ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 205 by ph10, Fri Aug 3 13:18:33 2007 UTC revision 977 by zherczeg, Sun Jun 17 06:20:52 2012 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4  Version 7.3 05-Jul-07  Version 8.31 02-June-2012
5    -------------------------
7    1.  Fixing a wrong JIT test case and some compiler warnings.
9    2.  Removed a bashism from the RunTest script.
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)".
15    4.  Partial matching support is added to the JIT compiler.
17    5.  Fixed several bugs concerned with partial matching of items that consist
18        of more than one character:
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.
24        (b) \R did not give a hard partial match if \r was found at the end of the
25            subject.
27        (c) \X did not give a hard partial match after matching one or more
28            characters at the end of the subject.
30        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
31            a partial match for the string "\r".
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.
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.
40    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
41        or /S+[+] with a digit between 1 and 7.
43    8.  OP_NOT now supports any UTF character not just single-byte ones.
45    9.  (*MARK) control verb is now supported by the JIT compiler.
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.)
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.
58    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
59        call.
61    14. As documented, (*COMMIT) is now confined to within a positive assertion.
63    15. It is now possible to link pcretest with libedit as an alternative to
64        libreadline.
66    16. (*COMMIT) control verb is now supported by the JIT compiler.
68    17. The Unicode data tables have been updated to Unicode 6.1.0.
70    18. Added --file-list option to pcregrep.
72    19. Added binary file support to pcregrep, including the -a, --binary-files,
73        -I, and --text options.
75    20. The madvise function is renamed for posix_madvise for QNX compatibility
76        reasons. Fixed by Giuseppe D'Angelo.
78    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
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.
83    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
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.
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).
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.
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.)
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.
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.
114    30. JIT compiler update.
116    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
117        Rill for the patch.
119    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
121    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
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.
127    35. Improve JIT code generation for greedy plus quantifier.
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.
133    37. Optimizing single character iterators in JIT.
136    Version 8.30 04-February-2012
137    -----------------------------
139    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
140        name is defined in ctype.h.
142    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
143        only in quite long subpatterns.
145    3.  Removed the function pcre_info(), which has been obsolete and deprecated
146        since it was replaced by pcre_fullinfo() in February 2000.
148    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
149        match a (*MARK), and the match failed at the start of the subject, a
150        reference to memory before the start of the subject could occur. This bug
151        was introduced by fix 17 of release 8.21.
153    5.  A reference to an unset group with zero minimum repetition was giving
154        totally wrong answers (in non-JavaScript-compatibility mode). For example,
155        /(another)?(\1?)test/ matched against "hello world test". This bug was
156        introduced in release 8.13.
158    6.  Add support for 16-bit character strings (a large amount of work involving
159        many changes and refactorings).
161    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
162        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
163        from a file.
165    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
166        rounding is not applied in this particular case).
168    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
169        if they appear, or are escaped, in patterns.
171    10. Get rid of a number of -Wunused-but-set-variable warnings.
173    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
174        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
175        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
176        also returns the mark "x". This bug applied to capturing parentheses,
177        non-capturing parentheses, and atomic parentheses. It also applied to some
178        assertions.
180    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
181        information out of configure.ac instead of relying on pcre.h.generic, which
182        is not stored in the repository.
184    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
185        -lreadline.
187    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
189    15. Applied Graycode's patch to put the top-level frame on the stack rather
190        than the heap when not using the stack for recursion. This gives a
191        performance improvement in many cases when recursion is not deep.
193    16. Experimental code added to "pcretest -C" to output the stack frame size.
196    Version 8.21 12-Dec-2011
197    ------------------------
199    1.  Updating the JIT compiler.
201    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
202        are added as well.
204    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
205        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
206        calling _pcre_jit_exec. Some extra comments are added.
208    4.  (*MARK) settings inside atomic groups that do not contain any capturing
209        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
210        was introduced by change 18 for 8.20.
212    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
213        ECMA-262 standard.
215    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
216        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
217        This bug was probably introduced by change 9 of 8.13.
219    7.  While fixing 6 above, I noticed that a number of other items were being
220        incorrectly rejected as "not fixed length". This arose partly because newer
221        opcodes had not been added to the fixed-length checking code. I have (a)
222        corrected the bug and added tests for these items, and (b) arranged for an
223        error to occur if an unknown opcode is encountered while checking for fixed
224        length instead of just assuming "not fixed length". The items that were
225        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
226        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
227        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
229    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
230        being incorrectly compiled and would have given unpredicatble results.
232    9.  A possessively repeated subpattern with minimum repeat count greater than
233        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
234        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
235        the first (A) could occur when it should not.
237    10. Add a cast and remove a redundant test from the code.
239    11. JIT should use pcre_malloc/pcre_free for allocation.
241    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
242        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
243        is anything other than /usr, -L is still shown).
245    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
247    14. Perl does not support \N without a following name in a [] class; PCRE now
248        also gives an error.
250    15. If a forward reference was repeated with an upper limit of around 2000,
251        it caused the error "internal error: overran compiling workspace". The
252        maximum number of forward references (including repeats) was limited by the
253        internal workspace, and dependent on the LINK_SIZE. The code has been
254        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
255        the default depends on LINK_SIZE. There is a new upper limit (for safety)
256        of around 200,000 forward references. While doing this, I also speeded up
257        the filling in of repeated forward references.
259    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
260        incorrectly expecting the subject to contain another "a" after the start.
262    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
263        in the match, the SKIP should be ignored. This was not happening; instead
264        the SKIP was being treated as NOMATCH. For patterns such as
265        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
266        tested.
268    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
269        now much more compatible with Perl, in particular in cases where the result
270        is a non-match for a non-anchored pattern. For example, if
271        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
272        "m", where previously it did not return a name. A side effect of this
273        change is that for partial matches, the last encountered mark name is
274        returned, as for non matches. A number of tests that were previously not
275        Perl-compatible have been moved into the Perl-compatible test files. The
276        refactoring has had the pleasing side effect of removing one argument from
277        the match() function, thus reducing its stack requirements.
279    19. If the /S+ option was used in pcretest to study a pattern using JIT,
280        subsequent uses of /S (without +) incorrectly behaved like /S+.
282    21. Retrieve executable code size support for the JIT compiler and fixing
283        some warnings.
285    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
286        not work when the shorter character appeared right at the end of the
287        subject string.
289    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
290        systems.
292    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
293        output it when the /M option is used in pcretest.
295    25. The CheckMan script was not being included in the distribution. Also, added
296        an explicit "perl" to run Perl scripts from the PrepareRelease script
297        because this is reportedly needed in Windows.
299    26. If study data was being save in a file and studying had not found a set of
300        "starts with" bytes for the pattern, the data written to the file (though
301        never used) was taken from uninitialized memory and so caused valgrind to
302        complain.
304    27. Updated RunTest.bat as provided by Sheri Pierce.
306    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
308    29. Computation of memory usage for the table of capturing group names was
309        giving an unnecessarily large value.
312    Version 8.20 21-Oct-2011
313    ------------------------
315    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
316        a POSIX class. After further experiments with Perl, which convinced me that
317        Perl has bugs and confusions, a closing square bracket is no longer allowed
318        in a POSIX name. This bug also affected patterns with classes that started
319        with full stops.
321    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
322        captured substring, but while checking the failing first alternative,
323        substring 1 is temporarily captured. If the output vector supplied to
324        pcre_exec() was not big enough for this capture, the yield of the function
325        was still zero ("insufficient space for captured substrings"). This cannot
326        be totally fixed without adding another stack variable, which seems a lot
327        of expense for a edge case. However, I have improved the situation in cases
328        such as /(a)(b)x|abc/ matched against "abc", where the return code
329        indicates that fewer than the maximum number of slots in the ovector have
330        been set.
