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

Diff of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 5 by nigel, Sat Feb 24 21:38:05 2007 UTC revision 55 by nigel, Sat Feb 24 21:39:46 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 3.6 23-Oct-01
5    ---------------------
6    
7    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
8    offsets passed as NULL with zero offset count.
9    
10    2. The config.guess and config.sub files had not been updated when I moved to
11    the latest autoconf.
12    
13    
14    Version 3.5 15-Aug-01
15    ---------------------
16    
17    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
18    had been forgotten.
19    
20    2. By using declared but undefined structures, we can avoid using "void"
21    definitions in pcre.h while keeping the internal definitions of the structures
22    private.
23    
24    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
25    user point of view, this means that both static and shared libraries are built
26    by default, but this can be individually controlled. More of the work of
27    handling this static/shared cases is now inside libtool instead of PCRE's make
28    file.
29    
30    4. The pcretest utility is now installed along with pcregrep because it is
31    useful for users (to test regexs) and by doing this, it automatically gets
32    relinked by libtool. The documentation has been turned into a man page, so
33    there are now .1, .txt, and .html versions in /doc.
34    
35    5. Upgrades to pcregrep:
36       (i)   Added long-form option names like gnu grep.
37       (ii)  Added --help to list all options with an explanatory phrase.
38       (iii) Added -r, --recursive to recurse into sub-directories.
39       (iv)  Added -f, --file to read patterns from a file.
40    
41    6. pcre_exec() was referring to its "code" argument before testing that
42    argument for NULL (and giving an error if it was NULL).
43    
44    7. Upgraded Makefile.in to allow for compiling in a different directory from
45    the source directory.
46    
47    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
48    options bits, the pointer it was passed was to an int instead of to an unsigned
49    long int. This mattered only on 64-bit systems.
50    
51    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
52    generated) instead of pcre.in, which it its source. Also made the same change
53    in several of the .c files.
54    
55    10. A new release of gcc defines printf() as a macro, which broke pcretest
56    because it had an ifdef in the middle of a string argument for printf(). Fixed
57    by using separate calls to printf().
58    
59    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
60    script, to force use of CR or LF instead of \n in the source. On non-Unix
61    systems, the value can be set in config.h.
62    
63    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
64    absolute limit. Changed the text of the error message to make this clear, and
65    likewise updated the man page.
66    
67    13. The limit of 99 on the number of capturing subpatterns has been removed.
68    The new limit is 65535, which I hope will not be a "real" limit.
69    
70    
71    Version 3.4 22-Aug-00
72    ---------------------
73    
74    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
75    
76    2. Diagnose condition (?(0) as an error instead of crashing on matching.
77    
78    
79    Version 3.3 01-Aug-00
80    ---------------------
81    
82    1. If an octal character was given, but the value was greater than \377, it
83    was not getting masked to the least significant bits, as documented. This could
84    lead to crashes in some systems.
85    
86    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
87    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
88    
89    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
90    These just pass their arguments on to (pcre_free)(), but they are provided
91    because some uses of PCRE bind it to non-C systems that can call its functions,
92    but cannot call free() or pcre_free() directly.
93    
94    4. Add "make test" as a synonym for "make check". Corrected some comments in
95    the Makefile.
96    
97    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
98    Makefile.
99    
100    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
101    command called pgrep for grepping around the active processes.
102    
103    7. Added the beginnings of support for UTF-8 character strings.
104    
105    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
106    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
107    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
108    out for the ar command.)
109    
110    
111    Version 3.2 12-May-00
112    ---------------------
113    
114    This is purely a bug fixing release.
115    
116    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
117    of ZA. This was just one example of several cases that could provoke this bug,
118    which was introduced by change 9 of version 2.00. The code for breaking
119    infinite loops after an iteration that matches an empty string was't working
120    correctly.
121    
122    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
123    when matched against abbab (for example). After matching an empty string, it
124    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
125    caused it to match further down the string than it should.
126    
127    3. The code contained an inclusion of sys/types.h. It isn't clear why this
128    was there because it doesn't seem to be needed, and it causes trouble on some
129    systems, as it is not a Standard C header. It has been removed.
