCRE_NEWLINE_CR: newline = CHAR_CR; break;
7453: case PCRE_NEWLINE_LF: newline = CHAR_NL; break;
7454: case PCRE_NEWLINE_CR+
7455: PCRE_NEWLINE_LF: newline = (CHAR_CR << 8) | CHAR_NL; break;
7456: case PCRE_NEWLINE_ANY: newline = -1; break;
7457: case PCRE_NEWLINE_ANYCRLF: newline = -2; break;
7458: default: errorcode = ERR56; goto PCRE_EARLY_ERROR_RETURN;
7459: }
7460:
7461: if (newline == -2)
7462: {
7463: cd->nltype = NLTYPE_ANYCRLF;
7464: }
7465: else if (newline < 0)
7466: {
7467: cd->nltype = NLTYPE_ANY;
7468: }
7469: else
7470: {
7471: cd->nltype = NLTYPE_FIXED;
7472: if (newline > 255)
7473: {
7474: cd->nllen = 2;
7475: cd->nl[0] = (newline >> 8) & 255;
7476: cd->nl[1] = newline & 255;
7477: }
7478: else
7479: {
7480: cd->nllen = 1;
7481: cd->nl[0] = newline;
7482: }
7483: }
7484:
7485: /* Maximum back reference and backref bitmap. The bitmap records up to 31 back
7486: references to help in deciding whether (.*) can be treated as anchored or not.
7487: */
7488:
7489: cd->top_backref = 0;
7490: cd->backref_map = 0;
7491:
7492: /* Reflect pattern for debugging output */
7493:
7494: DPRINTF(("------------------------------------------------------------------\n"));
7495: DPRINTF(("%s\n", pattern));
7496:
7497: /* Pretend to compile the pattern while actually just accumulating the length
7498: of memory required. This behaviour is triggered by passing a non-NULL final
7499: argument to compile_regex(). We pass a block of workspace (cworkspace) for it
7500: to compile parts of the pattern into; the compiled code is discarded when it is
7501: no longer needed, so hopefully this workspace will never overflow, though there
7502: is a test for its doing so. */
7503:
7504: cd->bracount = cd->final_bracount = 0;
7505: cd->names_found = 0;
7506: cd->name_entry_size = 0;
7507: cd->name_table = NULL;
7508: cd->start_code = cworkspace;
7509: cd->hwm = cworkspace;
7510: cd->start_workspace = cworkspace;
7511: cd->workspace_size = COMPILE_WORK_SIZE;
7512: cd->start_pattern = (const uschar *)pattern;
7513: cd->end_pattern = (const uschar *)(pattern + strlen(pattern));
7514: cd->req_varyopt = 0;
7515: cd->external_options = options;
7516: cd->external_flags = 0;
7517: cd->open_caps = NULL;
7518:
7519: /* Now do the pre-compile. On error, errorcode will be set non-zero, so we
7520: don't need to look at the result of the function here. The initial options have
7521: been put into the cd block so that they can be changed if an option setting is
7522: found within the regex right at the beginning. Bringing initial option settings
7523: outside can help speed up starting point checks. */
7524:
7525: ptr += skipatstart;
7526: code = cworkspace;
7527: *code = OP_BRA;
7528: (void)compile_regex(cd->external_options, &code, &ptr, &errorcode, FALSE,
7529: FALSE, 0, 0, &firstbyte, &reqbyte, NULL, cd, &length);
7530: if (errorcode != 0) goto PCRE_EARLY_ERROR_RETURN;
7531:
7532: DPRINTF(("end pre-compile: length=%d workspace=%d\n", length,
7533: cd->hwm - cworkspace));
7534:
7535: if (length > MAX_PATTERN_SIZE)
7536: {
7537: errorcode = ERR20;
7538: goto PCRE_EARLY_ERROR_RETURN;
7539: }
7540:
7541: /* Compute the size of data block needed and get it, either from malloc or
7542: externally provided function. Integer overflow should no longer be possible
7543: because nowadays we limit the maximum value of cd->names_found and
7544: cd->name_entry_size. */
7545:
7546: size = length + sizeof(real_pcre) + cd->names_found * cd->name_entry_size;
7547: re = (real_pcre *)(pcre_malloc)(size);
7548:
7549: if (re == NULL)
7550: {
7551: errorcode = ERR21;
7552: goto PCRE_EARLY_ERROR_RETURN;
7553: }
7554:
7555: /* Put in the magic number, and save the sizes, initial options, internal
7556: flags, and character table pointer. NULL is used for the default character
7557: tables. The nullpad field is at the end; it's there to help in the case when a
7558: regex compiled on a system with 4-byte pointers is run on another with 8-byte
7559: pointers. */
7560:
7561: re->magic_number = MAGIC_NUMBER;
7562: re->size = (int)size;
7563: re->options = cd->external_options;
7564: re->flags = cd->external_flags;
7565: re->dummy1 = 0;
7566: re->first_byte = 0;
7567: re->req_byte = 0;
7568: re->name_table_offset = sizeof(real_pcre);
7569: re->name_entry_size = cd->name_entry_size;
7570: re->name_count = cd->names_found;
7571: re->ref_count = 0;
7572: re->tables = (tables == _pcre_default_tables)? NULL : tables;
7573: re->nullpad = NULL;
7574:
7575: /* The starting points of the name/number translation table and of the code are
7576: passed around in the compile data block. The start/end pattern and initial
7577: options are already set from the pre-compile phase, as is the name_entry_size
7578: field. Reset the bracket count and the names_found field. Also reset the hwm
7579: field; this time it's used for remembering forward references to subpatterns.
7580: */
7581:
7582: cd->final_bracount = cd->bracount; /* Save for checking forward references */
7583: cd->assert_depth = 0;
7584: cd->bracount = 0;
7585: cd->names_found = 0;
7586: cd->name_table = (uschar *)re + re->name_table_offset;
7587: codestart = cd->name_table + re->name_entry_size * re->name_count;
7588: cd->start_code = codestart;
7589: cd->hwm = (uschar *)(cd->start_workspace);
7590: cd->req_varyopt = 0;
7591: cd->had_accept = FALSE;
7592: cd->check_lookbehind = FALSE;
7593: cd->open_caps = NULL;
7594:
7595: /* Set up a starting, non-extracting bracket, then compile the expression. On
7596: error, errorcode will be set non-zero, so we don't need to look at the result
7597: of the function here. */
7598:
7599: ptr = (const uschar *)pattern + skipatstart;
7600: code = (uschar *)codestart;
7601: *code = OP_BRA;
7602: (void)compile_regex(re->options, &code, &ptr, &errorcode, FALSE, FALSE, 0, 0,
7603: &firstbyte, &reqbyte, NULL, cd, NULL);
7604: re->top_bracket = cd->bracount;
7605: re->top_backref = cd->top_backref;
7606: re->flags = cd->external_flags;
7607:
7608: if (cd->had_accept) reqbyte = REQ_NONE; /* Must disable after (*ACCEPT) */
7609:
7610: /* If not reached end of pattern on success, there's an excess bracket. */
7611:
7612: if (errorcode == 0 && *ptr != 0) errorcode = ERR22;
7613:
7614: /* Fill in the terminating state and check for disastrous overflow, but
7615: if debugging, leave the test till after things are printed out. */
7616:
7617: *code++ = OP_END;
7618:
7619: #ifndef PCRE_DEBUG
7620: if (code - codestart > length) errorcode = ERR23;
7621: #endif
7622:
7623: /* Fill in any forward references that are required. There may be repeated
7624: references; optimize for them, as searching a large regex takes time. */
7625:
7626: if (cd->hwm > cd->start_workspace)
7627: {
7628: int prev_recno = -1;
7629: const uschar *groupptr = NULL;
7630: while (errorcode == 0 && cd->hwm > cd->start_workspace)