332    3.  Related to (2) above: when there are more back references in a pattern than
333        slots in the output vector, pcre_exec() uses temporary memory during
334        matching, and copies in the captures as far as possible afterwards. It was
335        using the entire output vector, but this conflicts with the specification
336        that only 2/3 is used for passing back captured substrings. Now it uses
337        only the first 2/3, for compatibility. This is, of course, another edge
338        case.
340    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
341        main code base, and can be used by building with --enable-jit. When this is
342        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
343        runtime --no-jit option is given.
345    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
346        ovector, the return from the function was zero, implying that there were
347        other matches that did not fit. The correct "exactly full" value is now
348        returned.
350    6.  If a subpattern that was called recursively or as a subroutine contained
351        (*PRUNE) or any other control that caused it to give a non-standard return,
352        invalid errors such as "Error -26 (nested recursion at the same subject
353        position)" or even infinite loops could occur.
355    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
356        computing the minimum length on reaching *ACCEPT, and so ended up with the
357        wrong value of 1 rather than 0. Further investigation indicates that
358        computing a minimum subject length in the presence of *ACCEPT is difficult
359        (think back references, subroutine calls), and so I have changed the code
360        so that no minimum is registered for a pattern that contains *ACCEPT.
362    8.  If (*THEN) was present in the first (true) branch of a conditional group,
363        it was not handled as intended. [But see 16 below.]
365    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
366        Sheri Pierce.
368    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
369        the first byte in a match must be "a".
371    11. Change 17 for 8.13 increased the recursion depth for patterns like
372        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
373        pattern contains any instances of (*THEN). If it does not, the old
374        optimizations are restored. It would be nice to do this on a per-group
375        basis, but at the moment that is not feasible.
377    12. In some environments, the output of pcretest -C is CRLF terminated. This
378        broke RunTest's code that checks for the link size. A single white space
379        character after the value is now allowed for.
381    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
382        For "fr", it uses the Windows-specific input and output files.
384    14. If (*THEN) appeared in a group that was called recursively or as a
385        subroutine, it did not work as intended. [But see next item.]
387    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
388        pattern fragments (but not containing any | characters). If A and B are
389        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
390        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
391        D. In other words, Perl considers parentheses that do not contain any |
392        characters to be part of a surrounding alternative, whereas PCRE was
393        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
394        differently. PCRE now behaves in the same way as Perl, except in the case
395        of subroutine/recursion calls such as (?1) which have in any case always
396        been different (but PCRE had them first :-).
398    16. Related to 15 above: Perl does not treat the | in a conditional group as
399        creating alternatives. Such a group is treated in the same way as an
400        ordinary group without any | characters when processing (*THEN). PCRE has
401        been changed to match Perl's behaviour.
403    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
404        RunGrepTest script failed.
406    18. Change 22 for version 13 caused atomic groups to use more stack. This is
407        inevitable for groups that contain captures, but it can lead to a lot of
408        stack use in large patterns. The old behaviour has been restored for atomic
409        groups that do not contain any capturing parentheses.
411    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
412        suppress the check for a minimum subject length at run time. (If it was
413        given to pcre_exec() or pcre_dfa_exec() it did work.)
415    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
416        fail to work when decoding hex characters in data strings in EBCDIC
417        environments.
419    21. It appears that in at least one Mac OS environment, the isxdigit() function
420        is implemented as a macro that evaluates to its argument more than once,
421        contravening the C 90 Standard (I haven't checked a later standard). There
422        was an instance in pcretest which caused it to go wrong when processing
423        \x{...} escapes in subject strings. The has been rewritten to avoid using
424        things like p++ in the argument of isxdigit().
427    Version 8.13 16-Aug-2011
428    ------------------------
430    1.  The Unicode data tables have been updated to Unicode 6.0.0.
432    2.  Two minor typos in pcre_internal.h have been fixed.
434    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
435        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
436        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
438    4.  There were a number of related bugs in the code for matching backrefences
439        caselessly in UTF-8 mode when codes for the characters concerned were
440        different numbers of bytes. For example, U+023A and U+2C65 are an upper
441        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
442        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
443        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
444        2-byte code at the end of the subject (it thought there wasn't enough data
445        left).
447    5.  Comprehensive information about what went wrong is now returned by
448        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
449        as the output vector has at least 2 elements. The offset of the start of
450        the failing character and a reason code are placed in the vector.
452    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
453        now returned is for the first byte of the failing character, instead of the
454        last byte inspected. This is an incompatible change, but I hope it is small
455        enough not to be a problem. It makes the returned offset consistent with
456        pcre_exec() and pcre_dfa_exec().
458    7.  pcretest now gives a text phrase as well as the error number when
459        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
460        failure, the offset and reason code are output.
462    8.  When \R was used with a maximizing quantifier it failed to skip backwards
463        over a \r\n pair if the subsequent match failed. Instead, it just skipped
464        back over a single character (\n). This seems wrong (because it treated the
465        two characters as a single entity when going forwards), conflicts with the
466        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
467        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
468        has been changed.
470    9.  Some internal refactoring has changed the processing so that the handling
471        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
472        time (the PCRE_DOTALL option was changed this way some time ago: version
473        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
474        which was always a bit of a fudge. It also means that there is one less
475        argument for the match() function, which reduces its stack requirements
476        slightly. This change also fixes an incompatibility with Perl: the pattern
477        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
479    10. More internal refactoring has drastically reduced the number of recursive
480        calls to match() for possessively repeated groups such as (abc)++ when
481        using pcre_exec().
483    11. While implementing 10, a number of bugs in the handling of groups were
484        discovered and fixed:
486        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
487        (a|)*(?1) gave a compile-time internal error.
488        ((a|)+)+  did not notice that the outer group could match an empty string.
489        (^a|^)+   was not marked as anchored.
490        (.*a|.*)+ was not marked as matching at start or after a newline.
492    12. Yet more internal refactoring has removed another argument from the match()
493        function. Special calls to this function are now indicated by setting a
494        value in a variable in the "match data" data block.
496    13. Be more explicit in pcre_study() instead of relying on "default" for
497        opcodes that mean there is no starting character; this means that when new
498        ones are added and accidentally left out of pcre_study(), testing should
499        pick them up.
501    14. The -s option of pcretest has been documented for ages as being an old
502        synonym of -m (show memory usage). I have changed it to mean "force study
503        for every regex", that is, assume /S for every regex. This is similar to -i
504        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
505        using it. It makes it easier to run collections of tests with and without
506        study enabled, and thereby test pcre_study() more easily. All the standard
507        tests are now run with and without -s (but some patterns can be marked as
508        "never study" - see 20 below).
510    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
511        restoration of the capturing data to the outer values was not happening
512        correctly.
514    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
515        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
516        pattern had matched an empty string, and so incorrectly returned a no
517        match.
519    17. There was optimizing code for the last branch of non-capturing parentheses,
520        and also for the obeyed branch of a conditional subexpression, which used
521        tail recursion to cut down on stack usage. Unfortunately, now that there is
522        the possibility of (*THEN) occurring in these branches, tail recursion is
523        no longer possible because the return has to be checked for (*THEN). These
524        two optimizations have therefore been removed. [But see 8.20/11 above.]
526    18. If a pattern containing \R was studied, it was assumed that \R always
527        matched two bytes, thus causing the minimum subject length to be
528        incorrectly computed because \R can also match just one byte.
530    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
531        was incorrectly computed.
533    20. If /S is present twice on a test pattern in pcretest input, it now
534        *disables* studying, thereby overriding the use of -s on the command line
535        (see 14 above). This is necessary for one or two tests to keep the output
536        identical in both cases.