130    
131    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
132    were reported on the Macintosh. The changes were from
133    
134      while ((c = *(++ptr)) != 0 && c != '\n');
135    to
136      while ((c = *(++ptr)) != 0 && c != '\n') ;
137    
138    Totally extraordinary, but if that's what it takes...
139    
140    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
141    available. Added HAVE_BCOPY and an autoconf test for it; if neither
142    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
143    assumes the way PCRE uses memmove() (always moving upwards).
144    
145    6. PCRE is being used in one environment where strchr() is not available. There
146    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
147    faster code anyway.
148    
149    
150    Version 3.1 09-Feb-00
151    ---------------------
152    
153    The only change in this release is the fixing of some bugs in Makefile.in for
154    the "install" target:
155    
156    (1) It was failing to install pcreposix.h.
157    
158    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
159    
160    
161    Version 3.0 01-Feb-00
162    ---------------------
163    
164    1. Add support for the /+ modifier to perltest (to output $` like it does in
165    pcretest).
166    
167    2. Add support for the /g modifier to perltest.
168    
169    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
170    matches null strings.
171    
172    4. Fix perltest so that it doesn't do unwanted things when fed an empty
173    pattern. Perl treats empty patterns specially - it reuses the most recent
174    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
175    effect.
176    
177    5. The POSIX interface was broken in that it was just handing over the POSIX
178    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
179    required a bigger vector, with some working space on the end. This means that
180    the POSIX wrapper now has to get and free some memory, and copy the results.
181    
182    6. Added some simple autoconf support, placing the test data and the
183    documentation in separate directories, re-organizing some of the
184    information files, and making it build pcre-config (a GNU standard). Also added
185    libtool support for building PCRE as a shared library, which is now the
186    default.
187    
188    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
189    09 are not valid octal constants. Single digits will be used for minor values
190    less than 10.
191    
192    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
193    existing programs that set these in the POSIX interface can use PCRE without
194    modification.
195    
196    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
197    return all that pcre_info() returns, plus additional data. The pcre_info()
198    function is retained for compatibility, but is considered to be obsolete.
199    
200    10. Added experimental recursion feature (?R) to handle one common case that
201    Perl 5.6 will be able to do with (?p{...}).
202    
203    11. Added support for POSIX character classes like [:alpha:], which Perl is
204    adopting.
205    
206    
207    Version 2.08 31-Aug-99
208    ----------------------
209    
210    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
211    trying to match at the startoffset position, but instead was moving forward to
212    the next newline as if a previous match had failed.
213    
214    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
215    and could get into a loop if a null string was matched other than at the start
216    of the subject.
217    
218    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
219    be distinguished at compile time, and for completeness also added PCRE_DATE.
220    
221    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
222    in GnuWin32 environments.
223    
224    
225    Version 2.07 29-Jul-99
226    ----------------------
227    
228    1. The documentation is now supplied in plain text form and HTML as well as in
229    the form of man page sources.
230    
231    2. C++ compilers don't like assigning (void *) values to other pointer types.
232    In particular this affects malloc(). Although there is no problem in Standard
233    C, I've put in casts to keep C++ compilers happy.
234    
235    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
236    should be (const char *).
237    
238    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
239    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
240    However, I haven't made this a standard facility. The documentation doesn't
241    mention it, and the Makefile doesn't support it.
242    
243    5. The Makefile now contains an "install" target, with editable destinations at
244    the top of the file. The pcretest program is not installed.
245    
246    6. pgrep -V now gives the PCRE version number and date.
247    
248    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
249    causing the entire string to be ignored, instead of just the last character.
250    
251    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
252    non-matching string, it can take a very, very long time, even for strings of
253    quite modest length, because of the nested recursion. PCRE now does better in
254    some of these cases. It does this by remembering the last required literal
255    character in the pattern, and pre-searching the subject to ensure it is present
256    before running the real match. In other words, it applies a heuristic to detect
257    some types of certain failure quickly, and in the above example, if presented
258    with a string that has no trailing " it gives "no match" very quickly.
259    
260    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
261    other alternatives are tried instead.
262    
263    
264    Version 2.06 09-Jun-99
265    ----------------------
266    
267    1. Change pcretest's output for amount of store used to show just the code
268    space, because the remainder (the data block) varies in size between 32-bit and
269    64-bit systems.
