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

Diff of /code/trunk/ChangeLog

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

revision 172 by ph10, Tue Jun 5 10:40:13 2007 UTC revision 971 by ph10, Fri Jun 1 16:29:43 2012 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 7.2 05-June-07  Version 8.31
5    -----------------------------
6    
7    1.  Fixing a wrong JIT test case and some compiler warnings.
8    
9    2.  Removed a bashism from the RunTest script.
10    
11    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
12        to unsigned type, result still unsigned" that was given by an MS compiler
13        on encountering the code "-sizeof(xxx)".
14    
15    4.  Partial matching support is added to the JIT compiler.
16    
17    5.  Fixed several bugs concerned with partial matching of items that consist
18        of more than one character:
19    
20        (a) /^(..)\1/ did not partially match "aba" because checking references was
21            done on an "all or nothing" basis. This also applied to repeated
22            references.
23    
24        (b) \R did not give a hard partial match if \r was found at the end of the
25            subject.
26    
27        (c) \X did not give a hard partial match after matching one or more
28            characters at the end of the subject.
29    
30        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
31            a partial match for the string "\r".
32    
33        (e) When newline was set to CRLF, the metacharacter "." did not recognize
34            a partial match for a CR character at the end of the subject string.
35    
36    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
37        running pcretest, the text "(JIT)" added to the output whenever JIT is
38        actually used to run the match.
39    
40    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
41        or /S+[+] with a digit between 1 and 7.
42    
43    8.  OP_NOT now supports any UTF character not just single-byte ones.
44    
45    9.  (*MARK) control verb is now supported by the JIT compiler.
46    
47    10. The command "./RunTest list" lists the available tests without actually
48        running any of them. (Because I keep forgetting what they all are.)
49    
50    11. Add PCRE_INFO_MAXLOOKBEHIND.
51    
52    12. Applied a (slightly modified) user-supplied patch that improves performance
53        when the heap is used for recursion (compiled with --disable-stack-for-
54        recursion). Instead of malloc and free for each heap frame each time a
55        logical recursion happens, frames are retained on a chain and re-used where
56        possible. This sometimes gives as much as 30% improvement.
57    
58    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
59        call.
60    
61    14. As documented, (*COMMIT) is now confined to within a positive assertion.
62    
63    15. It is now possible to link pcretest with libedit as an alternative to
64        libreadline.
65    
66    16. (*COMMIT) control verb is now supported by the JIT compiler.
67    
68    17. The Unicode data tables have been updated to Unicode 6.1.0.
69    
70    18. Added --file-list option to pcregrep.
71    
72    19. Added binary file support to pcregrep, including the -a, --binary-files,
73        -I, and --text options.
74    
75    20. The madvise function is renamed for posix_madvise for QNX compatibility
76        reasons. Fixed by Giuseppe D'Angelo.
77    
78    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
79    
80    22. Changed the option for creating symbolic links for 16-bit man pages from
81        -s to -sf so that re-installing does not cause issues.
82    
83    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
84    
85    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
86        matches in certain environments (the workspace was not being correctly
87        retained). Also added to pcre_dfa_exec() a simple plausibility check on
88        some of the workspace data at the beginning of a restart.
89    
90    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
91        was not doing so when it should - probably a typo introduced by SVN 528
92        (change 8.10/14).
93    
94    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
95        \w+ when the character tables indicated that \x{c4} was a word character.
96        There were several related cases, all because the tests for doing a table
97        lookup were testing for characters less than 127 instead of 255.
98    
99    27. If a pattern contains capturing parentheses that are not used in a match,
100        their slots in the ovector are set to -1. For those that are higher than
101        any matched groups, this happens at the end of processing. In the case when
102        there were back references that the ovector was too small to contain
103        (causing temporary malloc'd memory to be used during matching), and the
104        highest capturing number was not used, memory off the end of the ovector
105        was incorrectly being set to -1. (It was using the size of the temporary
106        memory instead of the true size.)