538    21. When (*ACCEPT) was used in an assertion that matched an empty string and
539        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
541    22. When an atomic group that contained a capturing parenthesis was
542        successfully matched, but the branch in which it appeared failed, the
543        capturing was not being forgotten if a higher numbered group was later
544        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
545        group 1 to "a", when in fact it should be unset. This applied to multi-
546        branched capturing and non-capturing groups, repeated or not, and also to
547        positive assertions (capturing in negative assertions does not happen
548        in PCRE) and also to nested atomic groups.
550    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
551        subject after a captured substring, to make it easier to tell which of a
552        number of identical substrings has been captured.
554    24. The way atomic groups are processed by pcre_exec() has been changed so that
555        if they are repeated, backtracking one repetition now resets captured
556        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
557        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
558        "aaa". Previously, it would have been "a". As part of this code
559        refactoring, the way recursive calls are handled has also been changed.
561    25. If an assertion condition captured any substrings, they were not passed
562        back unless some other capturing happened later. For example, if
563        (?(?=(a))a) was matched against "a", no capturing was returned.
565    26. When studying a pattern that contained subroutine calls or assertions,
566        the code for finding the minimum length of a possible match was handling
567        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
568        group 1 called group 2 while simultaneously a separate group 2 called group
569        1). A stack overflow occurred in this case. I have fixed this by limiting
570        the recursion depth to 10.
572    27. Updated RunTest.bat in the distribution to the version supplied by Tom
573        Fortmann. This supports explicit test numbers on the command line, and has
574        argument validation and error reporting.
576    28. An instance of \X with an unlimited repeat could fail if at any point the
577        first character it looked at was a mark character.
579    29. Some minor code refactoring concerning Unicode properties and scripts
580        should reduce the stack requirement of match() slightly.
582    30. Added the '=' option to pcretest to check the setting of unused capturing
583        slots at the end of the pattern, which are documented as being -1, but are
584        not included in the return count.
586    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
587        compiled something random. Now it gives a compile-time error (as does
588        Perl).
590    32. A *MARK encountered during the processing of a positive assertion is now
591        recorded and passed back (compatible with Perl).
593    33. If --only-matching or --colour was set on a pcregrep call whose pattern
594        had alternative anchored branches, the search for a second match in a line
595        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
596        matched the line "0102" twice. The same bug affected patterns that started
597        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
598        twice.
600    34. Previously, PCRE did not allow quantification of assertions. However, Perl
601        does, and because of capturing effects, quantifying parenthesized
602        assertions may at times be useful. Quantifiers are now allowed for
603        parenthesized assertions.
605    35. A minor code tidy in pcre_compile() when checking options for \R usage.
607    36. \g was being checked for fancy things in a character class, when it should
608        just be a literal "g".
610    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
611        appearance of a nested POSIX class supersedes an apparent external class.
612        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
613        unescaped square brackets may also appear as part of class names. For
614        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
615        more like Perl. (But see 8.20/1 above.)
617    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
618        was because it thought it was \N{name}, which is not supported).
620    39. Add minix to OS list not supporting the -S option in pcretest.
622    40. PCRE tries to detect cases of infinite recursion at compile time, but it
623        cannot analyze patterns in sufficient detail to catch mutual recursions
624        such as ((?1))((?2)). There is now a runtime test that gives an error if a
625        subgroup is called recursively as a subpattern for a second time at the
626        same position in the subject string. In previous releases this might have
627        been caught by the recursion limit, or it might have run out of stack.
629    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
630        happen only once. PCRE was, however incorrectly giving a compile time error
631        "recursive call could loop indefinitely" because it cannot analyze the
632        pattern in sufficient detail. The compile time test no longer happens when
633        PCRE is compiling a conditional subpattern, but actual runaway loops are
634        now caught at runtime (see 40 above).
636    42. It seems that Perl allows any characters other than a closing parenthesis
637        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
638        has been changed to be the same.
640    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
641        as not to get warnings when autogen.sh is called. Also changed
642        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
644    44. To help people who use pcregrep to scan files containing exceedingly long
645        lines, the following changes have been made:
647        (a) The default value of the buffer size parameter has been increased from
648            8K to 20K. (The actual buffer used is three times this size.)
650        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
651            PCRE is built.
653        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
654            to be set at run time.
656        (d) Numerical values in pcregrep options can be followed by K or M, for
657            example --buffer-size=50K.
659        (e) If a line being scanned overflows pcregrep's buffer, an error is now
660            given and the return code is set to 2.
662    45. Add a pointer to the latest mark to the callout data block.
664    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
665        partial match of an empty string instead of no match. This was specific to
666        the use of ".".
668    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
669        complete match instead of a partial match. This bug was dependent on both
670        the PCRE_UTF8 and PCRE_DOTALL options being set.
672    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
673        starting byte set, because \b was not being ignored.
676    Version 8.12 15-Jan-2011
677    ------------------------
679    1.  Fixed some typos in the markup of the man pages, and wrote a script that
680        checks for such things as part of the documentation building process.
682    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
683        --match-limit and --recursion-limit options (added for 8.11). In
684        particular, this made one of the standard tests fail. (The integer value
685        went into the wrong half of a long int.)
687    3.  If the --colour option was given to pcregrep with -v (invert match), it
688        did strange things, either producing crazy output, or crashing. It should,
689        of course, ignore a request for colour when reporting lines that do not
690        match.
692    4.  Another pcregrep bug caused similar problems if --colour was specified with
693        -M (multiline) and the pattern match finished with a line ending.
695    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
696        matched in multiline mode, the following line was shown as part of the
697        match. This seems wrong, so I have changed it.
699    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
700        the check for further matches in the same line (so they could be coloured)
701        to overrun the end of the current line. If another match was found, it was
702        incorrectly shown (and then shown again when found in the next line).
704    7.  If pcregrep was compiled under Windows, there was a reference to the
705        function pcregrep_exit() before it was defined. I am assuming this was
706        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
707        reported by a user. I've moved the definition above the reference.
710    Version 8.11 10-Dec-2010
711    ------------------------
713    1.  (*THEN) was not working properly if there were untried alternatives prior
714        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
715        backtracked to try for "b" instead of moving to the next alternative branch
716        at the same level (in this case, to look for "c"). The Perl documentation
717        is clear that when (*THEN) is backtracked onto, it goes to the "next
718        alternative in the innermost enclosing group".
720    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
721        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
722        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
723        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
724        (*THEN).
726    3.  If \s appeared in a character class, it removed the VT character from
727        the class, even if it had been included by some previous item, for example
728        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
729        of \s, but is part of the POSIX "space" class.)
731    4.  A partial match never returns an empty string (because you can always
732        match an empty string at the end of the subject); however the checking for
733        an empty string was starting at the "start of match" point. This has been
734        changed to the "earliest inspected character" point, because the returned
735        data for a partial match starts at this character. This means that, for
736        example, /(?<=abc)def/ gives a partial match for the subject "abc"
737        (previously it gave "no match").
739    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
740        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
741        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
742        has an implication that the given string is incomplete (because a partial
743        match is preferred over a full match). For this reason, these items now
744        give a partial match in this situation. [Aside: previously, the one case
745        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
746        match rather than a full match, which was wrong by the old rules, but is
747        now correct.]
749    6.  There was a bug in the handling of #-introduced comments, recognized when
750        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
751        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
752        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
753        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
754        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
755        places in pcre_compile().
757    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
758        comments when looking ahead for named forward references to subpatterns,
759        the only newline sequence it recognized was NL. It now handles newlines
760        according to the set newline convention.
762    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
763        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
764        cater for a lack of strerror(). These oversights have been fixed.
766    9.  Added --match-limit and --recursion-limit to pcregrep.
768    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
770    11. When the -o option was used, pcregrep was setting a return code of 1, even
771        when matches were found, and --line-buffered was not being honoured.
773    12. Added an optional parentheses number to the -o and --only-matching options
774        of pcregrep.
776    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
777        can match an empty string. The code to do it in pcretest and pcredemo
778        needed fixing:
780        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
781            only one byte after an empty string match just before CRLF (this case
782            just got forgotten; "any" and "anycrlf" were OK).
