/[pcre]/code/trunk/ChangeLog
ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 51 - (show annotations) (download)
Sat Feb 24 21:39:37 2007 UTC (7 years, 8 months ago) by nigel
File size: 24806 byte(s)
Load pcre-3.4 into code/trunk.

1 ChangeLog for PCRE
2 ------------------
3
4
5 Version 3.4 22-Aug-00
6 ---------------------
7
8 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
9
10 2. Diagnose condition (?(0) as an error instead of crashing on matching.
11
12
13 Version 3.3 01-Aug-00
14 ---------------------
15
16 1. If an octal character was given, but the value was greater than \377, it
17 was not getting masked to the least significant bits, as documented. This could
18 lead to crashes in some systems.
19
20 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
21 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
22
23 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
24 These just pass their arguments on to (pcre_free)(), but they are provided
25 because some uses of PCRE bind it to non-C systems that can call its functions,
26 but cannot call free() or pcre_free() directly.
27
28 4. Add "make test" as a synonym for "make check". Corrected some comments in
29 the Makefile.
30
31 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
32 Makefile.
33
34 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
35 command called pgrep for grepping around the active processes.
36
37 7. Added the beginnings of support for UTF-8 character strings.
38
39 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
40 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
41 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
42 out for the ar command.)
43
44
45 Version 3.2 12-May-00
46 ---------------------
47
48 This is purely a bug fixing release.
49
50 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
51 of ZA. This was just one example of several cases that could provoke this bug,
52 which was introduced by change 9 of version 2.00. The code for breaking
53 infinite loops after an iteration that matches an empty string was't working
54 correctly.
55
56 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
57 when matched against abbab (for example). After matching an empty string, it
58 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
59 caused it to match further down the string than it should.
60
61 3. The code contained an inclusion of sys/types.h. It isn't clear why this
62 was there because it doesn't seem to be needed, and it causes trouble on some
63 systems, as it is not a Standard C header. It has been removed.
64
65 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
66 were reported on the Macintosh. The changes were from
67
68 while ((c = *(++ptr)) != 0 && c != '\n');
69 to
70 while ((c = *(++ptr)) != 0 && c != '\n') ;
71
72 Totally extraordinary, but if that's what it takes...
73
74 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
75 available. Added HAVE_BCOPY and an autoconf test for it; if neither
76 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
77 assumes the way PCRE uses memmove() (always moving upwards).
78
79 6. PCRE is being used in one environment where strchr() is not available. There
80 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
81 faster code anyway.
82
83
84 Version 3.1 09-Feb-00
85 ---------------------
86
87 The only change in this release is the fixing of some bugs in Makefile.in for
88 the "install" target:
89
90 (1) It was failing to install pcreposix.h.
91
92 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
93
94
95 Version 3.0 01-Feb-00
96 ---------------------
97
98 1. Add support for the /+ modifier to perltest (to output $` like it does in
99 pcretest).
100
101 2. Add support for the /g modifier to perltest.
102
103 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
104 matches null strings.
105
106 4. Fix perltest so that it doesn't do unwanted things when fed an empty
107 pattern. Perl treats empty patterns specially - it reuses the most recent
108 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
109 effect.
110
111 5. The POSIX interface was broken in that it was just handing over the POSIX
112 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
113 required a bigger vector, with some working space on the end. This means that
114 the POSIX wrapper now has to get and free some memory, and copy the results.
115
116 6. Added some simple autoconf support, placing the test data and the
117 documentation in separate directories, re-organizing some of the
118 information files, and making it build pcre-config (a GNU standard). Also added
119 libtool support for building PCRE as a shared library, which is now the
120 default.
121
122 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
123 09 are not valid octal constants. Single digits will be used for minor values
124 less than 10.
125
126 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
127 existing programs that set these in the POSIX interface can use PCRE without
128 modification.