107    
108    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
109        ovector size, it uses memory at the end of the block that it has got.
110    
111    29. Check for an overlong MARK name and give an error at compile time. The
112        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
113    
114    30. JIT compiler update.
115    
116    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
117        Rill for the patch.
118    
119    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
120    
121    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
122    
123    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
124        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
125        was enabled.
126    
127    
128    Version 8.30 04-February-2012
129    -----------------------------
130    
131    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
132        name is defined in ctype.h.
133    
134    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
135        only in quite long subpatterns.
136    
137    3.  Removed the function pcre_info(), which has been obsolete and deprecated
138        since it was replaced by pcre_fullinfo() in February 2000.
139    
140    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
141        match a (*MARK), and the match failed at the start of the subject, a
142        reference to memory before the start of the subject could occur. This bug
143        was introduced by fix 17 of release 8.21.
144    
145    5.  A reference to an unset group with zero minimum repetition was giving
146        totally wrong answers (in non-JavaScript-compatibility mode). For example,
147        /(another)?(\1?)test/ matched against "hello world test". This bug was
148        introduced in release 8.13.
149    
150    6.  Add support for 16-bit character strings (a large amount of work involving
151        many changes and refactorings).
152    
153    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
154        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
155        from a file.
156    
157    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
158        rounding is not applied in this particular case).
159    
160    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
161        if they appear, or are escaped, in patterns.
162    
163    10. Get rid of a number of -Wunused-but-set-variable warnings.
164    
165    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
166        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
167        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
168        also returns the mark "x". This bug applied to capturing parentheses,
169        non-capturing parentheses, and atomic parentheses. It also applied to some
170        assertions.
171    
172    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
173        information out of configure.ac instead of relying on pcre.h.generic, which
174        is not stored in the repository.
175    
176    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
177        -lreadline.
178    
179    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
180    
181    15. Applied Graycode's patch to put the top-level frame on the stack rather
182        than the heap when not using the stack for recursion. This gives a
183        performance improvement in many cases when recursion is not deep.
184    
185    16. Experimental code added to "pcretest -C" to output the stack frame size.
186    
187    
188    Version 8.21 12-Dec-2011
189    ------------------------
190    
191    1.  Updating the JIT compiler.
192    
193    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
194        are added as well.
195    
196    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
197        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
198        calling _pcre_jit_exec. Some extra comments are added.
199    
200    4.  (*MARK) settings inside atomic groups that do not contain any capturing
201        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
202        was introduced by change 18 for 8.20.
203    
204    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
205        ECMA-262 standard.
206    
207    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
208        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
209        This bug was probably introduced by change 9 of 8.13.
210    
211    7.  While fixing 6 above, I noticed that a number of other items were being
212        incorrectly rejected as "not fixed length". This arose partly because newer
213        opcodes had not been added to the fixed-length checking code. I have (a)
214        corrected the bug and added tests for these items, and (b) arranged for an
215        error to occur if an unknown opcode is encountered while checking for fixed
216        length instead of just assuming "not fixed length". The items that were
217        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
218        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
219        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
220    
221    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
222        being incorrectly compiled and would have given unpredicatble results.
223    
224    9.  A possessively repeated subpattern with minimum repeat count greater than
225        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
226        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
227        the first (A) could occur when it should not.
228    
229    10. Add a cast and remove a redundant test from the code.
230    
231    11. JIT should use pcre_malloc/pcre_free for allocation.
232    
233    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
234        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
235        is anything other than /usr, -L is still shown).
236    
237    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
238    
239    14. Perl does not support \N without a following name in a [] class; PCRE now
240        also gives an error.
241    
242    15. If a forward reference was repeated with an upper limit of around 2000,
243        it caused the error "internal error: overran compiling workspace". The
244        maximum number of forward references (including repeats) was limited by the
245        internal workspace, and dependent on the LINK_SIZE. The code has been
246        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
247        the default depends on LINK_SIZE. There is a new upper limit (for safety)
248        of around 200,000 forward references. While doing this, I also speeded up
249        the filling in of repeated forward references.
