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

Diff of /code/trunk/ChangeLog

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

revision 349 by ph10, Wed Jul 2 18:42:11 2008 UTC revision 978 by ph10, Sun Jun 17 16:55:07 2012 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 8.0 02 Jul-08  Version 8.31 02-June-2012
5    -------------------------
6    
7    1.  Fixing a wrong JIT test case and some compiler warnings.
8    
9    2.  Removed a bashism from the RunTest script.
10    
11    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
12        to unsigned type, result still unsigned" that was given by an MS compiler
13        on encountering the code "-sizeof(xxx)".
14    
15    4.  Partial matching support is added to the JIT compiler.
16    
17    5.  Fixed several bugs concerned with partial matching of items that consist
18        of more than one character:
19    
20        (a) /^(..)\1/ did not partially match "aba" because checking references was
21            done on an "all or nothing" basis. This also applied to repeated
22            references.
23    
24        (b) \R did not give a hard partial match if \r was found at the end of the
25            subject.
26    
27        (c) \X did not give a hard partial match after matching one or more
28            characters at the end of the subject.
29    
30        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
31            a partial match for the string "\r".
32    
33        (e) When newline was set to CRLF, the metacharacter "." did not recognize
34            a partial match for a CR character at the end of the subject string.
35    
36    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
37        running pcretest, the text "(JIT)" added to the output whenever JIT is
38        actually used to run the match.
39    
40    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
41        or /S+[+] with a digit between 1 and 7.
42    
43    8.  OP_NOT now supports any UTF character not just single-byte ones.
44    
45    9.  (*MARK) control verb is now supported by the JIT compiler.
46    
47    10. The command "./RunTest list" lists the available tests without actually
48        running any of them. (Because I keep forgetting what they all are.)
49    
50    11. Add PCRE_INFO_MAXLOOKBEHIND.
51    
52    12. Applied a (slightly modified) user-supplied patch that improves performance
53        when the heap is used for recursion (compiled with --disable-stack-for-
54        recursion). Instead of malloc and free for each heap frame each time a
55        logical recursion happens, frames are retained on a chain and re-used where
56        possible. This sometimes gives as much as 30% improvement.
57    
58    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
59        call.
60    
61    14. As documented, (*COMMIT) is now confined to within a positive assertion.
62    
63    15. It is now possible to link pcretest with libedit as an alternative to
64        libreadline.
65    
66    16. (*COMMIT) control verb is now supported by the JIT compiler.
67    
68    17. The Unicode data tables have been updated to Unicode 6.1.0.
69    
70    18. Added --file-list option to pcregrep.
71    
72    19. Added binary file support to pcregrep, including the -a, --binary-files,
73        -I, and --text options.
74    
75    20. The madvise function is renamed for posix_madvise for QNX compatibility
76        reasons. Fixed by Giuseppe D'Angelo.
77    
78    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
79    
80    22. Changed the option for creating symbolic links for 16-bit man pages from
81        -s to -sf so that re-installing does not cause issues.
82    
83    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
84    
85    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
86        matches in certain environments (the workspace was not being correctly
87        retained). Also added to pcre_dfa_exec() a simple plausibility check on
88        some of the workspace data at the beginning of a restart.
89    
90    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
91        was not doing so when it should - probably a typo introduced by SVN 528
92        (change 8.10/14).
93    
94    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
95        \w+ when the character tables indicated that \x{c4} was a word character.
96        There were several related cases, all because the tests for doing a table
97        lookup were testing for characters less than 127 instead of 255.
98    
99    27. If a pattern contains capturing parentheses that are not used in a match,
100        their slots in the ovector are set to -1. For those that are higher than
101        any matched groups, this happens at the end of processing. In the case when
102        there were back references that the ovector was too small to contain
103        (causing temporary malloc'd memory to be used during matching), and the
104        highest capturing number was not used, memory off the end of the ovector
105        was incorrectly being set to -1. (It was using the size of the temporary
106        memory instead of the true size.)
107    
108    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
109        ovector size, it uses memory at the end of the block that it has got.
110    
111    29. Check for an overlong MARK name and give an error at compile time. The
112        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
113    
114    30. JIT compiler update.
115    
116    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
117        Rill for the patch.
118    
119    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
120    
121    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
122    
123    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
124        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
125        was enabled.
126    
127    35. Improve JIT code generation for greedy plus quantifier.
128    
129    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
130        1 to "aa" instead of to an empty string. The bug affected repeated groups
131        that could potentially match an empty string.
132    
133    37. Optimizing single character iterators in JIT.
134    
135    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
136        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
137        codepoints that are too big for the mode are faulted, and in a UTF mode,
138        disallowed codepoints are also faulted.
139    
140    
141    Version 8.30 04-February-2012
142    -----------------------------
143    
144    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
145        name is defined in ctype.h.
146    
147    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
148        only in quite long subpatterns.
149    
150    3.  Removed the function pcre_info(), which has been obsolete and deprecated
151        since it was replaced by pcre_fullinfo() in February 2000.
152    
153    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
154        match a (*MARK), and the match failed at the start of the subject, a
155        reference to memory before the start of the subject could occur. This bug
156        was introduced by fix 17 of release 8.21.
157    
158    5.  A reference to an unset group with zero minimum repetition was giving
159        totally wrong answers (in non-JavaScript-compatibility mode). For example,
160        /(another)?(\1?)test/ matched against "hello world test". This bug was
161        introduced in release 8.13.
162    
163    6.  Add support for 16-bit character strings (a large amount of work involving
164        many changes and refactorings).
165    
166    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
167        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
168        from a file.
169    
170    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
171        rounding is not applied in this particular case).
172    
173    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
174        if they appear, or are escaped, in patterns.
175    
176    10. Get rid of a number of -Wunused-but-set-variable warnings.
177    
178    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
179        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
180        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
181        also returns the mark "x". This bug applied to capturing parentheses,
182        non-capturing parentheses, and atomic parentheses. It also applied to some
183        assertions.
184    
185    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
186        information out of configure.ac instead of relying on pcre.h.generic, which
187        is not stored in the repository.
188    
189    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
190        -lreadline.
191    
192    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
193    
194    15. Applied Graycode's patch to put the top-level frame on the stack rather
195        than the heap when not using the stack for recursion. This gives a
196        performance improvement in many cases when recursion is not deep.
197    
198    16. Experimental code added to "pcretest -C" to output the stack frame size.
199    
200    
201    Version 8.21 12-Dec-2011
202    ------------------------
203    
204    1.  Updating the JIT compiler.
205    
206    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
207        are added as well.
208    
209    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
210        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
211        calling _pcre_jit_exec. Some extra comments are added.