129
130 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
131 return all that pcre_info() returns, plus additional data. The pcre_info()
132 function is retained for compatibility, but is considered to be obsolete.
133
134 10. Added experimental recursion feature (?R) to handle one common case that
135 Perl 5.6 will be able to do with (?p{...}).
136
137 11. Added support for POSIX character classes like [:alpha:], which Perl is
138 adopting.
139
140
141 Version 2.08 31-Aug-99
142 ----------------------
143
144 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
145 trying to match at the startoffset position, but instead was moving forward to
146 the next newline as if a previous match had failed.
147
148 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
149 and could get into a loop if a null string was matched other than at the start
150 of the subject.
151
152 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
153 be distinguished at compile time, and for completeness also added PCRE_DATE.
154
155 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
156 in GnuWin32 environments.
157
158
159 Version 2.07 29-Jul-99
160 ----------------------
161
162 1. The documentation is now supplied in plain text form and HTML as well as in
163 the form of man page sources.
164
165 2. C++ compilers don't like assigning (void *) values to other pointer types.
166 In particular this affects malloc(). Although there is no problem in Standard
167 C, I've put in casts to keep C++ compilers happy.
168
169 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
170 should be (const char *).
171
172 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
173 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
174 However, I haven't made this a standard facility. The documentation doesn't
175 mention it, and the Makefile doesn't support it.
176
177 5. The Makefile now contains an "install" target, with editable destinations at
178 the top of the file. The pcretest program is not installed.
179
180 6. pgrep -V now gives the PCRE version number and date.
181
182 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
183 causing the entire string to be ignored, instead of just the last character.
184
185 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
186 non-matching string, it can take a very, very long time, even for strings of
187 quite modest length, because of the nested recursion. PCRE now does better in
188 some of these cases. It does this by remembering the last required literal
189 character in the pattern, and pre-searching the subject to ensure it is present
190 before running the real match. In other words, it applies a heuristic to detect
191 some types of certain failure quickly, and in the above example, if presented
192 with a string that has no trailing " it gives "no match" very quickly.
193
194 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
195 other alternatives are tried instead.
196
197
198 Version 2.06 09-Jun-99
199 ----------------------
200
201 1. Change pcretest's output for amount of store used to show just the code
202 space, because the remainder (the data block) varies in size between 32-bit and
203 64-bit systems.
204
205 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
206 start matching at. This allows lookbehinds to work when searching for multiple
207 occurrences in a string.
208
209 3. Added additional options to pcretest for testing multiple occurrences:
210
211 /+ outputs the rest of the string that follows a match
212 /g loops for multiple occurrences, using the new startoffset argument
213 /G loops for multiple occurrences by passing an incremented pointer
214
215 4. PCRE wasn't doing the "first character" optimization for patterns starting
216 with \b or \B, though it was doing it for other lookbehind assertions. That is,
217 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
218 the letter 'x'. On long subject strings, this gives a significant speed-up.
219
220
221 Version 2.05 21-Apr-99
222 ----------------------
223
224 1. Changed the type of magic_number from int to long int so that it works
225 properly on 16-bit systems.
226
227 2. Fixed a bug which caused patterns starting with .* not to work correctly
228 when the subject string contained newline characters. PCRE was assuming
229 anchoring for such patterns in all cases, which is not correct because .* will
230 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
231 DOTALL is set at top level; otherwise it knows that patterns starting with .*
232 must be retried after every newline in the subject.
233
234
235 Version 2.04 18-Feb-99
236 ----------------------
237
238 1. For parenthesized subpatterns with repeats whose minimum was zero, the
239 computation of the store needed to hold the pattern was incorrect (too large).
240 If such patterns were nested a few deep, this could multiply and become a real
241 problem.
242
243 2. Added /M option to pcretest to show the memory requirement of a specific
244 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
245
246 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
247 compiled in such a way that the backtracking after subsequent failure was
248 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
249 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
250
251
252 Version 2.03 02-Feb-99
253 ----------------------
254
255 1. Fixed typo and small mistake in man page.