250    
251    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
252        incorrectly expecting the subject to contain another "a" after the start.
253    
254    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
255        in the match, the SKIP should be ignored. This was not happening; instead
256        the SKIP was being treated as NOMATCH. For patterns such as
257        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
258        tested.
259    
260    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
261        now much more compatible with Perl, in particular in cases where the result
262        is a non-match for a non-anchored pattern. For example, if
263        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
264        "m", where previously it did not return a name. A side effect of this
265        change is that for partial matches, the last encountered mark name is
266        returned, as for non matches. A number of tests that were previously not
267        Perl-compatible have been moved into the Perl-compatible test files. The
268        refactoring has had the pleasing side effect of removing one argument from
269        the match() function, thus reducing its stack requirements.
270    
271    19. If the /S+ option was used in pcretest to study a pattern using JIT,
272        subsequent uses of /S (without +) incorrectly behaved like /S+.
273    
274    21. Retrieve executable code size support for the JIT compiler and fixing
275        some warnings.
276    
277    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
278        not work when the shorter character appeared right at the end of the
279        subject string.
280    
281    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
282        systems.
283    
284    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
285        output it when the /M option is used in pcretest.
286    
287    25. The CheckMan script was not being included in the distribution. Also, added
288        an explicit "perl" to run Perl scripts from the PrepareRelease script
289        because this is reportedly needed in Windows.
290    
291    26. If study data was being save in a file and studying had not found a set of
292        "starts with" bytes for the pattern, the data written to the file (though
293        never used) was taken from uninitialized memory and so caused valgrind to
294        complain.
295    
296    27. Updated RunTest.bat as provided by Sheri Pierce.
297    
298    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
299    
300    29. Computation of memory usage for the table of capturing group names was
301        giving an unnecessarily large value.
302    
303    
304    Version 8.20 21-Oct-2011
305    ------------------------
306    
307    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
308        a POSIX class. After further experiments with Perl, which convinced me that
309        Perl has bugs and confusions, a closing square bracket is no longer allowed
310        in a POSIX name. This bug also affected patterns with classes that started
311        with full stops.
312    
313    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
314        captured substring, but while checking the failing first alternative,
315        substring 1 is temporarily captured. If the output vector supplied to
316        pcre_exec() was not big enough for this capture, the yield of the function
317        was still zero ("insufficient space for captured substrings"). This cannot
318        be totally fixed without adding another stack variable, which seems a lot
319        of expense for a edge case. However, I have improved the situation in cases
320        such as /(a)(b)x|abc/ matched against "abc", where the return code
321        indicates that fewer than the maximum number of slots in the ovector have
322        been set.
323    
324    3.  Related to (2) above: when there are more back references in a pattern than
325        slots in the output vector, pcre_exec() uses temporary memory during
326        matching, and copies in the captures as far as possible afterwards. It was
327        using the entire output vector, but this conflicts with the specification
328        that only 2/3 is used for passing back captured substrings. Now it uses
329        only the first 2/3, for compatibility. This is, of course, another edge
330        case.
331    
332    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
333        main code base, and can be used by building with --enable-jit. When this is
334        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
335        runtime --no-jit option is given.
336    
337    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
338        ovector, the return from the function was zero, implying that there were
339        other matches that did not fit. The correct "exactly full" value is now
340        returned.
341    
342    6.  If a subpattern that was called recursively or as a subroutine contained
343        (*PRUNE) or any other control that caused it to give a non-standard return,
344        invalid errors such as "Error -26 (nested recursion at the same subject
345        position)" or even infinite loops could occur.
346    
347    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
348        computing the minimum length on reaching *ACCEPT, and so ended up with the
349        wrong value of 1 rather than 0. Further investigation indicates that
350        computing a minimum subject length in the presence of *ACCEPT is difficult
351        (think back references, subroutine calls), and so I have changed the code
352        so that no minimum is registered for a pattern that contains *ACCEPT.