212    
213    4.  (*MARK) settings inside atomic groups that do not contain any capturing
214        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
215        was introduced by change 18 for 8.20.
216    
217    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
218        ECMA-262 standard.
219    
220    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
221        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
222        This bug was probably introduced by change 9 of 8.13.
223    
224    7.  While fixing 6 above, I noticed that a number of other items were being
225        incorrectly rejected as "not fixed length". This arose partly because newer
226        opcodes had not been added to the fixed-length checking code. I have (a)
227        corrected the bug and added tests for these items, and (b) arranged for an
228        error to occur if an unknown opcode is encountered while checking for fixed
229        length instead of just assuming "not fixed length". The items that were
230        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
231        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
232        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
233    
234    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
235        being incorrectly compiled and would have given unpredicatble results.
236    
237    9.  A possessively repeated subpattern with minimum repeat count greater than
238        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
239        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
240        the first (A) could occur when it should not.
241    
242    10. Add a cast and remove a redundant test from the code.
243    
244    11. JIT should use pcre_malloc/pcre_free for allocation.
245    
246    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
247        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
248        is anything other than /usr, -L is still shown).
249    
250    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
251    
252    14. Perl does not support \N without a following name in a [] class; PCRE now
253        also gives an error.
254    
255    15. If a forward reference was repeated with an upper limit of around 2000,
256        it caused the error "internal error: overran compiling workspace". The
257        maximum number of forward references (including repeats) was limited by the
258        internal workspace, and dependent on the LINK_SIZE. The code has been
259        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
260        the default depends on LINK_SIZE. There is a new upper limit (for safety)
261        of around 200,000 forward references. While doing this, I also speeded up
262        the filling in of repeated forward references.
263    
264    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
265        incorrectly expecting the subject to contain another "a" after the start.
266    
267    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
268        in the match, the SKIP should be ignored. This was not happening; instead
269        the SKIP was being treated as NOMATCH. For patterns such as
270        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
271        tested.
272    
273    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
274        now much more compatible with Perl, in particular in cases where the result
275        is a non-match for a non-anchored pattern. For example, if
276        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
277        "m", where previously it did not return a name. A side effect of this
278        change is that for partial matches, the last encountered mark name is
279        returned, as for non matches. A number of tests that were previously not
280        Perl-compatible have been moved into the Perl-compatible test files. The
281        refactoring has had the pleasing side effect of removing one argument from
282        the match() function, thus reducing its stack requirements.
283    
284    19. If the /S+ option was used in pcretest to study a pattern using JIT,
285        subsequent uses of /S (without +) incorrectly behaved like /S+.
286    
287    21. Retrieve executable code size support for the JIT compiler and fixing
288        some warnings.
289    
290    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
291        not work when the shorter character appeared right at the end of the
292        subject string.
293    
294    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
295        systems.
296    
297    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
298        output it when the /M option is used in pcretest.
299    
300    25. The CheckMan script was not being included in the distribution. Also, added
301        an explicit "perl" to run Perl scripts from the PrepareRelease script
302        because this is reportedly needed in Windows.
303    
304    26. If study data was being save in a file and studying had not found a set of
305        "starts with" bytes for the pattern, the data written to the file (though
306        never used) was taken from uninitialized memory and so caused valgrind to
307        complain.
308    
309    27. Updated RunTest.bat as provided by Sheri Pierce.
310    
311    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
312    
313    29. Computation of memory usage for the table of capturing group names was
314        giving an unnecessarily large value.
315    
316    
317    Version 8.20 21-Oct-2011
318    ------------------------
319    
320    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
321        a POSIX class. After further experiments with Perl, which convinced me that
322        Perl has bugs and confusions, a closing square bracket is no longer allowed
323        in a POSIX name. This bug also affected patterns with classes that started
324        with full stops.
325    
326    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
327        captured substring, but while checking the failing first alternative,
328        substring 1 is temporarily captured. If the output vector supplied to
329        pcre_exec() was not big enough for this capture, the yield of the function
330        was still zero ("insufficient space for captured substrings"). This cannot
331        be totally fixed without adding another stack variable, which seems a lot
332        of expense for a edge case. However, I have improved the situation in cases
333        such as /(a)(b)x|abc/ matched against "abc", where the return code
334        indicates that fewer than the maximum number of slots in the ovector have
335        been set.
336    
337    3.  Related to (2) above: when there are more back references in a pattern than
338        slots in the output vector, pcre_exec() uses temporary memory during
339        matching, and copies in the captures as far as possible afterwards. It was
340        using the entire output vector, but this conflicts with the specification
341        that only 2/3 is used for passing back captured substrings. Now it uses
342        only the first 2/3, for compatibility. This is, of course, another edge
343        case.
344    
345    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
346        main code base, and can be used by building with --enable-jit. When this is
347        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
348        runtime --no-jit option is given.
349    
350    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
351        ovector, the return from the function was zero, implying that there were
352        other matches that did not fit. The correct "exactly full" value is now
353        returned.
354    
355    6.  If a subpattern that was called recursively or as a subroutine contained
356        (*PRUNE) or any other control that caused it to give a non-standard return,
357        invalid errors such as "Error -26 (nested recursion at the same subject
358        position)" or even infinite loops could occur.
359    
360    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
361        computing the minimum length on reaching *ACCEPT, and so ended up with the
362        wrong value of 1 rather than 0. Further investigation indicates that
363        computing a minimum subject length in the presence of *ACCEPT is difficult
364        (think back references, subroutine calls), and so I have changed the code
365        so that no minimum is registered for a pattern that contains *ACCEPT.
366    
367    8.  If (*THEN) was present in the first (true) branch of a conditional group,
368        it was not handled as intended. [But see 16 below.]
369    
370    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
371        Sheri Pierce.
372    
373    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
374        the first byte in a match must be "a".
375    
376    11. Change 17 for 8.13 increased the recursion depth for patterns like
377        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
378        pattern contains any instances of (*THEN). If it does not, the old
379        optimizations are restored. It would be nice to do this on a per-group
380        basis, but at the moment that is not feasible.
381    
382    12. In some environments, the output of pcretest -C is CRLF terminated. This
383        broke RunTest's code that checks for the link size. A single white space
384        character after the value is now allowed for.
385    
386    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
387        For "fr", it uses the Windows-specific input and output files.
388    
389    14. If (*THEN) appeared in a group that was called recursively or as a
390        subroutine, it did not work as intended. [But see next item.]
391    
392    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
393        pattern fragments (but not containing any | characters). If A and B are
394        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
395        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
396        D. In other words, Perl considers parentheses that do not contain any |
397        characters to be part of a surrounding alternative, whereas PCRE was
398        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
399        differently. PCRE now behaves in the same way as Perl, except in the case
400        of subroutine/recursion calls such as (?1) which have in any case always
401        been different (but PCRE had them first :-).