256
257 2. Added 4th condition (GPL supersedes if conflict) and created separate
258 LICENCE file containing the conditions.
259
260 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
261 Perl, that is the internal \ allows the delimiter to be included in the
262 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
263 the final delimiter, add \ to the end of the pattern (to test the error).
264
265 4. Added the convenience functions for extracting substrings after a successful
266 match. Updated pcretest to make it able to test these functions.
267
268
269 Version 2.02 14-Jan-99
270 ----------------------
271
272 1. Initialized the working variables associated with each extraction so that
273 their saving and restoring doesn't refer to uninitialized store.
274
275 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
276 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
277 fix the problem.
278
279 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
280 calls, and wasn't printing the correct value for compiling calls. Increased the
281 default value of LOOPREPEAT, and the number of significant figures in the
282 times.
283
284 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
285
286 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
287 a building problem on Windows NT with a FAT file system.
288
289
290 Version 2.01 21-Oct-98
291 ----------------------
292
293 1. Changed the API for pcre_compile() to allow for the provision of a pointer
294 to character tables built by pcre_maketables() in the current locale. If NULL
295 is passed, the default tables are used.
296
297
298 Version 2.00 24-Sep-98
299 ----------------------
300
301 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
302 it any more.
303
304 2. Allow quantification of (?>) groups, and make it work correctly.
305
306 3. The first character computation wasn't working for (?>) groups.
307
308 4. Correct the implementation of \Z (it is permitted to match on the \n at the
309 end of the subject) and add 5.005's \z, which really does match only at the
310 very end of the subject.
311
312 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
313
314 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
315 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
316 localized options. All options to pcre_study() were also removed.
317
318 7. Add other new features from 5.005:
319
320 $(?<= positive lookbehind
321 $(?<! negative lookbehind
322 (?imsx-imsx) added the unsetting capability
323 such a setting is global if at outer level; local otherwise
324 (?imsx-imsx:) non-capturing groups with option setting
325 (?(cond)re|re) conditional pattern matching
326
327 A backreference to itself in a repeated group matches the previous
328 captured string.
329
330 8. General tidying up of studying (both automatic and via "study")
331 consequential on the addition of new assertions.
332
333 9. As in 5.005, unlimited repeated groups that could match an empty substring
334 are no longer faulted at compile time. Instead, the loop is forcibly broken at
335 runtime if any iteration does actually match an empty substring.
336
337 10. Include the RunTest script in the distribution.
338
339 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
340 discrepancies, some of which were old and were also with respect to 5.004. They
341 have now been fixed.
342
343
344 Version 1.09 28-Apr-98
345 ----------------------
346
347 1. A negated single character class followed by a quantifier with a minimum
348 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
349 program crashes, or just wrong answers. This did not apply to negated classes
350 containing more than one character, or to minima other than one.
351
352
353 Version 1.08 27-Mar-98
354 ----------------------
355
356 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
357
358 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
359 latter must appear before anything that relies on it in the pattern.
360
361
362 Version 1.07 16-Feb-98
363 ----------------------
364
365 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
366 repeat of a potentially empty string).
367
368
369 Version 1.06 23-Jan-98
370 ----------------------
371
372 1. Added Markus Oberhumer's little patches for C++.
373
374 2. Literal strings longer than 255 characters were broken.
375
376
377 Version 1.05 23-Dec-97
378 ----------------------
379
380 1. Negated character classes containing more than one character were failing if
381 PCRE_CASELESS was set at run time.
382
383
384 Version 1.04 19-Dec-97
385 ----------------------
386
387 1. Corrected the man page, where some "const" qualifiers had been omitted.
388
389 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
390 input syntax.
391
392 3. Fixed memory leak which occurred when a regex with back references was
393 matched with an offsets vector that wasn't big enough. The temporary memory
394 that is used in this case wasn't being freed if the match failed.