353    
354    8.  If (*THEN) was present in the first (true) branch of a conditional group,
355        it was not handled as intended. [But see 16 below.]
356    
357    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
358        Sheri Pierce.
359    
360    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
361        the first byte in a match must be "a".
362    
363    11. Change 17 for 8.13 increased the recursion depth for patterns like
364        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
365        pattern contains any instances of (*THEN). If it does not, the old
366        optimizations are restored. It would be nice to do this on a per-group
367        basis, but at the moment that is not feasible.
368    
369    12. In some environments, the output of pcretest -C is CRLF terminated. This
370        broke RunTest's code that checks for the link size. A single white space
371        character after the value is now allowed for.
372    
373    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
374        For "fr", it uses the Windows-specific input and output files.
375    
376    14. If (*THEN) appeared in a group that was called recursively or as a
377        subroutine, it did not work as intended. [But see next item.]
378    
379    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
380        pattern fragments (but not containing any | characters). If A and B are
381        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
382        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
383        D. In other words, Perl considers parentheses that do not contain any |
384        characters to be part of a surrounding alternative, whereas PCRE was
385        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
386        differently. PCRE now behaves in the same way as Perl, except in the case
387        of subroutine/recursion calls such as (?1) which have in any case always
388        been different (but PCRE had them first :-).
389    
390    16. Related to 15 above: Perl does not treat the | in a conditional group as
391        creating alternatives. Such a group is treated in the same way as an
392        ordinary group without any | characters when processing (*THEN). PCRE has
393        been changed to match Perl's behaviour.
394    
395    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
396        RunGrepTest script failed.
397    
398    18. Change 22 for version 13 caused atomic groups to use more stack. This is
399        inevitable for groups that contain captures, but it can lead to a lot of
400        stack use in large patterns. The old behaviour has been restored for atomic
401        groups that do not contain any capturing parentheses.
402    
403    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
404        suppress the check for a minimum subject length at run time. (If it was
405        given to pcre_exec() or pcre_dfa_exec() it did work.)
406    
407    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
408        fail to work when decoding hex characters in data strings in EBCDIC
409        environments.
410    
411    21. It appears that in at least one Mac OS environment, the isxdigit() function
412        is implemented as a macro that evaluates to its argument more than once,
413        contravening the C 90 Standard (I haven't checked a later standard). There
414        was an instance in pcretest which caused it to go wrong when processing
415        \x{...} escapes in subject strings. The has been rewritten to avoid using
416        things like p++ in the argument of isxdigit().
417    
418    
419    Version 8.13 16-Aug-2011
420    ------------------------
421    
422    1.  The Unicode data tables have been updated to Unicode 6.0.0.
423    
424    2.  Two minor typos in pcre_internal.h have been fixed.
425    
426    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
427        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
428        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
429    
430    4.  There were a number of related bugs in the code for matching backrefences
431        caselessly in UTF-8 mode when codes for the characters concerned were
432        different numbers of bytes. For example, U+023A and U+2C65 are an upper
433        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
434        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
435        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
436        2-byte code at the end of the subject (it thought there wasn't enough data
437        left).
438    
439    5.  Comprehensive information about what went wrong is now returned by
440        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
441        as the output vector has at least 2 elements. The offset of the start of
442        the failing character and a reason code are placed in the vector.
443    
444    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
445        now returned is for the first byte of the failing character, instead of the
446        last byte inspected. This is an incompatible change, but I hope it is small
447        enough not to be a problem. It makes the returned offset consistent with
448        pcre_exec() and pcre_dfa_exec().
449    
450    7.  pcretest now gives a text phrase as well as the error number when
451        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
452        failure, the offset and reason code are output.