402    
403    16. Related to 15 above: Perl does not treat the | in a conditional group as
404        creating alternatives. Such a group is treated in the same way as an
405        ordinary group without any | characters when processing (*THEN). PCRE has
406        been changed to match Perl's behaviour.
407    
408    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
409        RunGrepTest script failed.
410    
411    18. Change 22 for version 13 caused atomic groups to use more stack. This is
412        inevitable for groups that contain captures, but it can lead to a lot of
413        stack use in large patterns. The old behaviour has been restored for atomic
414        groups that do not contain any capturing parentheses.
415    
416    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
417        suppress the check for a minimum subject length at run time. (If it was
418        given to pcre_exec() or pcre_dfa_exec() it did work.)
419    
420    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
421        fail to work when decoding hex characters in data strings in EBCDIC
422        environments.
423    
424    21. It appears that in at least one Mac OS environment, the isxdigit() function
425        is implemented as a macro that evaluates to its argument more than once,
426        contravening the C 90 Standard (I haven't checked a later standard). There
427        was an instance in pcretest which caused it to go wrong when processing
428        \x{...} escapes in subject strings. The has been rewritten to avoid using
429        things like p++ in the argument of isxdigit().
430    
431    
432    Version 8.13 16-Aug-2011
433    ------------------------
434    
435    1.  The Unicode data tables have been updated to Unicode 6.0.0.
436    
437    2.  Two minor typos in pcre_internal.h have been fixed.
438    
439    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
440        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
441        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
442    
443    4.  There were a number of related bugs in the code for matching backrefences
444        caselessly in UTF-8 mode when codes for the characters concerned were
445        different numbers of bytes. For example, U+023A and U+2C65 are an upper
446        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
447        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
448        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
449        2-byte code at the end of the subject (it thought there wasn't enough data
450        left).
451    
452    5.  Comprehensive information about what went wrong is now returned by
453        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
454        as the output vector has at least 2 elements. The offset of the start of
455        the failing character and a reason code are placed in the vector.
456    
457    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
458        now returned is for the first byte of the failing character, instead of the
459        last byte inspected. This is an incompatible change, but I hope it is small
460        enough not to be a problem. It makes the returned offset consistent with
461        pcre_exec() and pcre_dfa_exec().
462    
463    7.  pcretest now gives a text phrase as well as the error number when
464        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
465        failure, the offset and reason code are output.
466    
467    8.  When \R was used with a maximizing quantifier it failed to skip backwards
468        over a \r\n pair if the subsequent match failed. Instead, it just skipped
469        back over a single character (\n). This seems wrong (because it treated the
470        two characters as a single entity when going forwards), conflicts with the
471        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
472        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
473        has been changed.
474    
475    9.  Some internal refactoring has changed the processing so that the handling
476        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
477        time (the PCRE_DOTALL option was changed this way some time ago: version
478        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
479        which was always a bit of a fudge. It also means that there is one less
480        argument for the match() function, which reduces its stack requirements
481        slightly. This change also fixes an incompatibility with Perl: the pattern
482        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
483    
484    10. More internal refactoring has drastically reduced the number of recursive
485        calls to match() for possessively repeated groups such as (abc)++ when
486        using pcre_exec().
487    
488    11. While implementing 10, a number of bugs in the handling of groups were
489        discovered and fixed:
490    
491        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
492        (a|)*(?1) gave a compile-time internal error.
493        ((a|)+)+  did not notice that the outer group could match an empty string.
494        (^a|^)+   was not marked as anchored.
495        (.*a|.*)+ was not marked as matching at start or after a newline.
496    
497    12. Yet more internal refactoring has removed another argument from the match()
498        function. Special calls to this function are now indicated by setting a
499        value in a variable in the "match data" data block.
500    
501    13. Be more explicit in pcre_study() instead of relying on "default" for
502        opcodes that mean there is no starting character; this means that when new
503        ones are added and accidentally left out of pcre_study(), testing should
504        pick them up.
505    
506    14. The -s option of pcretest has been documented for ages as being an old
507        synonym of -m (show memory usage). I have changed it to mean "force study
508        for every regex", that is, assume /S for every regex. This is similar to -i
509        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
510        using it. It makes it easier to run collections of tests with and without
511        study enabled, and thereby test pcre_study() more easily. All the standard
512        tests are now run with and without -s (but some patterns can be marked as
513        "never study" - see 20 below).
514    
515    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
516        restoration of the capturing data to the outer values was not happening
517        correctly.
518    
519    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
520        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
521        pattern had matched an empty string, and so incorrectly returned a no
522        match.
523    
524    17. There was optimizing code for the last branch of non-capturing parentheses,
525        and also for the obeyed branch of a conditional subexpression, which used
526        tail recursion to cut down on stack usage. Unfortunately, now that there is
527        the possibility of (*THEN) occurring in these branches, tail recursion is
528        no longer possible because the return has to be checked for (*THEN). These
529        two optimizations have therefore been removed. [But see 8.20/11 above.]
530    
531    18. If a pattern containing \R was studied, it was assumed that \R always
532        matched two bytes, thus causing the minimum subject length to be
533        incorrectly computed because \R can also match just one byte.
534    
535    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
536        was incorrectly computed.
537    
538    20. If /S is present twice on a test pattern in pcretest input, it now
539        *disables* studying, thereby overriding the use of -s on the command line
540        (see 14 above). This is necessary for one or two tests to keep the output
541        identical in both cases.
542    
543    21. When (*ACCEPT) was used in an assertion that matched an empty string and
544        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
545    
546    22. When an atomic group that contained a capturing parenthesis was
547        successfully matched, but the branch in which it appeared failed, the
548        capturing was not being forgotten if a higher numbered group was later
549        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
550        group 1 to "a", when in fact it should be unset. This applied to multi-
551        branched capturing and non-capturing groups, repeated or not, and also to
552        positive assertions (capturing in negative assertions does not happen
553        in PCRE) and also to nested atomic groups.
554    
555    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
556        subject after a captured substring, to make it easier to tell which of a
557        number of identical substrings has been captured.
558    
559    24. The way atomic groups are processed by pcre_exec() has been changed so that
560        if they are repeated, backtracking one repetition now resets captured
561        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
562        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
563        "aaa". Previously, it would have been "a". As part of this code
564        refactoring, the way recursive calls are handled has also been changed.
565    
566    25. If an assertion condition captured any substrings, they were not passed
567        back unless some other capturing happened later. For example, if
568        (?(?=(a))a) was matched against "a", no capturing was returned.