395
396 4. Tidied pcretest to ensure it frees memory that it gets.
397
398 5. Temporary memory was being obtained in the case where the passed offsets
399 vector was exactly big enough.
400
401 6. Corrected definition of offsetof() from change 5 below.
402
403 7. I had screwed up change 6 below and broken the rules for the use of
404 setjmp(). Now fixed.
405
406
407 Version 1.03 18-Dec-97
408 ----------------------
409
410 1. A erroneous regex with a missing opening parenthesis was correctly
411 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
412 on some systems.
413
414 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
415 it was reported that one broken compiler failed on the former because "code" is
416 also an independent variable.
417
418 3. The erroneous regex a[]b caused an array overrun reference.
419
420 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
421 fail on data ending with that character. (It was going on too far, and checking
422 the next character, typically a binary zero.) This was specific to the
423 optimized code for single-character negative classes.
424
425 5. Added a contributed patch from the TIN world which does the following:
426
427 + Add an undef for memmove, in case the the system defines a macro for it.
428
429 + Add a definition of offsetof(), in case there isn't one. (I don't know
430 the reason behind this - offsetof() is part of the ANSI standard - but
431 it does no harm).
432
433 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
434 most of the places where whitespace preceded '#'. I have given up and
435 allowed the remaining 2 cases to be at the margin.
436
437 + Rename some variables in pcre to eliminate shadowing. This seems very
438 pedantic, but does no harm, of course.
439
440 6. Moved the call to setjmp() into its own function, to get rid of warnings
441 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
442
443 7. Constructs such as \d{8,} were compiling into the equivalent of
444 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
445 outcome, but in this particular case used more store than had been allocated,
446 which caused the bug to be discovered because it threw up an internal error.
447
448 8. The debugging code in both pcre and pcretest for outputting the compiled
449 form of a regex was going wrong in the case of back references followed by
450 curly-bracketed repeats.
451
452
453 Version 1.02 12-Dec-97
454 ----------------------
455
456 1. Typos in pcre.3 and comments in the source fixed.
457
458 2. Applied a contributed patch to get rid of places where it used to remove
459 'const' from variables, and fixed some signed/unsigned and uninitialized
460 variable warnings.
461
462 3. Added the "runtest" target to Makefile.
463
464 4. Set default compiler flag to -O2 rather than just -O.
465
466
467 Version 1.01 19-Nov-97
468 ----------------------
469
470 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
471 like /([ab]*)*/, that is, for classes with more than one character in them.
472
473 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
474 as /((?>a*))*/ (a PCRE_EXTRA facility).
475
476
477 Version 1.00 18-Nov-97
478 ----------------------
479
480 1. Added compile-time macros to support systems such as SunOS4 which don't have
481 memmove() or strerror() but have other things that can be used instead.
482
483 2. Arranged that "make clean" removes the executables.
484
485
486 Version 0.99 27-Oct-97
487 ----------------------
488
489 1. Fixed bug in code for optimizing classes with only one character. It was
490 initializing a 32-byte map regardless, which could cause it to run off the end
491 of the memory it had got.
492
493 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
494
495
496 Version 0.98 22-Oct-97
497 ----------------------
498
499 1. Fixed bug in code for handling temporary memory usage when there are more
500 back references than supplied space in the ovector. This could cause segfaults.
501
502
503 Version 0.97 21-Oct-97
504 ----------------------
505
506 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
507
508 2. Optimized negated single characters not to use a bit map.
509
510 3. Brought error texts together as macro definitions; clarified some of them;
511 fixed one that was wrong - it said "range out of order" when it meant "invalid
512 escape sequence".
513
514 4. Changed some char * arguments to const char *.
515
516 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
517
518 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
519 pcretest.
520
521
522 Version 0.96 16-Oct-97
523 ----------------------
524
525 1. Added a simple "pgrep" utility to the distribution.
