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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 47 - (show annotations) (download)
Sat Feb 24 21:39:29 2007 UTC (7 years, 9 months ago) by nigel
File size: 23292 byte(s)
Load pcre-3.2 into code/trunk.

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

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12