569    
570    26. When studying a pattern that contained subroutine calls or assertions,
571        the code for finding the minimum length of a possible match was handling
572        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
573        group 1 called group 2 while simultaneously a separate group 2 called group
574        1). A stack overflow occurred in this case. I have fixed this by limiting
575        the recursion depth to 10.
576    
577    27. Updated RunTest.bat in the distribution to the version supplied by Tom
578        Fortmann. This supports explicit test numbers on the command line, and has
579        argument validation and error reporting.
580    
581    28. An instance of \X with an unlimited repeat could fail if at any point the
582        first character it looked at was a mark character.
583    
584    29. Some minor code refactoring concerning Unicode properties and scripts
585        should reduce the stack requirement of match() slightly.
586    
587    30. Added the '=' option to pcretest to check the setting of unused capturing
588        slots at the end of the pattern, which are documented as being -1, but are
589        not included in the return count.
590    
591    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
592        compiled something random. Now it gives a compile-time error (as does
593        Perl).
594    
595    32. A *MARK encountered during the processing of a positive assertion is now
596        recorded and passed back (compatible with Perl).
597    
598    33. If --only-matching or --colour was set on a pcregrep call whose pattern
599        had alternative anchored branches, the search for a second match in a line
600        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
601        matched the line "0102" twice. The same bug affected patterns that started
602        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
603        twice.
604    
605    34. Previously, PCRE did not allow quantification of assertions. However, Perl
606        does, and because of capturing effects, quantifying parenthesized
607        assertions may at times be useful. Quantifiers are now allowed for
608        parenthesized assertions.
609    
610    35. A minor code tidy in pcre_compile() when checking options for \R usage.
611    
612    36. \g was being checked for fancy things in a character class, when it should
613        just be a literal "g".
614    
615    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
616        appearance of a nested POSIX class supersedes an apparent external class.
617        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
618        unescaped square brackets may also appear as part of class names. For
619        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
620        more like Perl. (But see 8.20/1 above.)
621    
622    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
623        was because it thought it was \N{name}, which is not supported).
624    
625    39. Add minix to OS list not supporting the -S option in pcretest.
626    
627    40. PCRE tries to detect cases of infinite recursion at compile time, but it
628        cannot analyze patterns in sufficient detail to catch mutual recursions
629        such as ((?1))((?2)). There is now a runtime test that gives an error if a
630        subgroup is called recursively as a subpattern for a second time at the
631        same position in the subject string. In previous releases this might have
632        been caught by the recursion limit, or it might have run out of stack.
633    
634    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
635        happen only once. PCRE was, however incorrectly giving a compile time error
636        "recursive call could loop indefinitely" because it cannot analyze the
637        pattern in sufficient detail. The compile time test no longer happens when
638        PCRE is compiling a conditional subpattern, but actual runaway loops are
639        now caught at runtime (see 40 above).
640    
641    42. It seems that Perl allows any characters other than a closing parenthesis
642        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
643        has been changed to be the same.
644    
645    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
646        as not to get warnings when autogen.sh is called. Also changed
647        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
648    
649    44. To help people who use pcregrep to scan files containing exceedingly long
650        lines, the following changes have been made:
651    
652        (a) The default value of the buffer size parameter has been increased from
653            8K to 20K. (The actual buffer used is three times this size.)
654    
655        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
656            PCRE is built.
657    
658        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
659            to be set at run time.
660    
661        (d) Numerical values in pcregrep options can be followed by K or M, for
662            example --buffer-size=50K.
663    
664        (e) If a line being scanned overflows pcregrep's buffer, an error is now
665            given and the return code is set to 2.
666    
667    45. Add a pointer to the latest mark to the callout data block.
668    
669    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
670        partial match of an empty string instead of no match. This was specific to
671        the use of ".".
672    
673    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
674        complete match instead of a partial match. This bug was dependent on both
675        the PCRE_UTF8 and PCRE_DOTALL options being set.
676    
677    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
678        starting byte set, because \b was not being ignored.
679    
680    
681    Version 8.12 15-Jan-2011
682    ------------------------
683    
684    1.  Fixed some typos in the markup of the man pages, and wrote a script that
685        checks for such things as part of the documentation building process.
686    
687    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
688        --match-limit and --recursion-limit options (added for 8.11). In
689        particular, this made one of the standard tests fail. (The integer value
690        went into the wrong half of a long int.)
691    
692    3.  If the --colour option was given to pcregrep with -v (invert match), it
693        did strange things, either producing crazy output, or crashing. It should,
694        of course, ignore a request for colour when reporting lines that do not
695        match.
696    
697    4.  Another pcregrep bug caused similar problems if --colour was specified with
698        -M (multiline) and the pattern match finished with a line ending.
699    
700    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
701        matched in multiline mode, the following line was shown as part of the
702        match. This seems wrong, so I have changed it.
703    
704    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
705        the check for further matches in the same line (so they could be coloured)
706        to overrun the end of the current line. If another match was found, it was
707        incorrectly shown (and then shown again when found in the next line).
708    
709    7.  If pcregrep was compiled under Windows, there was a reference to the
710        function pcregrep_exit() before it was defined. I am assuming this was
711        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
712        reported by a user. I've moved the definition above the reference.
713    
714    
715    Version 8.11 10-Dec-2010
716    ------------------------
717    
718    1.  (*THEN) was not working properly if there were untried alternatives prior
719        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
720        backtracked to try for "b" instead of moving to the next alternative branch
721        at the same level (in this case, to look for "c"). The Perl documentation
722        is clear that when (*THEN) is backtracked onto, it goes to the "next
723        alternative in the innermost enclosing group".
724    
725    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
726        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
727        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
728        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
729        (*THEN).
730    
731    3.  If \s appeared in a character class, it removed the VT character from
732        the class, even if it had been included by some previous item, for example
733        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
734        of \s, but is part of the POSIX "space" class.)
735    
736    4.  A partial match never returns an empty string (because you can always
737        match an empty string at the end of the subject); however the checking for
738        an empty string was starting at the "start of match" point. This has been
739        changed to the "earliest inspected character" point, because the returned
740        data for a partial match starts at this character. This means that, for
741        example, /(?<=abc)def/ gives a partial match for the subject "abc"
742        (previously it gave "no match").
743    
744    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
745        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
746        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
747        has an implication that the given string is incomplete (because a partial
748        match is preferred over a full match). For this reason, these items now
749        give a partial match in this situation. [Aside: previously, the one case
750        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
751        match rather than a full match, which was wrong by the old rules, but is
752        now correct.]
753    
754    6.  There was a bug in the handling of #-introduced comments, recognized when
755        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
756        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
757        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
758        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
759        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
760        places in pcre_compile().