453    
454    8.  When \R was used with a maximizing quantifier it failed to skip backwards
455        over a \r\n pair if the subsequent match failed. Instead, it just skipped
456        back over a single character (\n). This seems wrong (because it treated the
457        two characters as a single entity when going forwards), conflicts with the
458        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
459        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
460        has been changed.
461    
462    9.  Some internal refactoring has changed the processing so that the handling
463        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
464        time (the PCRE_DOTALL option was changed this way some time ago: version
465        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
466        which was always a bit of a fudge. It also means that there is one less
467        argument for the match() function, which reduces its stack requirements
468        slightly. This change also fixes an incompatibility with Perl: the pattern
469        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
470    
471    10. More internal refactoring has drastically reduced the number of recursive
472        calls to match() for possessively repeated groups such as (abc)++ when
473        using pcre_exec().
474    
475    11. While implementing 10, a number of bugs in the handling of groups were
476        discovered and fixed:
477    
478        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
479        (a|)*(?1) gave a compile-time internal error.
480        ((a|)+)+  did not notice that the outer group could match an empty string.
481        (^a|^)+   was not marked as anchored.
482        (.*a|.*)+ was not marked as matching at start or after a newline.
483    
484    12. Yet more internal refactoring has removed another argument from the match()
485        function. Special calls to this function are now indicated by setting a
486        value in a variable in the "match data" data block.
487    
488    13. Be more explicit in pcre_study() instead of relying on "default" for
489        opcodes that mean there is no starting character; this means that when new
490        ones are added and accidentally left out of pcre_study(), testing should
491        pick them up.
492    
493    14. The -s option of pcretest has been documented for ages as being an old
494        synonym of -m (show memory usage). I have changed it to mean "force study
495        for every regex", that is, assume /S for every regex. This is similar to -i
496        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
497        using it. It makes it easier to run collections of tests with and without
498        study enabled, and thereby test pcre_study() more easily. All the standard
499        tests are now run with and without -s (but some patterns can be marked as
500        "never study" - see 20 below).
501    
502    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
503        restoration of the capturing data to the outer values was not happening
504        correctly.
505    
506    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
507        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
508        pattern had matched an empty string, and so incorrectly returned a no
509        match.
510    
511    17. There was optimizing code for the last branch of non-capturing parentheses,
512        and also for the obeyed branch of a conditional subexpression, which used
513        tail recursion to cut down on stack usage. Unfortunately, now that there is
514        the possibility of (*THEN) occurring in these branches, tail recursion is
515        no longer possible because the return has to be checked for (*THEN). These
516        two optimizations have therefore been removed. [But see 8.20/11 above.]
517    
518    18. If a pattern containing \R was studied, it was assumed that \R always
519        matched two bytes, thus causing the minimum subject length to be
520        incorrectly computed because \R can also match just one byte.
521    
522    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
523        was incorrectly computed.
524    
525    20. If /S is present twice on a test pattern in pcretest input, it now
526        *disables* studying, thereby overriding the use of -s on the command line
527        (see 14 above). This is necessary for one or two tests to keep the output
528        identical in both cases.
529    
530    21. When (*ACCEPT) was used in an assertion that matched an empty string and
531        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
532    
533    22. When an atomic group that contained a capturing parenthesis was
534        successfully matched, but the branch in which it appeared failed, the
535        capturing was not being forgotten if a higher numbered group was later
536        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
537        group 1 to "a", when in fact it should be unset. This applied to multi-
538        branched capturing and non-capturing groups, repeated or not, and also to
539        positive assertions (capturing in negative assertions does not happen
540        in PCRE) and also to nested atomic groups.
541    
542    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
543        subject after a captured substring, to make it easier to tell which of a
544        number of identical substrings has been captured.
545    
546    24. The way atomic groups are processed by pcre_exec() has been changed so that
547        if they are repeated, backtracking one repetition now resets captured
548        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
549        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
550        "aaa". Previously, it would have been "a". As part of this code
551        refactoring, the way recursive calls are handled has also been changed.