784        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
785            mode when an empty string match preceded an ASCII character followed by
786            a non-ASCII character. (The code for advancing by one character rather
787            than one byte was nonsense.)
789        (c) The pcredemo.c sample program did not have any code at all to handle
790            the cases when CRLF is a valid newline sequence.
792    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
793        as a starting offset was within the subject string. There is now a new
794        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
795        negative or greater than the length of the string. In order to test this,
796        pcretest is extended to allow the setting of negative starting offsets.
798    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
799        starting offset points to the beginning of a UTF-8 character was
800        unnecessarily clumsy. I tidied it up.
802    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
803        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
805    17. Nobody had reported that the --include_dir option, which was added in
806        release 7.7 should have been called --include-dir (hyphen, not underscore)
807        for compatibility with GNU grep. I have changed it to --include-dir, but
808        left --include_dir as an undocumented synonym, and the same for
809        --exclude-dir, though that is not available in GNU grep, at least as of
810        release 2.5.4.
812    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
813        characters from a string of bytes) have been redefined so as not to use
814        loops, in order to improve performance in some environments. At the same
815        time, I abstracted some of the common code into auxiliary macros to save
816        repetition (this should not affect the compiled code).
818    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
819        compile-time error is now given if \c is not followed by an ASCII
820        character, that is, a byte less than 128. (In EBCDIC mode, the code is
821        different, and any byte value is allowed.)
823    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
824        START_OPTIMIZE option, which is now allowed at compile time - but just
825        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
826        to pcregrep and other applications that have no direct access to PCRE
827        options. The new /Y option in pcretest sets this option when calling
828        pcre_compile().
830    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
831        back references. Groups containing recursive back references were forced to
832        be atomic by that change, but in the case of named groups, the amount of
833        memory required was incorrectly computed, leading to "Failed: internal
834        error: code overflow". This has been fixed.
836    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
837        pcretest.c, to avoid build problems in some Borland environments.
840    Version 8.10 25-Jun-2010
841    ------------------------
843    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
844        THEN.
846    2.  (*ACCEPT) was not working when inside an atomic group.
848    3.  Inside a character class, \B is treated as a literal by default, but
849        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
850        causes the error). The code is unchanged, but I tidied the documentation.
852    4.  Inside a character class, PCRE always treated \R and \X as literals,
853        whereas Perl faults them if its -w option is set. I have changed PCRE so
854        that it faults them when PCRE_EXTRA is set.
856    5.  Added support for \N, which always matches any character other than
857        newline. (It is the same as "." when PCRE_DOTALL is not set.)
859    6.  When compiling pcregrep with newer versions of gcc which may have
860        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
861        declared with attribute warn_unused_result" were given. Just casting the
862        result to (void) does not stop the warnings; a more elaborate fudge is
863        needed. I've used a macro to implement this.
865    7.  Minor change to pcretest.c to avoid a compiler warning.
867    8.  Added four artifical Unicode properties to help with an option to make
868        \s etc use properties (see next item). The new properties are: Xan
869        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
871    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
872        use Unicode properties. (*UCP) at the start of a pattern can be used to set
873        this option. Modified pcretest to add /W to test this facility. Added
874        REG_UCP to make it available via the POSIX interface.
876    10. Added --line-buffered to pcregrep.
878    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
879        studied, and the match started with a letter with a code point greater than
880        127 whose first byte was different to the first byte of the other case of
881        the letter, the other case of this starting letter was not recognized
882        (#976).
884    12. If a pattern that was studied started with a repeated Unicode property
885        test, for example, \p{Nd}+, there was the theoretical possibility of
886        setting up an incorrect bitmap of starting bytes, but fortunately it could
887        not have actually happened in practice until change 8 above was made (it
888        added property types that matched character-matching opcodes).
890    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
891        possible starting bytes for non-anchored patterns.
893    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
894        \R, and also a number of cases that involve Unicode properties, both
895        explicit and implicit when PCRE_UCP is set.
897    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
898        input, it could crash or give wrong results if characters with values
899        greater than 0xc0 were present in the subject string. (Detail: it assumed
900        UTF-8 input when processing these items.)
902    16. Added a lot of (int) casts to avoid compiler warnings in systems where
903        size_t is 64-bit (#991).
905    17. Added a check for running out of memory when PCRE is compiled with
906        --disable-stack-for-recursion (#990).
908    18. If the last data line in a file for pcretest does not have a newline on
909        the end, a newline was missing in the output.
911    19. The default pcre_chartables.c file recognizes only ASCII characters (values
912        less than 128) in its various bitmaps. However, there is a facility for
913        generating tables according to the current locale when PCRE is compiled. It
914        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
915        characters, are recognized by isspace() and therefore were getting set in
916        these tables, and indeed these tables seem to approximate to ISO 8859. This
917        caused a problem in UTF-8 mode when pcre_study() was used to create a list
918        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
919        which of course cannot start UTF-8 characters. I have changed the code so
920        that only real ASCII characters (less than 128) and the correct starting
921        bytes for UTF-8 encodings are set for characters greater than 127 when in
922        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
923        altogether.)
925    20. Added the /T option to pcretest so as to be able to run tests with non-
926        standard character tables, thus making it possible to include the tests
927        used for 19 above in the standard set of tests.
929    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
930        reference to a subpattern the other side of a comment that contains an
931        opening parenthesis caused either an internal compiling error, or a
932        reference to the wrong subpattern.
935    Version 8.02 19-Mar-2010
936    ------------------------
938    1.  The Unicode data tables have been updated to Unicode 5.2.0.
940    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
941        configured.
943    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
944        original author of that file, following a query about its status.
946    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
947        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
949    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
950        quantifier applied to a forward-referencing subroutine call, could compile
951        incorrect code or give the error "internal error: previously-checked
952        referenced subpattern not found".
954    6.  Both MS Visual Studio and Symbian OS have problems with initializing
955        variables to point to external functions. For these systems, therefore,
956        pcre_malloc etc. are now initialized to local functions that call the
957        relevant global functions.
959    7.  There were two entries missing in the vectors called coptable and poptable
960        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
961        I've fixed the data, and added a kludgy way of testing at compile time that
962        the lengths are correct (equal to the number of opcodes).
964    8.  Following on from 7, I added a similar kludge to check the length of the
965        eint vector in pcreposix.c.
967    9.  Error texts for pcre_compile() are held as one long string to avoid too
968        much relocation at load time. To find a text, the string is searched,
969        counting zeros. There was no check for running off the end of the string,
970        which could happen if a new error number was added without updating the
971        string.
973    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
975    11. \K was not working if it appeared in an atomic group or in a group that
976        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
977        \K is "not well defined" if used in an assertion. PCRE now accepts it if
978        the assertion is positive, but not if it is negative.
980    12. Change 11 fortuitously reduced the size of the stack frame used in the
981        "match()" function of pcre_exec.c by one pointer. Forthcoming
982        implementation of support for (*MARK) will need an extra pointer on the
983        stack; I have reserved it now, so that the stack frame size does not
984        decrease.
986    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
987        item in branch that calls a recursion is a subroutine call - as in the
988        second branch in the above example - was incorrectly given the compile-
989        time error "recursive call could loop indefinitely" because pcre_compile()
990        was not correctly checking the subroutine for matching a non-empty string.
992    14. The checks for overrunning compiling workspace could trigger after an
993        overrun had occurred. This is a "should never occur" error, but it can be
994        triggered by pathological patterns such as hundreds of nested parentheses.
995        The checks now trigger 100 bytes before the end of the workspace.
997    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1000    Version 8.01 19-Jan-2010
1001    ------------------------
1003    1.  If a pattern contained a conditional subpattern with only one branch (in
1004        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1005        computed the wrong minimum data length (which is of course zero for such
1006        subpatterns). This could cause incorrect "no match" results.