761    
762    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
763        comments when looking ahead for named forward references to subpatterns,
764        the only newline sequence it recognized was NL. It now handles newlines
765        according to the set newline convention.
766    
767    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
768        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
769        cater for a lack of strerror(). These oversights have been fixed.
770    
771    9.  Added --match-limit and --recursion-limit to pcregrep.
772    
773    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
774    
775    11. When the -o option was used, pcregrep was setting a return code of 1, even
776        when matches were found, and --line-buffered was not being honoured.
777    
778    12. Added an optional parentheses number to the -o and --only-matching options
779        of pcregrep.
780    
781    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
782        can match an empty string. The code to do it in pcretest and pcredemo
783        needed fixing:
784    
785        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
786            only one byte after an empty string match just before CRLF (this case
787            just got forgotten; "any" and "anycrlf" were OK).
788    
789        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
790            mode when an empty string match preceded an ASCII character followed by
791            a non-ASCII character. (The code for advancing by one character rather
792            than one byte was nonsense.)
793    
794        (c) The pcredemo.c sample program did not have any code at all to handle
795            the cases when CRLF is a valid newline sequence.
796    
797    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
798        as a starting offset was within the subject string. There is now a new
799        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
800        negative or greater than the length of the string. In order to test this,
801        pcretest is extended to allow the setting of negative starting offsets.
802    
803    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
804        starting offset points to the beginning of a UTF-8 character was
805        unnecessarily clumsy. I tidied it up.
806    
807    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
808        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
809    
810    17. Nobody had reported that the --include_dir option, which was added in
811        release 7.7 should have been called --include-dir (hyphen, not underscore)
812        for compatibility with GNU grep. I have changed it to --include-dir, but
813        left --include_dir as an undocumented synonym, and the same for
814        --exclude-dir, though that is not available in GNU grep, at least as of
815        release 2.5.4.
816    
817    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
818        characters from a string of bytes) have been redefined so as not to use
819        loops, in order to improve performance in some environments. At the same
820        time, I abstracted some of the common code into auxiliary macros to save
821        repetition (this should not affect the compiled code).
822    
823    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
824        compile-time error is now given if \c is not followed by an ASCII
825        character, that is, a byte less than 128. (In EBCDIC mode, the code is
826        different, and any byte value is allowed.)
827    
828    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
829        START_OPTIMIZE option, which is now allowed at compile time - but just
830        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
831        to pcregrep and other applications that have no direct access to PCRE
832        options. The new /Y option in pcretest sets this option when calling
833        pcre_compile().
834    
835    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
836        back references. Groups containing recursive back references were forced to
837        be atomic by that change, but in the case of named groups, the amount of
838        memory required was incorrectly computed, leading to "Failed: internal
839        error: code overflow". This has been fixed.
840    
841    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
842        pcretest.c, to avoid build problems in some Borland environments.
843    
844    
845    Version 8.10 25-Jun-2010
846    ------------------------
847    
848    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
849        THEN.
850    
851    2.  (*ACCEPT) was not working when inside an atomic group.
852    
853    3.  Inside a character class, \B is treated as a literal by default, but
854        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
855        causes the error). The code is unchanged, but I tidied the documentation.
856    
857    4.  Inside a character class, PCRE always treated \R and \X as literals,
858        whereas Perl faults them if its -w option is set. I have changed PCRE so
859        that it faults them when PCRE_EXTRA is set.
860    
861    5.  Added support for \N, which always matches any character other than
862        newline. (It is the same as "." when PCRE_DOTALL is not set.)
863    
864    6.  When compiling pcregrep with newer versions of gcc which may have
865        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
866        declared with attribute warn_unused_result" were given. Just casting the
867        result to (void) does not stop the warnings; a more elaborate fudge is
868        needed. I've used a macro to implement this.
869    
870    7.  Minor change to pcretest.c to avoid a compiler warning.
871    
872    8.  Added four artifical Unicode properties to help with an option to make
873        \s etc use properties (see next item). The new properties are: Xan
874        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
875    
876    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
877        use Unicode properties. (*UCP) at the start of a pattern can be used to set
878        this option. Modified pcretest to add /W to test this facility. Added
879        REG_UCP to make it available via the POSIX interface.
880    
881    10. Added --line-buffered to pcregrep.
882    
883    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
884        studied, and the match started with a letter with a code point greater than
885        127 whose first byte was different to the first byte of the other case of
886        the letter, the other case of this starting letter was not recognized
887        (#976).
888    
889    12. If a pattern that was studied started with a repeated Unicode property
890        test, for example, \p{Nd}+, there was the theoretical possibility of
891        setting up an incorrect bitmap of starting bytes, but fortunately it could
892        not have actually happened in practice until change 8 above was made (it
893        added property types that matched character-matching opcodes).
894    
895    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
896        possible starting bytes for non-anchored patterns.
897    
898    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
899        \R, and also a number of cases that involve Unicode properties, both
900        explicit and implicit when PCRE_UCP is set.
901    
902    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
903        input, it could crash or give wrong results if characters with values
904        greater than 0xc0 were present in the subject string. (Detail: it assumed
905        UTF-8 input when processing these items.)
906    
907    16. Added a lot of (int) casts to avoid compiler warnings in systems where
908        size_t is 64-bit (#991).
909    
910    17. Added a check for running out of memory when PCRE is compiled with
911        --disable-stack-for-recursion (#990).
912    
913    18. If the last data line in a file for pcretest does not have a newline on
914        the end, a newline was missing in the output.
915    
916    19. The default pcre_chartables.c file recognizes only ASCII characters (values
917        less than 128) in its various bitmaps. However, there is a facility for
918        generating tables according to the current locale when PCRE is compiled. It
919        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
920        characters, are recognized by isspace() and therefore were getting set in
921        these tables, and indeed these tables seem to approximate to ISO 8859. This
922        caused a problem in UTF-8 mode when pcre_study() was used to create a list
923        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
924        which of course cannot start UTF-8 characters. I have changed the code so
925        that only real ASCII characters (less than 128) and the correct starting
926        bytes for UTF-8 encodings are set for characters greater than 127 when in
927        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
928        altogether.)
929    
930    20. Added the /T option to pcretest so as to be able to run tests with non-
931        standard character tables, thus making it possible to include the tests
932        used for 19 above in the standard set of tests.
933    
934    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
935        reference to a subpattern the other side of a comment that contains an
936        opening parenthesis caused either an internal compiling error, or a
937        reference to the wrong subpattern.
938    
939    
940    Version 8.02 19-Mar-2010
941    ------------------------
942    
943    1.  The Unicode data tables have been updated to Unicode 5.2.0.