552    
553    25. If an assertion condition captured any substrings, they were not passed
554        back unless some other capturing happened later. For example, if
555        (?(?=(a))a) was matched against "a", no capturing was returned.
556    
557    26. When studying a pattern that contained subroutine calls or assertions,
558        the code for finding the minimum length of a possible match was handling
559        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
560        group 1 called group 2 while simultaneously a separate group 2 called group
561        1). A stack overflow occurred in this case. I have fixed this by limiting
562        the recursion depth to 10.
563    
564    27. Updated RunTest.bat in the distribution to the version supplied by Tom
565        Fortmann. This supports explicit test numbers on the command line, and has
566        argument validation and error reporting.
567    
568    28. An instance of \X with an unlimited repeat could fail if at any point the
569        first character it looked at was a mark character.
570    
571    29. Some minor code refactoring concerning Unicode properties and scripts
572        should reduce the stack requirement of match() slightly.
573    
574    30. Added the '=' option to pcretest to check the setting of unused capturing
575        slots at the end of the pattern, which are documented as being -1, but are
576        not included in the return count.
577    
578    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
579        compiled something random. Now it gives a compile-time error (as does
580        Perl).
581    
582    32. A *MARK encountered during the processing of a positive assertion is now
583        recorded and passed back (compatible with Perl).
584    
585    33. If --only-matching or --colour was set on a pcregrep call whose pattern
586        had alternative anchored branches, the search for a second match in a line
587        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
588        matched the line "0102" twice. The same bug affected patterns that started
589        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
590        twice.
591    
592    34. Previously, PCRE did not allow quantification of assertions. However, Perl
593        does, and because of capturing effects, quantifying parenthesized
594        assertions may at times be useful. Quantifiers are now allowed for
595        parenthesized assertions.
596    
597    35. A minor code tidy in pcre_compile() when checking options for \R usage.
598    
599    36. \g was being checked for fancy things in a character class, when it should
600        just be a literal "g".
601    
602    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
603        appearance of a nested POSIX class supersedes an apparent external class.
604        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
605        unescaped square brackets may also appear as part of class names. For
606        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
607        more like Perl. (But see 8.20/1 above.)
608    
609    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
610        was because it thought it was \N{name}, which is not supported).
611    
612    39. Add minix to OS list not supporting the -S option in pcretest.
613    
614    40. PCRE tries to detect cases of infinite recursion at compile time, but it
615        cannot analyze patterns in sufficient detail to catch mutual recursions
616        such as ((?1))((?2)). There is now a runtime test that gives an error if a
617        subgroup is called recursively as a subpattern for a second time at the
618        same position in the subject string. In previous releases this might have
619        been caught by the recursion limit, or it might have run out of stack.
620    
621    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
622        happen only once. PCRE was, however incorrectly giving a compile time error
623        "recursive call could loop indefinitely" because it cannot analyze the
624        pattern in sufficient detail. The compile time test no longer happens when
625        PCRE is compiling a conditional subpattern, but actual runaway loops are
626        now caught at runtime (see 40 above).
627    
628    42. It seems that Perl allows any characters other than a closing parenthesis
629        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
630        has been changed to be the same.
631    
632    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
633        as not to get warnings when autogen.sh is called. Also changed
634        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
635    
636    44. To help people who use pcregrep to scan files containing exceedingly long
637        lines, the following changes have been made:
638    
639        (a) The default value of the buffer size parameter has been increased from
640            8K to 20K. (The actual buffer used is three times this size.)
641    
642        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
643            PCRE is built.
644    
645        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
646            to be set at run time.
647    
648        (d) Numerical values in pcregrep options can be followed by K or M, for
649            example --buffer-size=50K.
650    
651        (e) If a line being scanned overflows pcregrep's buffer, an error is now
652            given and the return code is set to 2.
653    
654    45. Add a pointer to the latest mark to the callout data block.
655    
656    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
657        partial match of an empty string instead of no match. This was specific to
658        the use of ".".