270    
271    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
272    start matching at. This allows lookbehinds to work when searching for multiple
273    occurrences in a string.
274    
275    3. Added additional options to pcretest for testing multiple occurrences:
276    
277       /+   outputs the rest of the string that follows a match
278       /g   loops for multiple occurrences, using the new startoffset argument
279       /G   loops for multiple occurrences by passing an incremented pointer
280    
281    4. PCRE wasn't doing the "first character" optimization for patterns starting
282    with \b or \B, though it was doing it for other lookbehind assertions. That is,
283    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
284    the letter 'x'. On long subject strings, this gives a significant speed-up.
285    
286    
287    Version 2.05 21-Apr-99
288    ----------------------
289    
290    1. Changed the type of magic_number from int to long int so that it works
291    properly on 16-bit systems.
292    
293    2. Fixed a bug which caused patterns starting with .* not to work correctly
294    when the subject string contained newline characters. PCRE was assuming
295    anchoring for such patterns in all cases, which is not correct because .* will
296    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
297    DOTALL is set at top level; otherwise it knows that patterns starting with .*
298    must be retried after every newline in the subject.
299    
300    
301    Version 2.04 18-Feb-99
302    ----------------------
303    
304    1. For parenthesized subpatterns with repeats whose minimum was zero, the
305    computation of the store needed to hold the pattern was incorrect (too large).
306    If such patterns were nested a few deep, this could multiply and become a real
307    problem.
308    
309    2. Added /M option to pcretest to show the memory requirement of a specific
310    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
311    
312    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
313    compiled in such a way that the backtracking after subsequent failure was
314    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
315    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
316    
317    
318    Version 2.03 02-Feb-99
319    ----------------------
320    
321    1. Fixed typo and small mistake in man page.
322    
323    2. Added 4th condition (GPL supersedes if conflict) and created separate
324    LICENCE file containing the conditions.
325    
326    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
327    Perl, that is the internal \ allows the delimiter to be included in the
328    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
329    the final delimiter, add \ to the end of the pattern (to test the error).
330    
331    4. Added the convenience functions for extracting substrings after a successful
332    match. Updated pcretest to make it able to test these functions.
333    
334    
335    Version 2.02 14-Jan-99
336    ----------------------
337    
338    1. Initialized the working variables associated with each extraction so that
339    their saving and restoring doesn't refer to uninitialized store.
340    
341    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
342    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
343    fix the problem.
344    
345    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
346    calls, and wasn't printing the correct value for compiling calls. Increased the
347    default value of LOOPREPEAT, and the number of significant figures in the
348    times.
349    
350    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
351    
352    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
353    a building problem on Windows NT with a FAT file system.
354    
355    
356    Version 2.01 21-Oct-98
357    ----------------------
358    
359    1. Changed the API for pcre_compile() to allow for the provision of a pointer
360    to character tables built by pcre_maketables() in the current locale. If NULL
361    is passed, the default tables are used.
362    
363    
364    Version 2.00 24-Sep-98
365    ----------------------
366    
367    1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
368    it any more.
369    
370    2. Allow quantification of (?>) groups, and make it work correctly.
371    
372    3. The first character computation wasn't working for (?>) groups.
373    
374    4. Correct the implementation of \Z (it is permitted to match on the \n at the
375    end of the subject) and add 5.005's \z, which really does match only at the
376    very end of the subject.
377    
378    5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
379    
380    6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
381    DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
382    localized options. All options to pcre_study() were also removed.
383    
384    7. Add other new features from 5.005:
385    
386       $(?<=           positive lookbehind
387       $(?<!           negative lookbehind
388       (?imsx-imsx)    added the unsetting capability
389                       such a setting is global if at outer level; local otherwise
390       (?imsx-imsx:)   non-capturing groups with option setting
391       (?(cond)re|re)  conditional pattern matching
392    
393       A backreference to itself in a repeated group matches the previous
394       captured string.
395    
396    8. General tidying up of studying (both automatic and via "study")
397    consequential on the addition of new assertions.
398    
399    9. As in 5.005, unlimited repeated groups that could match an empty substring
400    are no longer faulted at compile time. Instead, the loop is forcibly broken at
401    runtime if any iteration does actually match an empty substring.
402    
403    10. Include the RunTest script in the distribution.