944    
945    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
946        configured.
947    
948    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
949        original author of that file, following a query about its status.
950    
951    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
952        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
953    
954    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
955        quantifier applied to a forward-referencing subroutine call, could compile
956        incorrect code or give the error "internal error: previously-checked
957        referenced subpattern not found".
958    
959    6.  Both MS Visual Studio and Symbian OS have problems with initializing
960        variables to point to external functions. For these systems, therefore,
961        pcre_malloc etc. are now initialized to local functions that call the
962        relevant global functions.
963    
964    7.  There were two entries missing in the vectors called coptable and poptable
965        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
966        I've fixed the data, and added a kludgy way of testing at compile time that
967        the lengths are correct (equal to the number of opcodes).
968    
969    8.  Following on from 7, I added a similar kludge to check the length of the
970        eint vector in pcreposix.c.
971    
972    9.  Error texts for pcre_compile() are held as one long string to avoid too
973        much relocation at load time. To find a text, the string is searched,
974        counting zeros. There was no check for running off the end of the string,
975        which could happen if a new error number was added without updating the
976        string.
977    
978    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
979    
980    11. \K was not working if it appeared in an atomic group or in a group that
981        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
982        \K is "not well defined" if used in an assertion. PCRE now accepts it if
983        the assertion is positive, but not if it is negative.
984    
985    12. Change 11 fortuitously reduced the size of the stack frame used in the
986        "match()" function of pcre_exec.c by one pointer. Forthcoming
987        implementation of support for (*MARK) will need an extra pointer on the
988        stack; I have reserved it now, so that the stack frame size does not
989        decrease.
990    
991    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
992        item in branch that calls a recursion is a subroutine call - as in the
993        second branch in the above example - was incorrectly given the compile-
994        time error "recursive call could loop indefinitely" because pcre_compile()
995        was not correctly checking the subroutine for matching a non-empty string.
996    
997    14. The checks for overrunning compiling workspace could trigger after an
998        overrun had occurred. This is a "should never occur" error, but it can be
999        triggered by pathological patterns such as hundreds of nested parentheses.
1000        The checks now trigger 100 bytes before the end of the workspace.
1001    
1002    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1003    
1004    
1005    Version 8.01 19-Jan-2010
1006    ------------------------
1007    
1008    1.  If a pattern contained a conditional subpattern with only one branch (in
1009        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1010        computed the wrong minimum data length (which is of course zero for such
1011        subpatterns). This could cause incorrect "no match" results.
1012    
1013    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1014        the pattern is reset in the first branch, pcre_compile() failed with
1015        "internal error: code overflow at offset...". This happened only when
1016        the reset was to the original external option setting. (An optimization
1017        abstracts leading options settings into an external setting, which was the
1018        cause of this.)
1019    
1020    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1021        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1022        assertion pattern did not match (meaning that the assertion was true), it
1023        was incorrectly treated as false if the SKIP had been reached during the
1024        matching. This also applied to assertions used as conditions.
1025    
1026    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1027        assertion subpattern, including such a pattern used as a condition,
1028        unpredictable results occurred, instead of the error return
1029        PCRE_ERROR_DFA_UITEM.
1030    
1031    5.  The C++ GlobalReplace function was not working like Perl for the special
1032        situation when an empty string is matched. It now does the fancy magic
1033        stuff that is necessary.
1034    
1035    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1036        removed. (These were left over from very, very early versions of PCRE.)
1037    
1038    7.  Some cosmetic changes to the code to make life easier when compiling it
1039        as part of something else:
1040    
1041        (a) Change DEBUG to PCRE_DEBUG.
1042    
1043        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1044            called "current" as "current_branch", to prevent a collision with the
1045            Linux macro when compiled as a kernel module.
1046    
1047        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1048            prevent a collision with the Linux macro when compiled as a kernel
1049            module.
1050    
1051    8.  In pcre_compile() there are some checks for integer overflows that used to
1052        cast potentially large values to (double). This has been changed to that
1053        when building, a check for int64_t is made, and if it is found, it is used
1054        instead, thus avoiding the use of floating point arithmetic. (There is no
1055        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1056        double.
1057    
1058    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1059        2005 (difference between two addresses compared to an unsigned value).
1060    
1061    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1062        custom one, because of the following reported problem in Windows:
1063    
1064          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1065              under Win32.
1066          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1067              therefore missing the function definition.
1068          - The compiler thus generates a "C" signature for the test function.
1069          - The linker fails to find the "C" function.
1070          - PCRE fails to configure if asked to do so against libbz2.
1071    
1072    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1073        messages were output:
1074    
1075          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1076          rerunning libtoolize, to keep the correct libtool macros in-tree.
1077          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1078    
1079        I have done both of these things.
1080    
1081    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1082        most of the time, it *can* run out if it is given a pattern that contains a
1083        runaway infinite recursion. I updated the discussion in the pcrestack man
1084        page.
1085    
1086    13. Now that we have gone to the x.xx style of version numbers, the minor
1087        version may start with zero. Using 08 or 09 is a bad idea because users
1088        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1089        interpreted as invalid octal numbers. I've updated the previous comment in
1090        configure.ac, and also added a check that gives an error if 08 or 09 are
1091        used.
1092    
1093    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1094        causing partial matching to fail when the end of the subject matched \W
1095        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1096    
1097    15. There were some discrepancies between the declarations in pcre_internal.h
1098        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1099        their definitions. The declarations used "const uschar *" and the
1100        definitions used USPTR. Even though USPTR is normally defined as "const
1101        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1102        reported that: "This difference in casting confuses some C++ compilers, for
1103        example, SunCC recognizes above declarations as different functions and
1104        generates broken code for hbpcre." I have changed the declarations to use
1105        USPTR.
1106    
1107    16. GNU libtool is named differently on some systems. The autogen.sh script now
1108        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1109        (FreeBSD).
1110    
1111    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1112        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1113        comment: "Figure out how to create a longlong from a string: strtoll and
1114        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1115        instance, but it only takes 2 args instead of 3!"
1116    
1117    18. A subtle bug concerned with back references has been fixed by a change of
1118        specification, with a corresponding code fix. A pattern such as
1119        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1120        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1121        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1122        same bug. Such groups have to be quantified to be useful, or contained
1123        inside another quantified group. (If there's no repetition, the reference
1124        can never match.) The problem arises because, having left the group and
1125        moved on to the rest of the pattern, a later failure that backtracks into
1126        the group uses the captured value from the final iteration of the group
1127        rather than the correct earlier one. I have fixed this in PCRE by forcing
1128        any group that contains a reference to itself to be an atomic group; that
1129        is, there cannot be any backtracking into it once it has completed. This is
1130        similar to recursive and subroutine calls.