659    
660    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
661        complete match instead of a partial match. This bug was dependent on both
662        the PCRE_UTF8 and PCRE_DOTALL options being set.
663    
664    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
665        starting byte set, because \b was not being ignored.
666    
667    
668    Version 8.12 15-Jan-2011
669    ------------------------
670    
671    1.  Fixed some typos in the markup of the man pages, and wrote a script that
672        checks for such things as part of the documentation building process.
673    
674    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
675        --match-limit and --recursion-limit options (added for 8.11). In
676        particular, this made one of the standard tests fail. (The integer value
677        went into the wrong half of a long int.)
678    
679    3.  If the --colour option was given to pcregrep with -v (invert match), it
680        did strange things, either producing crazy output, or crashing. It should,
681        of course, ignore a request for colour when reporting lines that do not
682        match.
683    
684    4.  Another pcregrep bug caused similar problems if --colour was specified with
685        -M (multiline) and the pattern match finished with a line ending.
686    
687    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
688        matched in multiline mode, the following line was shown as part of the
689        match. This seems wrong, so I have changed it.
690    
691    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
692        the check for further matches in the same line (so they could be coloured)
693        to overrun the end of the current line. If another match was found, it was
694        incorrectly shown (and then shown again when found in the next line).
695    
696    7.  If pcregrep was compiled under Windows, there was a reference to the
697        function pcregrep_exit() before it was defined. I am assuming this was
698        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
699        reported by a user. I've moved the definition above the reference.
700    
701    
702    Version 8.11 10-Dec-2010
703    ------------------------
704    
705    1.  (*THEN) was not working properly if there were untried alternatives prior
706        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
707        backtracked to try for "b" instead of moving to the next alternative branch
708        at the same level (in this case, to look for "c"). The Perl documentation
709        is clear that when (*THEN) is backtracked onto, it goes to the "next
710        alternative in the innermost enclosing group".
711    
712    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
713        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
714        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
715        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
716        (*THEN).
717    
718    3.  If \s appeared in a character class, it removed the VT character from
719        the class, even if it had been included by some previous item, for example
720        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
721        of \s, but is part of the POSIX "space" class.)
722    
723    4.  A partial match never returns an empty string (because you can always
724        match an empty string at the end of the subject); however the checking for
725        an empty string was starting at the "start of match" point. This has been
726        changed to the "earliest inspected character" point, because the returned
727        data for a partial match starts at this character. This means that, for
728        example, /(?<=abc)def/ gives a partial match for the subject "abc"
729        (previously it gave "no match").
730    
731    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
732        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
733        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
734        has an implication that the given string is incomplete (because a partial
735        match is preferred over a full match). For this reason, these items now
736        give a partial match in this situation. [Aside: previously, the one case
737        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
738        match rather than a full match, which was wrong by the old rules, but is
739        now correct.]
740    
741    6.  There was a bug in the handling of #-introduced comments, recognized when
742        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
743        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
744        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
745        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
746        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
747        places in pcre_compile().
748    
749    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
750        comments when looking ahead for named forward references to subpatterns,
751        the only newline sequence it recognized was NL. It now handles newlines
752        according to the set newline convention.
753    
754    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
755        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
756        cater for a lack of strerror(). These oversights have been fixed.
757    
758    9.  Added --match-limit and --recursion-limit to pcregrep.
759    
760    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
761    
762    11. When the -o option was used, pcregrep was setting a return code of 1, even
763        when matches were found, and --line-buffered was not being honoured.
764    
765    12. Added an optional parentheses number to the -o and --only-matching options
766        of pcregrep.
767    
768    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
769        can match an empty string. The code to do it in pcretest and pcredemo
770        needed fixing:
771    
772        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
773            only one byte after an empty string match just before CRLF (this case
774            just got forgotten; "any" and "anycrlf" were OK).
775    
776        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
777            mode when an empty string match preceded an ASCII character followed by
778            a non-ASCII character. (The code for advancing by one character rather
779            than one byte was nonsense.)