404    
405    11. Added tests from the Perl 5.005_02 distribution. This showed up a few
406    discrepancies, some of which were old and were also with respect to 5.004. They
407    have now been fixed.
408    
409    
410    Version 1.09 28-Apr-98
411    ----------------------
412    
413    1. A negated single character class followed by a quantifier with a minimum
414    value of one (e.g.  [^x]{1,6}  ) was not compiled correctly. This could lead to
415    program crashes, or just wrong answers. This did not apply to negated classes
416    containing more than one character, or to minima other than one.
417    
418    
419    Version 1.08 27-Mar-98
420    ----------------------
421    
422    1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
423    
424    2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
425    latter must appear before anything that relies on it in the pattern.
426    
427    
428    Version 1.07 16-Feb-98
429    ----------------------
430    
431    1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
432    repeat of a potentially empty string).
433    
434    
435    Version 1.06 23-Jan-98
436    ----------------------
437    
438    1. Added Markus Oberhumer's little patches for C++.
439    
440    2. Literal strings longer than 255 characters were broken.
441    
442    
443    Version 1.05 23-Dec-97
444    ----------------------
445    
446    1. Negated character classes containing more than one character were failing if
447    PCRE_CASELESS was set at run time.
448    
449    
450    Version 1.04 19-Dec-97
451    ----------------------
452    
453    1. Corrected the man page, where some "const" qualifiers had been omitted.
454    
455    2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
456    input syntax.
457    
458    3. Fixed memory leak which occurred when a regex with back references was
459    matched with an offsets vector that wasn't big enough. The temporary memory
460    that is used in this case wasn't being freed if the match failed.
461    
462    4. Tidied pcretest to ensure it frees memory that it gets.
463    
464    5. Temporary memory was being obtained in the case where the passed offsets
465    vector was exactly big enough.
466    
467    6. Corrected definition of offsetof() from change 5 below.
468    
469    7. I had screwed up change 6 below and broken the rules for the use of
470    setjmp(). Now fixed.
471    
472    
473    Version 1.03 18-Dec-97
474    ----------------------
475    
476    1. A erroneous regex with a missing opening parenthesis was correctly
477    diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
478    on some systems.
479    
480    2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
481    it was reported that one broken compiler failed on the former because "code" is
482    also an independent variable.
483    
484    3. The erroneous regex a[]b caused an array overrun reference.
485    
486    4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
487    fail on data ending with that character. (It was going on too far, and checking
488    the next character, typically a binary zero.) This was specific to the
489    optimized code for single-character negative classes.
490    
491    5. Added a contributed patch from the TIN world which does the following:
492    
493      + Add an undef for memmove, in case the the system defines a macro for it.
494    
495      + Add a definition of offsetof(), in case there isn't one. (I don't know
496        the reason behind this - offsetof() is part of the ANSI standard - but
497        it does no harm).
498    
499      + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
500        most of the places where whitespace preceded '#'. I have given up and
501        allowed the remaining 2 cases to be at the margin.
502    
503      + Rename some variables in pcre to eliminate shadowing. This seems very
504        pedantic, but does no harm, of course.
505    
506    6. Moved the call to setjmp() into its own function, to get rid of warnings
507    from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
508    
509    7. Constructs such as \d{8,} were compiling into the equivalent of
510    \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
511    outcome, but in this particular case used more store than had been allocated,
512    which caused the bug to be discovered because it threw up an internal error.
513    
514    8. The debugging code in both pcre and pcretest for outputting the compiled
515    form of a regex was going wrong in the case of back references followed by
516    curly-bracketed repeats.
517    
518    
519    Version 1.02 12-Dec-97
520    ----------------------
521    
522    1. Typos in pcre.3 and comments in the source fixed.
523    
524    2. Applied a contributed patch to get rid of places where it used to remove
525    'const' from variables, and fixed some signed/unsigned and uninitialized
526    variable warnings.
527    
528    3. Added the "runtest" target to Makefile.
529    
530    4. Set default compiler flag to -O2 rather than just -O.
531    
532    
533  Version 1.01 19-Nov-97  Version 1.01 19-Nov-97
534  ----------------------  ----------------------

Legend:
Removed from v.5  
changed lines
  Added in v.55

  ViewVC Help
Powered by ViewVC 1.1.5