1131    
1132    
1133    Version 8.00 19-Oct-09
1134    ----------------------
1135    
1136    1.  The table for translating pcre_compile() error codes into POSIX error codes
1137        was out-of-date, and there was no check on the pcre_compile() error code
1138        being within the table. This could lead to an OK return being given in
1139        error.
1140    
1141    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1142        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1143        in a Windows environment.
1144    
1145    3.  The pcregrep --count option prints the count for each file even when it is
1146        zero, as does GNU grep. However, pcregrep was also printing all files when
1147        --files-with-matches was added. Now, when both options are given, it prints
1148        counts only for those files that have at least one match. (GNU grep just
1149        prints the file name in this circumstance, but including the count seems
1150        more useful - otherwise, why use --count?) Also ensured that the
1151        combination -clh just lists non-zero counts, with no names.
1152    
1153    4.  The long form of the pcregrep -F option was incorrectly implemented as
1154        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1155        but it seems right to fix it, and I didn't think it was worth preserving
1156        the old behaviour.
1157    
1158    5.  The command line items --regex=pattern and --regexp=pattern were not
1159        recognized by pcregrep, which required --regex pattern or --regexp pattern
1160        (with a space rather than an '='). The man page documented the '=' forms,
1161        which are compatible with GNU grep; these now work.
1162    
1163    6.  No libpcreposix.pc file was created for pkg-config; there was just
1164        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1165    
1166    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1167        when UCP support is not needed, by modifying the Python script that
1168        generates it from Unicode data files. This should not matter if the module
1169        is correctly used as a library, but I received one complaint about 50K of
1170        unwanted data. My guess is that the person linked everything into his
1171        program rather than using a library. Anyway, it does no harm.
1172    
1173    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1174        was a minimum greater than 1 for a wide character in a possessive
1175        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1176        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1177        character. Chaos in the form of incorrect output or a compiling loop could
1178        result.
1179    
1180    9.  The restrictions on what a pattern can contain when partial matching is
1181        requested for pcre_exec() have been removed. All patterns can now be
1182        partially matched by this function. In addition, if there are at least two
1183        slots in the offset vector, the offset of the earliest inspected character
1184        for the match and the offset of the end of the subject are set in them when
1185        PCRE_ERROR_PARTIAL is returned.
1186    
1187    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1188        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1189        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1190        and may be more useful for multi-segment matching.
1191    
1192    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1193        used to be given if ever the end of the subject was reached; now it is
1194        given only if matching could not proceed because another character was
1195        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1196        string "Z", which now yields "no match" instead of "partial match". In the
1197        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1198        final character ended with (*FAIL).
1199    
1200    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1201        if the pattern had a "must contain" character that was already found in the
1202        earlier partial match, unless partial matching was again requested. For
1203        example, with the pattern /dog.(body)?/, the "must contain" character is
1204        "g". If the first part-match was for the string "dog", restarting with
1205        "sbody" failed. This bug has been fixed.
1206    
1207    13. The string returned by pcre_dfa_exec() after a partial match has been
1208        changed so that it starts at the first inspected character rather than the
1209        first character of the match. This makes a difference only if the pattern
1210        starts with a lookbehind assertion or \b or \B (\K is not supported by
1211        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1212        matching functions compatible, and I think it's the right thing to do.
1213    
1214    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1215        so that the demonstration program is easily available in environments where
1216        PCRE has not been installed from source.
1217    
1218    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1219        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1220        library.
1221    
1222    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1223        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1224        is not the first non-POSIX option to be added. Clearly some people find
1225        these options useful.
1226    
1227    17. If a caller to the POSIX matching function regexec() passes a non-zero
1228        value for nmatch with a NULL value for pmatch, the value of
1229        nmatch is forced to zero.
1230    
1231    18. RunGrepTest did not have a test for the availability of the -u option of
1232        the diff command, as RunTest does. It now checks in the same way as
1233        RunTest, and also checks for the -b option.
1234    
1235    19. If an odd number of negated classes containing just a single character
1236        interposed, within parentheses, between a forward reference to a named
1237        subpattern and the definition of the subpattern, compilation crashed with
1238        an internal error, complaining that it could not find the referenced
1239        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1240        [The bug was that it was starting one character too far in when skipping
1241        over the character class, thus treating the ] as data rather than
1242        terminating the class. This meant it could skip too much.]
1243    
1244    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1245        /g option in pcretest when the pattern contains \K, which makes it possible
1246        to have an empty string match not at the start, even when the pattern is
1247        anchored. Updated pcretest and pcredemo to use this option.
1248    
1249    21. If the maximum number of capturing subpatterns in a recursion was greater
1250        than the maximum at the outer level, the higher number was returned, but
1251        with unset values at the outer level. The correct (outer level) value is
1252        now given.
1253    
1254    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1255        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1256        make it do so. The string so far is captured, making this feature
1257        compatible with Perl.
1258    
1259    23. The tests have been re-organized, adding tests 11 and 12, to make it
1260        possible to check the Perl 5.10 features against Perl 5.10.
1261    
1262    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1263        pattern matches a fixed length string. PCRE did not allow this; now it
1264        does. Neither allows recursion.
1265    
1266    25. I finally figured out how to implement a request to provide the minimum
1267        length of subject string that was needed in order to match a given pattern.
1268        (It was back references and recursion that I had previously got hung up
1269        on.) This code has now been added to pcre_study(); it finds a lower bound
1270        to the length of subject needed. It is not necessarily the greatest lower
1271        bound, but using it to avoid searching strings that are too short does give
1272        some useful speed-ups. The value is available to calling programs via
1273        pcre_fullinfo().
1274    
1275    26. While implementing 25, I discovered to my embarrassment that pcretest had
1276        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1277        study optimizations had never been tested with that matching function.
1278        Oops. What is worse, even when it was passed study data, there was a bug in
1279        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1280        were also very few tests of studied patterns with pcre_dfa_exec().
1281    
1282    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1283        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1284        on the other side of the coin, they are no longer allowed to have different
1285        names, because these cannot be distinguished in PCRE, and this has caused
1286        confusion. (This is a difference from Perl.)
1287    
1288    28. When duplicate subpattern names are present (necessarily with different
1289        numbers, as required by 27 above), and a test is made by name in a
1290        conditional pattern, either for a subpattern having been matched, or for
1291        recursion in such a pattern, all the associated numbered subpatterns are
1292        tested, and the overall condition is true if the condition is true for any
1293        one of them. This is the way Perl works, and is also more like the way
1294        testing by number works.