1008    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1009        the pattern is reset in the first branch, pcre_compile() failed with
1010        "internal error: code overflow at offset...". This happened only when
1011        the reset was to the original external option setting. (An optimization
1012        abstracts leading options settings into an external setting, which was the
1013        cause of this.)
1015    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1016        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1017        assertion pattern did not match (meaning that the assertion was true), it
1018        was incorrectly treated as false if the SKIP had been reached during the
1019        matching. This also applied to assertions used as conditions.
1021    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1022        assertion subpattern, including such a pattern used as a condition,
1023        unpredictable results occurred, instead of the error return
1026    5.  The C++ GlobalReplace function was not working like Perl for the special
1027        situation when an empty string is matched. It now does the fancy magic
1028        stuff that is necessary.
1030    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1031        removed. (These were left over from very, very early versions of PCRE.)
1033    7.  Some cosmetic changes to the code to make life easier when compiling it
1034        as part of something else:
1036        (a) Change DEBUG to PCRE_DEBUG.
1038        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1039            called "current" as "current_branch", to prevent a collision with the
1040            Linux macro when compiled as a kernel module.
1042        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1043            prevent a collision with the Linux macro when compiled as a kernel
1044            module.
1046    8.  In pcre_compile() there are some checks for integer overflows that used to
1047        cast potentially large values to (double). This has been changed to that
1048        when building, a check for int64_t is made, and if it is found, it is used
1049        instead, thus avoiding the use of floating point arithmetic. (There is no
1050        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1051        double.
1053    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1054        2005 (difference between two addresses compared to an unsigned value).
1056    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1057        custom one, because of the following reported problem in Windows:
1059          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1060              under Win32.
1061          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1062              therefore missing the function definition.
1063          - The compiler thus generates a "C" signature for the test function.
1064          - The linker fails to find the "C" function.
1065          - PCRE fails to configure if asked to do so against libbz2.
1067    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1068        messages were output:
1070          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1071          rerunning libtoolize, to keep the correct libtool macros in-tree.
1072          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1074        I have done both of these things.
1076    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1077        most of the time, it *can* run out if it is given a pattern that contains a
1078        runaway infinite recursion. I updated the discussion in the pcrestack man
1079        page.
1081    13. Now that we have gone to the x.xx style of version numbers, the minor
1082        version may start with zero. Using 08 or 09 is a bad idea because users
1083        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1084        interpreted as invalid octal numbers. I've updated the previous comment in
1085        configure.ac, and also added a check that gives an error if 08 or 09 are
1086        used.
1088    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1089        causing partial matching to fail when the end of the subject matched \W
1090        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1092    15. There were some discrepancies between the declarations in pcre_internal.h
1093        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1094        their definitions. The declarations used "const uschar *" and the
1095        definitions used USPTR. Even though USPTR is normally defined as "const
1096        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1097        reported that: "This difference in casting confuses some C++ compilers, for
1098        example, SunCC recognizes above declarations as different functions and
1099        generates broken code for hbpcre." I have changed the declarations to use
1100        USPTR.
1102    16. GNU libtool is named differently on some systems. The autogen.sh script now
1103        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1104        (FreeBSD).
1106    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1107        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1108        comment: "Figure out how to create a longlong from a string: strtoll and
1109        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1110        instance, but it only takes 2 args instead of 3!"
1112    18. A subtle bug concerned with back references has been fixed by a change of
1113        specification, with a corresponding code fix. A pattern such as
1114        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1115        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1116        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1117        same bug. Such groups have to be quantified to be useful, or contained
1118        inside another quantified group. (If there's no repetition, the reference
1119        can never match.) The problem arises because, having left the group and
1120        moved on to the rest of the pattern, a later failure that backtracks into
1121        the group uses the captured value from the final iteration of the group
1122        rather than the correct earlier one. I have fixed this in PCRE by forcing
1123        any group that contains a reference to itself to be an atomic group; that
1124        is, there cannot be any backtracking into it once it has completed. This is
1125        similar to recursive and subroutine calls.
1128    Version 8.00 19-Oct-09
1129    ----------------------
1131    1.  The table for translating pcre_compile() error codes into POSIX error codes
1132        was out-of-date, and there was no check on the pcre_compile() error code
1133        being within the table. This could lead to an OK return being given in
1134        error.
1136    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1137        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1138        in a Windows environment.
1140    3.  The pcregrep --count option prints the count for each file even when it is
1141        zero, as does GNU grep. However, pcregrep was also printing all files when
1142        --files-with-matches was added. Now, when both options are given, it prints
1143        counts only for those files that have at least one match. (GNU grep just
1144        prints the file name in this circumstance, but including the count seems
1145        more useful - otherwise, why use --count?) Also ensured that the
1146        combination -clh just lists non-zero counts, with no names.
1148    4.  The long form of the pcregrep -F option was incorrectly implemented as
1149        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1150        but it seems right to fix it, and I didn't think it was worth preserving
1151        the old behaviour.
1153    5.  The command line items --regex=pattern and --regexp=pattern were not
1154        recognized by pcregrep, which required --regex pattern or --regexp pattern
1155        (with a space rather than an '='). The man page documented the '=' forms,
1156        which are compatible with GNU grep; these now work.
1158    6.  No libpcreposix.pc file was created for pkg-config; there was just
1159        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1161    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1162        when UCP support is not needed, by modifying the Python script that
1163        generates it from Unicode data files. This should not matter if the module
1164        is correctly used as a library, but I received one complaint about 50K of
1165        unwanted data. My guess is that the person linked everything into his
1166        program rather than using a library. Anyway, it does no harm.
1168    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1169        was a minimum greater than 1 for a wide character in a possessive
1170        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1171        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1172        character. Chaos in the form of incorrect output or a compiling loop could
1173        result.
1175    9.  The restrictions on what a pattern can contain when partial matching is
1176        requested for pcre_exec() have been removed. All patterns can now be
1177        partially matched by this function. In addition, if there are at least two
1178        slots in the offset vector, the offset of the earliest inspected character
1179        for the match and the offset of the end of the subject are set in them when
1180        PCRE_ERROR_PARTIAL is returned.
1182    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1183        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1184        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1185        and may be more useful for multi-segment matching.
1187    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1188        used to be given if ever the end of the subject was reached; now it is
1189        given only if matching could not proceed because another character was
1190        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1191        string "Z", which now yields "no match" instead of "partial match". In the
1192        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1193        final character ended with (*FAIL).
1195    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1196        if the pattern had a "must contain" character that was already found in the
1197        earlier partial match, unless partial matching was again requested. For
1198        example, with the pattern /dog.(body)?/, the "must contain" character is
1199        "g". If the first part-match was for the string "dog", restarting with
1200        "sbody" failed. This bug has been fixed.
1202    13. The string returned by pcre_dfa_exec() after a partial match has been
1203        changed so that it starts at the first inspected character rather than the
1204        first character of the match. This makes a difference only if the pattern
1205        starts with a lookbehind assertion or \b or \B (\K is not supported by
1206        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1207        matching functions compatible, and I think it's the right thing to do.
1209    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1210        so that the demonstration program is easily available in environments where
1211        PCRE has not been installed from source.
1213    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1214        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1215        library.
1217    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1218        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1219        is not the first non-POSIX option to be added. Clearly some people find
1220        these options useful.
1222    17. If a caller to the POSIX matching function regexec() passes a non-zero
1223        value for nmatch with a NULL value for pmatch, the value of
1224        nmatch is forced to zero.