526
527 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
528 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
529 where "ddd" means "one or more decimal digits".
530
531 3. Fixed serious bug. If a pattern had a back reference, but the call to
532 pcre_exec() didn't supply a large enough ovector to record the related
533 identifying subpattern, the match always failed. PCRE now remembers the number
534 of the largest back reference, and gets some temporary memory in which to save
535 the offsets during matching if necessary, in order to ensure that
536 backreferences always work.
537
538 4. Increased the compatibility with Perl in a number of ways:
539
540 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
541 to request this handling. The option can be set at compile or exec time.
542
543 (b) $ matches before a terminating newline by default; an option
544 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
545 mode). The option can be set at compile or exec time.
546
547 (c) The handling of \ followed by a digit other than 0 is now supposed to be
548 the same as Perl's. If the decimal number it represents is less than 10
549 or there aren't that many previous left capturing parentheses, an octal
550 escape is read. Inside a character class, it's always an octal escape,
551 even if it is a single digit.
552
553 (d) An escaped but undefined alphabetic character is taken as a literal,
554 unless PCRE_EXTRA is set. Currently this just reserves the remaining
555 escapes.
556
557 (e) {0} is now permitted. (The previous item is removed from the compiled
558 pattern).
559
560 5. Changed all the names of code files so that the basic parts are no longer
561 than 10 characters, and abolished the teeny "globals.c" file.
562
563 6. Changed the handling of character classes; they are now done with a 32-byte
564 bit map always.
565
566 7. Added the -d and /D options to pcretest to make it possible to look at the
567 internals of compilation without having to recompile pcre.
568
569
570 Version 0.95 23-Sep-97
571 ----------------------
572
573 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
574 \x20 at the start of a run of normal characters. These were being treated as
575 real characters, instead of the source characters being re-checked.
576
577
578 Version 0.94 18-Sep-97
579 ----------------------
580
581 1. The functions are now thread-safe, with the caveat that the global variables
582 containing pointers to malloc() and free() or alternative functions are the
583 same for all threads.
584
585 2. Get pcre_study() to generate a bitmap of initial characters for non-
586 anchored patterns when this is possible, and use it if passed to pcre_exec().
587
588
589 Version 0.93 15-Sep-97
590 ----------------------
591
592 1. /(b)|(:+)/ was computing an incorrect first character.
593
594 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
595 but not actually doing anything yet.
596
597 3. Treat "-" characters in classes that cannot be part of ranges as literals,
598 as Perl does (e.g. [-az] or [az-]).
599
600 4. Set the anchored flag if a branch starts with .* or .*? because that tests
601 all possible positions.
602
603 5. Split up into different modules to avoid including unneeded functions in a
604 compiled binary. However, compile and exec are still in one module. The "study"
605 function is split off.
606
607 6. The character tables are now in a separate module whose source is generated
608 by an auxiliary program - but can then be edited by hand if required. There are
609 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
610 toupper() in the code.
611
612 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
613 make them global. Abolish the function for setting them, as the caller can now
614 set them directly.
615
616
617 Version 0.92 11-Sep-97
618 ----------------------
619
620 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
621 (e.g. /a{1,3}/) was broken (I mis-optimized it).
622
623 2. Caseless matching was not working in character classes if the characters in
624 the pattern were in upper case.
625
626 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
627
628 4. Make PCRE_ANCHORED public and accept as a compile option.
629
630 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
631 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
632 pass them.
633
634 6. Give an error if bad option bits passed at compile or run time.
635
636 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
637 pcretest to cause it to pass that flag.
638
639 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
640 options, and the first character, if set.
641
642 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
643
644
645 Version 0.91 10-Sep-97
646 ----------------------
647
648 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
649 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
650
651 2. PCRE was looping on encountering an indefinitely repeated back reference to
652 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
653 Perl does - treats the match as successful.
654
655 ****

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12