1295    
1296    
1297    Version 7.9 11-Apr-09
1298    ---------------------
1299    
1300    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1301        (pcretest), all targets were linked against these libraries. This included
1302        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1303        libraries. This caused unwanted dependencies to be created. This problem
1304        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1305        pcretest is linked with readline.
1306    
1307    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1308        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1309        moved outside it again, because FALSE and TRUE are already defined in AIX,
1310        but BOOL is not.
1311    
1312    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1313        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1314    
1315    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1316        hyphens) following file names and line numbers when outputting matching
1317        lines. This is not true; no spaces are inserted. I have also clarified the
1318        wording for the --colour (or --color) option.
1319    
1320    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1321        was not coloured; this is different to GNU grep, so I have changed it to be
1322        the same.
1323    
1324    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1325        each matching line was coloured. Now it goes on to look for further matches
1326        of any of the test patterns, which is the same behaviour as GNU grep.
1327    
1328    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1329        doesn't make sense to accept an empty string match in pcregrep, so I have
1330        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1331        seems to be how GNU grep behaves.
1332    
1333    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1334        start or after a newline", because the conditional assertion was not being
1335        correctly handled. The rule now is that both the assertion and what follows
1336        in the first alternative must satisfy the test.
1337    
1338    9.  If auto-callout was enabled in a pattern with a conditional group whose
1339        condition was an assertion, PCRE could crash during matching, both with
1340        pcre_exec() and pcre_dfa_exec().
1341    
1342    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1343        used for matching.
1344    
1345    11. Unicode property support in character classes was not working for
1346        characters (bytes) greater than 127 when not in UTF-8 mode.
1347    
1348    12. Added the -M command line option to pcretest.
1349    
1350    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1351    
1352    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1353    
1354    16. Added comments and documentation about mis-use of no_arg in the C++
1355        wrapper.
1356    
1357    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1358        from Martin Jerabek that uses macro names for all relevant character and
1359        string constants.
1360    
1361    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1362        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1363        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1364        these, but not everybody uses configure.
1365    
1366    19. A conditional group that had only one branch was not being correctly
1367        recognized as an item that could match an empty string. This meant that an
1368        enclosing group might also not be so recognized, causing infinite looping
1369        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1370        with the subject "ab", where knowledge that the repeated group can match
1371        nothing is needed in order to break the loop.
1372    
1373    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1374        exec(), but without supplying a callout function, matching went wrong.
1375    
1376    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1377        leak if the size of the offset vector was greater than 30. When the vector
1378        is smaller, the saved offsets during recursion go onto a local stack
1379        vector, but for larger vectors malloc() is used. It was failing to free
1380        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1381        error, in fact).
1382    
1383    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1384        heapframe that is used only when UTF-8 support is enabled. This caused no
1385        problem, but was untidy.
1386    
1387    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1388        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1389        included within another project.
1390    
1391    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1392        slightly modified by me:
1393    
1394          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1395              not building pcregrep.
1396    
1397          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1398              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1399    
1400    25. Forward references, both numeric and by name, in patterns that made use of
1401        duplicate group numbers, could behave incorrectly or give incorrect errors,
1402        because when scanning forward to find the reference group, PCRE was not
1403        taking into account the duplicate group numbers. A pattern such as
1404        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1405    
1406    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1407        the feature of a custom pointer more persuasive (as requested by a user).
1408    
1409    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1410        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1411        pre-defined.
1412    
1413    28. Added support for (*UTF8) at the start of a pattern.
1414    
1415    29. Arrange for flags added by the "release type" setting in CMake to be shown
1416        in the configuration summary.
1417    
1418    
1419    Version 7.8 05-Sep-08
1420  ---------------------  ---------------------
1421    
1422  1.  Replaced UCP searching code with optimized version as implemented for Ad  1.  Replaced UCP searching code with optimized version as implemented for Ad
1423      Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-      Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1424      stage table and inline lookup instead of a function, giving speed ups of 2      stage table and inline lookup instead of a function, giving speed ups of 2
1425      to 5 times on some simple patterns that I tested. Permission was given to      to 5 times on some simple patterns that I tested. Permission was given to
1426      distribute the MultiStage2.py script that generates the tables (it's not in      distribute the MultiStage2.py script that generates the tables (it's not in
1427      the tarball, but is in the Subversion repository).      the tarball, but is in the Subversion repository).
1428    
1429    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1430        scripts.
1431    
1432    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1433        a group with a zero qualifier. The result of the study could be incorrect,
1434        or the function might crash, depending on the pattern.
1435    
1436    4.  Caseless matching was not working for non-ASCII characters in back
1437        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1438        It now works when Unicode Property Support is available.
1439    
1440    5.  In pcretest, an escape such as \x{de} in the data was always generating
1441        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1442        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1443        truncation.
1444    
1445    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1446    
1447    7.  Added two (int) casts to pcregrep when printing the difference of two
1448        pointers, in case they are 64-bit values.
1449    
1450    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1451        test 2 if it fails.
1452    
1453    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1454        and a #define of that name to empty if it is not externally set. This is to
1455        allow users of MSVC to set it if necessary.
1456    
1457    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1458        the convenience functions in the pcre_get.c source file.
1459    
1460    11. An option change at the start of a pattern that had top-level alternatives
1461        could cause overwriting and/or a crash. This command provoked a crash in
1462        some environments:
1463    
1464          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1465    
1466        This potential security problem was recorded as CVE-2008-2371.
1467    
1468    12. For a pattern where the match had to start at the beginning or immediately
1469        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1470        pcre_dfa_exec() could read past the end of the passed subject if there was
1471        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1472        pcretest so that it places the subject at the end of its malloc-ed buffer.
1473    
1474    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1475        exec() might read past the end of the data buffer in UTF-8 mode.
1476    
1477    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1478        the data contained the byte 0x85 as part of a UTF-8 character within its
1479        first line. This applied both to normal and DFA matching.
1480    
1481    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1482        /^[^d]*?$/8 failed to match "abc".
1483    
1484    16. Added a missing copyright notice to pcrecpp_internal.h.
1485    
1486    17. Make it more clear in the documentation that values returned from
1487        pcre_exec() in ovector are byte offsets, not character counts.
1488    
1489    18. Tidied a few places to stop certain compilers from issuing warnings.
1490    
1491    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1492        supplied by Stefan Weber. I made a further small update for 7.8 because
1493        there is a change of source arrangements: the pcre_searchfuncs.c module is
1494        replaced by pcre_ucd.c.
1495    
1496    
1497  Version 7.7 07-May-08  Version 7.7 07-May-08

Legend:
Removed from v.349  
changed lines
  Added in v.978

  ViewVC Help
Powered by ViewVC 1.1.5