1226    18. RunGrepTest did not have a test for the availability of the -u option of
1227        the diff command, as RunTest does. It now checks in the same way as
1228        RunTest, and also checks for the -b option.
1230    19. If an odd number of negated classes containing just a single character
1231        interposed, within parentheses, between a forward reference to a named
1232        subpattern and the definition of the subpattern, compilation crashed with
1233        an internal error, complaining that it could not find the referenced
1234        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1235        [The bug was that it was starting one character too far in when skipping
1236        over the character class, thus treating the ] as data rather than
1237        terminating the class. This meant it could skip too much.]
1239    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1240        /g option in pcretest when the pattern contains \K, which makes it possible
1241        to have an empty string match not at the start, even when the pattern is
1242        anchored. Updated pcretest and pcredemo to use this option.
1244    21. If the maximum number of capturing subpatterns in a recursion was greater
1245        than the maximum at the outer level, the higher number was returned, but
1246        with unset values at the outer level. The correct (outer level) value is
1247        now given.
1249    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1250        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1251        make it do so. The string so far is captured, making this feature
1252        compatible with Perl.
1254    23. The tests have been re-organized, adding tests 11 and 12, to make it
1255        possible to check the Perl 5.10 features against Perl 5.10.
1257    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1258        pattern matches a fixed length string. PCRE did not allow this; now it
1259        does. Neither allows recursion.
1261    25. I finally figured out how to implement a request to provide the minimum
1262        length of subject string that was needed in order to match a given pattern.
1263        (It was back references and recursion that I had previously got hung up
1264        on.) This code has now been added to pcre_study(); it finds a lower bound
1265        to the length of subject needed. It is not necessarily the greatest lower
1266        bound, but using it to avoid searching strings that are too short does give
1267        some useful speed-ups. The value is available to calling programs via
1268        pcre_fullinfo().
1270    26. While implementing 25, I discovered to my embarrassment that pcretest had
1271        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1272        study optimizations had never been tested with that matching function.
1273        Oops. What is worse, even when it was passed study data, there was a bug in
1274        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1275        were also very few tests of studied patterns with pcre_dfa_exec().
1277    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1278        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1279        on the other side of the coin, they are no longer allowed to have different
1280        names, because these cannot be distinguished in PCRE, and this has caused
1281        confusion. (This is a difference from Perl.)
1283    28. When duplicate subpattern names are present (necessarily with different
1284        numbers, as required by 27 above), and a test is made by name in a
1285        conditional pattern, either for a subpattern having been matched, or for
1286        recursion in such a pattern, all the associated numbered subpatterns are
1287        tested, and the overall condition is true if the condition is true for any
1288        one of them. This is the way Perl works, and is also more like the way
1289        testing by number works.
1292    Version 7.9 11-Apr-09
1293    ---------------------
1295    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1296        (pcretest), all targets were linked against these libraries. This included
1297        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1298        libraries. This caused unwanted dependencies to be created. This problem
1299        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1300        pcretest is linked with readline.
1302    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1303        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1304        moved outside it again, because FALSE and TRUE are already defined in AIX,
1305        but BOOL is not.
1307    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1308        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1310    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1311        hyphens) following file names and line numbers when outputting matching
1312        lines. This is not true; no spaces are inserted. I have also clarified the
1313        wording for the --colour (or --color) option.
1315    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1316        was not coloured; this is different to GNU grep, so I have changed it to be
1317        the same.
1319    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1320        each matching line was coloured. Now it goes on to look for further matches
1321        of any of the test patterns, which is the same behaviour as GNU grep.
1323    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1324        doesn't make sense to accept an empty string match in pcregrep, so I have
1325        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1326        seems to be how GNU grep behaves.
1328    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1329        start or after a newline", because the conditional assertion was not being
1330        correctly handled. The rule now is that both the assertion and what follows
1331        in the first alternative must satisfy the test.
1333    9.  If auto-callout was enabled in a pattern with a conditional group whose
1334        condition was an assertion, PCRE could crash during matching, both with
1335        pcre_exec() and pcre_dfa_exec().
1337    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1338        used for matching.
1340    11. Unicode property support in character classes was not working for
1341        characters (bytes) greater than 127 when not in UTF-8 mode.
1343    12. Added the -M command line option to pcretest.
1345    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1347    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1349    16. Added comments and documentation about mis-use of no_arg in the C++
1350        wrapper.
1352    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1353        from Martin Jerabek that uses macro names for all relevant character and
1354        string constants.
1356    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1357        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1358        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1359        these, but not everybody uses configure.
1361    19. A conditional group that had only one branch was not being correctly
1362        recognized as an item that could match an empty string. This meant that an
1363        enclosing group might also not be so recognized, causing infinite looping
1364        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1365        with the subject "ab", where knowledge that the repeated group can match
1366        nothing is needed in order to break the loop.
1368    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1369        exec(), but without supplying a callout function, matching went wrong.
1371    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1372        leak if the size of the offset vector was greater than 30. When the vector
1373        is smaller, the saved offsets during recursion go onto a local stack
1374        vector, but for larger vectors malloc() is used. It was failing to free
1375        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1376        error, in fact).
1378    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1379        heapframe that is used only when UTF-8 support is enabled. This caused no
1380        problem, but was untidy.
1382    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1383        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1384        included within another project.
1386    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1387        slightly modified by me:
1389          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1390              not building pcregrep.
1392          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1393              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1395    25. Forward references, both numeric and by name, in patterns that made use of
1396        duplicate group numbers, could behave incorrectly or give incorrect errors,
1397        because when scanning forward to find the reference group, PCRE was not
1398        taking into account the duplicate group numbers. A pattern such as
1399        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1401    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1402        the feature of a custom pointer more persuasive (as requested by a user).
1404    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1405        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1406        pre-defined.
1408    28. Added support for (*UTF8) at the start of a pattern.
1410    29. Arrange for flags added by the "release type" setting in CMake to be shown
1411        in the configuration summary.
1414    Version 7.8 05-Sep-08
1415    ---------------------
1417    1.  Replaced UCP searching code with optimized version as implemented for Ad
1418        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1419        stage table and inline lookup instead of a function, giving speed ups of 2
1420        to 5 times on some simple patterns that I tested. Permission was given to
1421        distribute the MultiStage2.py script that generates the tables (it's not in
1422        the tarball, but is in the Subversion repository).
1424    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1425        scripts.
1427    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1428        a group with a zero qualifier. The result of the study could be incorrect,
1429        or the function might crash, depending on the pattern.
1431    4.  Caseless matching was not working for non-ASCII characters in back
1432        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1433        It now works when Unicode Property Support is available.
1435    5.  In pcretest, an escape such as \x{de} in the data was always generating
1436        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1437        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1438        truncation.
1440    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1442    7.  Added two (int) casts to pcregrep when printing the difference of two
1443        pointers, in case they are 64-bit values.
1445    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1446        test 2 if it fails.
1448    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1449        and a #define of that name to empty if it is not externally set. This is to
1450        allow users of MSVC to set it if necessary.
1452    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1453        the convenience functions in the pcre_get.c source file.
1455    11. An option change at the start of a pattern that had top-level alternatives
1456        could cause overwriting and/or a crash. This command provoked a crash in
1457        some environments:
1459          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1461        This potential security problem was recorded as CVE-2008-2371.
1463    12. For a pattern where the match had to start at the beginning or immediately
1464        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1465        pcre_dfa_exec() could read past the end of the passed subject if there was
1466        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1467        pcretest so that it places the subject at the end of its malloc-ed buffer.
1469    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1470        exec() might read past the end of the data buffer in UTF-8 mode.
1472    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1473        the data contained the byte 0x85 as part of a UTF-8 character within its
1474        first line. This applied both to normal and DFA matching.
1476    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1477        /^[^d]*?$/8 failed to match "abc".
1479    16. Added a missing copyright notice to pcrecpp_internal.h.
1481    17. Make it more clear in the documentation that values returned from
1482        pcre_exec() in ovector are byte offsets, not character counts.
1484    18. Tidied a few places to stop certain compilers from issuing warnings.