7631: {
7632: int offset, recno;
7633: cd->hwm -= LINK_SIZE;
7634: offset = GET(cd->hwm, 0);
7635: recno = GET(codestart, offset);
7636: if (recno != prev_recno)
7637: {
7638: groupptr = _pcre_find_bracket(codestart, utf8, recno);
7639: prev_recno = recno;
7640: }
7641: if (groupptr == NULL) errorcode = ERR53;
7642: else PUT(((uschar *)codestart), offset, (int)(groupptr - codestart));
7643: }
7644: }
7645:
7646: /* If the workspace had to be expanded, free the new memory. */
7647:
7648: if (cd->workspace_size > COMPILE_WORK_SIZE)
7649: (pcre_free)((void *)cd->start_workspace);
7650:
7651: /* Give an error if there's back reference to a non-existent capturing
7652: subpattern. */
7653:
7654: if (errorcode == 0 && re->top_backref > re->top_bracket) errorcode = ERR15;
7655:
7656: /* If there were any lookbehind assertions that contained OP_RECURSE
7657: (recursions or subroutine calls), a flag is set for them to be checked here,
7658: because they may contain forward references. Actual recursions can't be fixed
7659: length, but subroutine calls can. It is done like this so that those without
7660: OP_RECURSE that are not fixed length get a diagnosic with a useful offset. The
7661: exceptional ones forgo this. We scan the pattern to check that they are fixed
7662: length, and set their lengths. */
7663:
7664: if (cd->check_lookbehind)
7665: {
7666: uschar *cc = (uschar *)codestart;
7667:
7668: /* Loop, searching for OP_REVERSE items, and process those that do not have
7669: their length set. (Actually, it will also re-process any that have a length
7670: of zero, but that is a pathological case, and it does no harm.) When we find
7671: one, we temporarily terminate the branch it is in while we scan it. */
7672:
7673: for (cc = (uschar *)_pcre_find_bracket(codestart, utf8, -1);
7674: cc != NULL;
7675: cc = (uschar *)_pcre_find_bracket(cc, utf8, -1))
7676: {
7677: if (GET(cc, 1) == 0)
7678: {
7679: int fixed_length;
7680: uschar *be = cc - 1 - LINK_SIZE + GET(cc, -LINK_SIZE);
7681: int end_op = *be;
7682: *be = OP_END;
7683: fixed_length = find_fixedlength(cc, (re->options & PCRE_UTF8) != 0, TRUE,
7684: cd);
7685: *be = end_op;
7686: DPRINTF(("fixed length = %d\n", fixed_length));
7687: if (fixed_length < 0)
7688: {
7689: errorcode = (fixed_length == -2)? ERR36 :
7690: (fixed_length == -4)? ERR70 : ERR25;
7691: break;
7692: }
7693: PUT(cc, 1, fixed_length);
7694: }
7695: cc += 1 + LINK_SIZE;
7696: }
7697: }
7698:
7699: /* Failed to compile, or error while post-processing */
7700:
7701: if (errorcode != 0)
7702: {
7703: (pcre_free)(re);
7704: PCRE_EARLY_ERROR_RETURN:
7705: *erroroffset = (int)(ptr - (const uschar *)pattern);
7706: PCRE_EARLY_ERROR_RETURN2:
7707: *errorptr = find_error_text(errorcode);
7708: if (errorcodeptr != NULL) *errorcodeptr = errorcode;
7709: return NULL;
7710: }
7711:
7712: /* If the anchored option was not passed, set the flag if we can determine that
7713: the pattern is anchored by virtue of ^ characters or \A or anything else (such
7714: as starting with .* when DOTALL is set).
7715:
7716: Otherwise, if we know what the first byte has to be, save it, because that
7717: speeds up unanchored matches no end. If not, see if we can set the
7718: PCRE_STARTLINE flag. This is helpful for multiline matches when all branches
7719: start with ^. and also when all branches start with .* for non-DOTALL matches.
