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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 47 - (hide annotations) (download)
Sat Feb 24 21:39:29 2007 UTC (7 years, 4 months ago) by nigel
File size: 23292 byte(s)
Load pcre-3.2 into code/trunk.

1 nigel 3 ChangeLog for PCRE
2     ------------------
3    
4 nigel 5
5 nigel 47 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 nigel 45 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 nigel 43 Version 3.0 01-Feb-00
56     ---------------------
57 nigel 41
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 nigel 43 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 nigel 41
82 nigel 43 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 nigel 39 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 nigel 37 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 nigel 35 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 nigel 33 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 nigel 31 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 nigel 29 Version 2.03 02-Feb-99
213     ----------------------
214    
215     1. Fixed typo and small mistake in man page.
216    
217 nigel 31 2. Added 4th condition (GPL supersedes if conflict) and created separate
218     LICENCE file containing the conditions.
219 nigel 29
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 nigel 27 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 nigel 25 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 nigel 23 Version 2.00 24-Sep-98
259 nigel 21 ----------------------
260    
261 nigel 23 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 nigel 21 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 nigel 19 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 nigel 17 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 nigel 15 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 nigel 13 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 nigel 11 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 nigel 9 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 nigel 11 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
405 nigel 9 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 nigel 7 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 nigel 5 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 nigel 3 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