1486    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1487        supplied by Stefan Weber. I made a further small update for 7.8 because
1488        there is a change of source arrangements: the pcre_searchfuncs.c module is
1489        replaced by pcre_ucd.c.
1492    Version 7.7 07-May-08
1493    ---------------------
1495    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1496        a string to a long long, pretend we don't even have a long long." This is
1497        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1499    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1500        pre-7.6 versions, which defined a global no_arg variable instead of putting
1501        it in the RE class. (See also #8 below.)
1503    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1504        Lopes.
1506    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1508        (1) The include/exclude patterns were being applied to the whole pathnames
1509            of files, instead of just to the final components.
1511        (2) If there was more than one level of directory, the subdirectories were
1512            skipped unless they satisfied the include/exclude conditions. This is
1513            inconsistent with GNU grep (and could even be seen as contrary to the
1514            pcregrep specification - which I improved to make it absolutely clear).
1515            The action now is always to scan all levels of directory, and just
1516            apply the include/exclude patterns to regular files.
1518    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1519        --exclude_dir in the tests to avoid scanning .svn directories.
1521    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1522        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1523        doesn't support NULs in patterns.
1525    7.  Added some missing "const"s to declarations of static tables in
1526        pcre_compile.c and pcre_dfa_exec.c.
1528    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1529        caused by fix #2  above. (Subsequently also a second patch to fix the
1530        first patch. And a third patch - this was a messy problem.)
1532    9.  Applied Craig's patch to remove the use of push_back().
1534    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1535        matching function regexec().
1537    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1538        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1539        references. PCRE supports relative numbers with this syntax (I don't think
1540        Oniguruma does).
1542    12. Previously, a group with a zero repeat such as (...){0} was completely
1543        omitted from the compiled regex. However, this means that if the group
1544        was called as a subroutine from elsewhere in the pattern, things went wrong
1545        (an internal error was given). Such groups are now left in the compiled
1546        pattern, with a new opcode that causes them to be skipped at execution
1547        time.
1549    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1550        to the way PCRE behaves:
1552        (a) A lone ] character is dis-allowed (Perl treats it as data).
1554        (b) A back reference to an unmatched subpattern matches an empty string
1555            (Perl fails the current match path).
1557        (c) A data ] in a character class must be notated as \] because if the
1558            first data character in a class is ], it defines an empty class. (In
1559            Perl it is not possible to have an empty class.) The empty class []
1560            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1561            The negative empty class [^] matches any one character, independently
1562            of the DOTALL setting.
1564    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1565        non-existent subpattern following a character class starting with ']' and
1566        containing () gave an internal compiling error instead of "reference to
1567        non-existent subpattern". Fortunately, when the pattern did exist, the
1568        compiled code was correct. (When scanning forwards to check for the
1569        existencd of the subpattern, it was treating the data ']' as terminating
1570        the class, so got the count wrong. When actually compiling, the reference
1571        was subsequently set up correctly.)
1573    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1574        it was being rejected as not supported by pcre_dfa_exec(), even though
1575        other assertions are supported. I have made pcre_dfa_exec() support
1576        (*FAIL).
1578    16. The implementation of 13c above involved the invention of a new opcode,
1579        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1580        cannot be changed at match time, I realized I could make a small
1581        improvement to matching performance by compiling OP_ALLANY instead of
1582        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1583        on the OP_ANY path.
1585    17. Compiling pcretest on Windows with readline support failed without the
1586        following two fixes: (1) Make the unistd.h include conditional on
1587        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1589    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1590        ncurses library to be included for pcretest when ReadLine support is
1591        requested, but also to allow for it to be overridden. This patch came from
1592        Daniel Bergström.
1594    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1595        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1596        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1597        spotting this.
1600    Version 7.6 28-Jan-08
1601    ---------------------
1603    1.  A character class containing a very large number of characters with
1604        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1605        overflow.
1607    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1608        HAVE_LONG_LONG is not defined.
1610    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1611        bring them up to date and include new features. This patch includes:
1613        - Fixed PH's badly added libz and libbz2 support.
1614        - Fixed a problem with static linking.
1615        - Added pcredemo. [But later removed - see 7 below.]
1616        - Fixed dftables problem and added an option.
1617        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1618            HAVE_LONG_LONG.
1619        - Added readline support for pcretest.
1620        - Added an listing of the option settings after cmake has run.
1622    4.  A user submitted a patch to Makefile that makes it easy to create
1623        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1624        Makefile.am that cause it to include this special target, without
1625        affecting anything else. Note that the same mingw target plus all
1626        the other distribution libraries and programs are now supported
1627        when configuring with CMake (see 6 below) instead of with
1628        Configure/Make.
1630    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1631        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1632        exported in the Windows port". It has not yet been confirmed that the patch
1633        solves the problem, but it does no harm.
1635    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1636        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1637        with CMake, and also correct the comment about stack recursion.
1639    7.  Remove the automatic building of pcredemo from the ./configure system and
1640        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1641        of a program that users should build themselves after PCRE is installed, so
1642        building it automatically is not really right. What is more, it gave
1643        trouble in some build environments.
1645    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1648    Version 7.5 10-Jan-08
1649    ---------------------
1651    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1652        values in parens when parsing an RE using the C++ wrapper."
1654    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1655        Characters greater than 255 were excluded from the class instead of being
1656        included.
1658    3.  The same bug as (2) above applied to negated POSIX classes such as
1659        [:^space:].
1661    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1662        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1663        I have changed it.
1665    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1666        first named subpattern) and a construct such as (?&a) would reference the
1667        first named subpattern whose name started with "a" (in other words, the
1668        length check was missing). Both these problems are fixed. "Subpattern name
1669        expected" is now given for (?&) (a zero-length name), and this patch also
1670        makes it give the same error for \k'' (previously it complained that that
1671        was a reference to a non-existent subpattern).
1673    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1674        this is right because (?- can be followed by option settings as well as by
1675        digits. I have, however, made the messages clearer.
1677    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1678        than the number used in the conditional) now cause a compile-time error.
1679        This is actually not compatible with Perl, which accepts such patterns, but
1680        treats the conditional as always being FALSE (as PCRE used to), but it
1681        seems to me that giving a diagnostic is better.
1683    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1684        and messages.
1686    9.  Fix two occurrences of "backslash" in comments that should have been
1687        "backspace".
1689    10. Remove two redundant lines of code that can never be obeyed (their function
1690        was moved elsewhere).
1692    11. The program that makes PCRE's Unicode character property table had a bug
1693        which caused it to generate incorrect table entries for sequences of
1694        characters that have the same character type, but are in different scripts.
1695        It amalgamated them into a single range, with the script of the first of
1696        them. In other words, some characters were in the wrong script. There were
1697        thirteen such cases, affecting characters in the following ranges:
1699          U+002b0 - U+002c1
1700          U+0060c - U+0060d
1701          U+0061e - U+00612
1702          U+0064b - U+0065e
1703          U+0074d - U+0076d
1704          U+01800 - U+01805
1705          U+01d00 - U+01d77
1706          U+01d9b - U+01dbf
1707          U+0200b - U+0200f
1708          U+030fc - U+030fe
1709          U+03260 - U+0327f
1710          U+0fb46 - U+0fbb1
1711          U+10450 - U+1049d
1713    12. The -o option (show only the matching part of a line) for pcregrep was not
1714        compatible with GNU grep in that, if there was more than one match in a
1715        line, it showed only the first of them. It now behaves in the same way as
1716        GNU grep.
1718    13. If the -o and -v options were combined for pcregrep, it printed a blank
1719        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1720        does the same. The return code can be used to tell if there were any
1721        non-matching lines.
1723    14. Added --file-offsets and --line-offsets to pcregrep.
1725    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1726        infinitely looping recursion. The bug was that positive lookaheads were not
1727        being skipped when checking for a possible empty match (negative lookaheads
1728        and both kinds of lookbehind were skipped).