7720: */
7721:
7722: if ((re->options & PCRE_ANCHORED) == 0)
7723: {
7724: if (is_anchored(codestart, 0, cd->backref_map))
7725: re->options |= PCRE_ANCHORED;
7726: else
7727: {
7728: if (firstbyte < 0)
7729: firstbyte = find_firstassertedchar(codestart, FALSE);
7730: if (firstbyte >= 0) /* Remove caseless flag for non-caseable chars */
7731: {
7732: int ch = firstbyte & 255;
7733: re->first_byte = ((firstbyte & REQ_CASELESS) != 0 &&
7734: cd->fcc[ch] == ch)? ch : firstbyte;
7735: re->flags |= PCRE_FIRSTSET;
7736: }
7737: else if (is_startline(codestart, 0, cd->backref_map))
7738: re->flags |= PCRE_STARTLINE;
7739: }
7740: }
7741:
7742: /* For an anchored pattern, we use the "required byte" only if it follows a
7743: variable length item in the regex. Remove the caseless flag for non-caseable
7744: bytes. */
7745:
7746: if (reqbyte >= 0 &&
7747: ((re->options & PCRE_ANCHORED) == 0 || (reqbyte & REQ_VARY) != 0))
7748: {
7749: int ch = reqbyte & 255;
7750: re->req_byte = ((reqbyte & REQ_CASELESS) != 0 &&
7751: cd->fcc[ch] == ch)? (reqbyte & ~REQ_CASELESS) : reqbyte;
7752: re->flags |= PCRE_REQCHSET;
7753: }
7754:
7755: /* Print out the compiled data if debugging is enabled. This is never the
7756: case when building a production library. */
7757:
7758: #ifdef PCRE_DEBUG
7759: printf("Length = %d top_bracket = %d top_backref = %d\n",
7760: length, re->top_bracket, re->top_backref);
7761:
7762: printf("Options=%08x\n", re->options);
7763:
7764: if ((re->flags & PCRE_FIRSTSET) != 0)
7765: {
7766: int ch = re->first_byte & 255;
7767: const char *caseless = ((re->first_byte & REQ_CASELESS) == 0)?
7768: "" : " (caseless)";
7769: if (isprint(ch)) printf("First char = %c%s\n", ch, caseless);
7770: else printf("First char = \\x%02x%s\n", ch, caseless);
7771: }
7772:
7773: if ((re->flags & PCRE_REQCHSET) != 0)
7774: {
7775: int ch = re->req_byte & 255;
7776: const char *caseless = ((re->req_byte & REQ_CASELESS) == 0)?
7777: "" : " (caseless)";
7778: if (isprint(ch)) printf("Req char = %c%s\n", ch, caseless);
7779: else printf("Req char = \\x%02x%s\n", ch, caseless);
7780: }
7781:
7782: pcre_printint(re, stdout, TRUE);
7783:
7784: /* This check is done here in the debugging case so that the code that
7785: was compiled can be seen. */
7786:
7787: if (code - codestart > length)
7788: {
7789: (pcre_free)(re);
7790: *errorptr = find_error_text(ERR23);
7791: *erroroffset = ptr - (uschar *)pattern;
7792: if (errorcodeptr != NULL) *errorcodeptr = ERR23;
7793: return NULL;
7794: }
7795: #endif /* PCRE_DEBUG */
7796:
7797: return (pcre *)re;
7798: }
7799:
7800: /* End of pcre_compile.c */
FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>
![[BACK]](/icons/cvsweb/back.gif){kind=icon}
Return to pcre_compile.c CVS log ![[TXT]](/icons/cvsweb/text.gif){kind=icon}
![[DIR]](/icons/cvsweb/dir.gif){kind=icon}
Up to [ELWIX - Embedded LightWeight unIX -] / embedaddon / pcre