1730    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1731        inclusion of <windows.h> to before rather than after the definition of
1732        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1734    17. Specifying a possessive quantifier with a specific limit for a Unicode
1735        character property caused pcre_compile() to compile bad code, which led at
1736        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1737        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1738        caused the error; without that there was no problem.
1740    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1742    19. Added --enable-pcretest-libreadline.
1744    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1745        RE::GlobalReplace(). As a result, the number of replacements returned was
1746        double what it should be. I removed one of the increments, but Craig sent a
1747        later patch that removed the other one (the right fix) and added unit tests
1748        that check the return values (which was not done before).
1750    21. Several CMake things:
1752        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1753            the names libpcre and libpcreposix, not just pcre and pcreposix.
1755        (2) The above change means that pcretest and pcregrep are now correctly
1756            linked with the newly-built libraries, not previously installed ones.
1760    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1761        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1762        UTF-8 newline character). The key issue is that the pattern starts .*;
1763        this means that the match must be either at the beginning, or after a
1764        newline. The bug was in the code for advancing after a failed match and
1765        checking that the new position followed a newline. It was not taking
1766        account of UTF-8 characters correctly.
1768    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1769        character classes. PCRE was not treating the sequence [:...:] as a
1770        character class unless the ... were all letters. Perl, however, seems to
1771        allow any characters between [: and :], though of course it rejects as
1772        unknown any "names" that contain non-letters, because all the known class
1773        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1774        for example, whereas PCRE did not - it did not recognize a POSIX character
1775        class. This seemed a bit dangerous, so the code has been changed to be
1776        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1777        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1778        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1779        Perl does, and where it didn't before.
1781    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1782        Windows environments %n is disabled by default.
1785    Version 7.4 21-Sep-07
1786    ---------------------
1788    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1789        means that a class such as [\s] counted as "explicit reference to CR or
1790        LF". That isn't really right - the whole point of the change was to try to
1791        help when there was an actual mention of one of the two characters. So now
1792        the change happens only if \r or \n (or a literal CR or LF) character is
1793        encountered.
1795    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1796        of both had grown to the point where there were only 3 bits left.
1797        Fortunately, there was spare space in the data structure, and so I have
1798        moved the internal flags into a new 16-bit field to free up more option
1799        bits.
1801    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1802        but did not set the internal JCHANGED flag - either of these is enough to
1803        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1804        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1805        start sets both bits.
1807    4.  Added options (at build time, compile time, exec time) to change \R from
1808        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1810    5.  doc/pcresyntax.html was missing from the distribution.
1812    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1813        compatibility, even though it is no longer used.
1815    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1816        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1817        windows.h file is present (where different names are used). [This was
1818        reversed later after testing - see 16 below.]
1820    8.  Changed all #include <config.h> to #include "config.h". There were also
1821        some further <pcre.h> cases that I changed to "pcre.h".
1823    9.  When pcregrep was used with the --colour option, it missed the line ending
1824        sequence off the lines that it output.
1826    10. It was pointed out to me that arrays of string pointers cause lots of
1827        relocations when a shared library is dynamically loaded. A technique of
1828        using a single long string with a table of offsets can drastically reduce
1829        these. I have refactored PCRE in four places to do this. The result is
1830        dramatic:
1832          Originally:                          290
1833          After changing UCP table:            187
1834          After changing error message table:   43
1835          After changing table of "verbs"       36
1836          After changing table of Posix names   22
1838        Thanks to the folks working on Gregex for glib for this insight.
1840    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1841        unicode-properties was also set.
1843    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1845    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1846        checked only for CRLF.
1848    14. Added casts to pcretest.c to avoid compiler warnings.
1850    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1852    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1853        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1854        entirely. This removes changes made in 7 above.
1856    17. The CMake files have been updated, and there is now more information about
1857        building with CMake in the NON-UNIX-USE document.
1860    Version 7.3 28-Aug-07
1861  ---------------------  ---------------------
1863   1. In the rejigging of the build system that eventually resulted in 7.1, the   1. In the rejigging of the build system that eventually resulted in 7.1, the
# Line 64  Version 7.3 05-Jul-07 Line 1920  Version 7.3 05-Jul-07
1920      dynamic way, which I have now done. The artificial limitation on group      dynamic way, which I have now done. The artificial limitation on group
1921      length has been removed - we now have only the limit on the total length of      length has been removed - we now have only the limit on the total length of
1922      the compiled pattern, which depends on the LINK_SIZE setting.      the compiled pattern, which depends on the LINK_SIZE setting.
1924  10. Fixed a bug in the documentation for get/copy named substring when  10. Fixed a bug in the documentation for get/copy named substring when
1925      duplicate names are permitted. If none of the named substrings are set, the      duplicate names are permitted. If none of the named substrings are set, the
1926      functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an      functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1927      empty string.      empty string.
1929  11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E  11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1930      instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,      instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1931      because the ] is interpreted as the first data character and the      because the ] is interpreted as the first data character and the
1932      terminating ] is not found. PCRE has been made compatible with Perl in this      terminating ] is not found. PCRE has been made compatible with Perl in this
1933      regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could      regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1934      cause memory overwriting.      cause memory overwriting.
1936    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1937        string has been matched (to stop an infinite loop). It was not recognizing
1938        a conditional subpattern that could match an empty string if that
1939        subpattern was within another subpattern. For example, it looped when
1940        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1941        condition was not nested. This bug has been fixed.
1943    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1944        past the start of the subject in the presence of bytes with the top bit
1945        set, for example "\x8aBCD".
1947    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1948        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1950    14. Optimized (?!) to (*FAIL).
1952    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1953        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1954        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1955        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1956        does: it's just the validity check that is more restrictive.
1958    16. Inserted checks for integer overflows during escape sequence (backslash)
1959        processing, and also fixed erroneous offset values for syntax errors during
1960        backslash processing.
1962    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1963        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1965    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1966        caused an overrun.
1968    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1969        something other than just ASCII characters) inside a group that had an
1970        unlimited repeat caused a loop at compile time (while checking to see
1971        whether the group could match an empty string).
1973    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1974        [\P{Any}] did so. (Error in the code for printing property names.)
1976    21. An orphan \E inside a character class could cause a crash.
1978    22. A repeated capturing bracket such as (A)? could cause a wild memory
1979        reference during compilation.
1981    23. There are several functions in pcre_compile() that scan along a compiled
1982        expression for various reasons (e.g. to see if it's fixed length for look
1983        behind). There were bugs in these functions when a repeated \p or \P was
1984        present in the pattern. These operators have additional parameters compared
1985        with \d, etc, and these were not being taken into account when moving along
1986        the compiled data. Specifically:
1988        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1989            length.
1991        (b) An item such as \pL+ within a repeated group could cause crashes or
1992            loops.
1994        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1995            "reference to non-existent subpattern" error.
1997        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1999    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2000        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2002    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2004    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2005        character were causing crashes (broken optimization).
2007    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2008        \p or \P) caused a compile-time loop.
2010    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2011        break. For example, the unstudied pattern [\r\n]A does not match the string
2012        "\r\nA" because change 7.0/46 below moves the current point on by two
2013        characters after failing to match at the start. However, the pattern \nA
2014        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2015        the same is true. There doesn't seem any very clean way out of this, but
2016        what I have chosen to do makes the common cases work: PCRE now takes note
2017        of whether there can be an explicit match for \r or \n anywhere in the
2018        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2019        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2020        pattern has explicit CR or LF references.
2022    29. Added (*CR) etc for changing newline setting at start of pattern.
2025  Version 7.2 19-Jun-07  Version 7.2 19-Jun-07

Removed from v.205  
changed lines
  Added in v.977

  ViewVC Help
Powered by ViewVC 1.1.5