780    
781        (c) The pcredemo.c sample program did not have any code at all to handle
782            the cases when CRLF is a valid newline sequence.
783    
784    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
785        as a starting offset was within the subject string. There is now a new
786        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
787        negative or greater than the length of the string. In order to test this,
788        pcretest is extended to allow the setting of negative starting offsets.
789    
790    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
791        starting offset points to the beginning of a UTF-8 character was
792        unnecessarily clumsy. I tidied it up.
793    
794    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
795        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
796    
797    17. Nobody had reported that the --include_dir option, which was added in
798        release 7.7 should have been called --include-dir (hyphen, not underscore)
799        for compatibility with GNU grep. I have changed it to --include-dir, but
800        left --include_dir as an undocumented synonym, and the same for
801        --exclude-dir, though that is not available in GNU grep, at least as of
802        release 2.5.4.
803    
804    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
805        characters from a string of bytes) have been redefined so as not to use
806        loops, in order to improve performance in some environments. At the same
807        time, I abstracted some of the common code into auxiliary macros to save
808        repetition (this should not affect the compiled code).
809    
810    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
811        compile-time error is now given if \c is not followed by an ASCII
812        character, that is, a byte less than 128. (In EBCDIC mode, the code is
813        different, and any byte value is allowed.)
814    
815    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
816        START_OPTIMIZE option, which is now allowed at compile time - but just
817        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
818        to pcregrep and other applications that have no direct access to PCRE
819        options. The new /Y option in pcretest sets this option when calling
820        pcre_compile().
821    
822    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
823        back references. Groups containing recursive back references were forced to
824        be atomic by that change, but in the case of named groups, the amount of
825        memory required was incorrectly computed, leading to "Failed: internal
826        error: code overflow". This has been fixed.
827    
828    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
829        pcretest.c, to avoid build problems in some Borland environments.
830    
831    
832    Version 8.10 25-Jun-2010
833    ------------------------
834    
835    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
836        THEN.
837    
838    2.  (*ACCEPT) was not working when inside an atomic group.
839    
840    3.  Inside a character class, \B is treated as a literal by default, but
841        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
842        causes the error). The code is unchanged, but I tidied the documentation.
843    
844    4.  Inside a character class, PCRE always treated \R and \X as literals,
845        whereas Perl faults them if its -w option is set. I have changed PCRE so
846        that it faults them when PCRE_EXTRA is set.
847    
848    5.  Added support for \N, which always matches any character other than
849        newline. (It is the same as "." when PCRE_DOTALL is not set.)
850    
851    6.  When compiling pcregrep with newer versions of gcc which may have
852        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
853        declared with attribute warn_unused_result" were given. Just casting the
854        result to (void) does not stop the warnings; a more elaborate fudge is
855        needed. I've used a macro to implement this.
856    
857    7.  Minor change to pcretest.c to avoid a compiler warning.
858    
859    8.  Added four artifical Unicode properties to help with an option to make
860        \s etc use properties (see next item). The new properties are: Xan
861        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
862    
863    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
864        use Unicode properties. (*UCP) at the start of a pattern can be used to set
865        this option. Modified pcretest to add /W to test this facility. Added
866        REG_UCP to make it available via the POSIX interface.
867    
868    10. Added --line-buffered to pcregrep.
869    
870    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
871        studied, and the match started with a letter with a code point greater than
872        127 whose first byte was different to the first byte of the other case of
873        the letter, the other case of this starting letter was not recognized
874        (#976).
875    
876    12. If a pattern that was studied started with a repeated Unicode property
877        test, for example, \p{Nd}+, there was the theoretical possibility of
878        setting up an incorrect bitmap of starting bytes, but fortunately it could
879        not have actually happened in practice until change 8 above was made (it
880        added property types that matched character-matching opcodes).
881    
882    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
883        possible starting bytes for non-anchored patterns.
884    
885    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
886        \R, and also a number of cases that involve Unicode properties, both
887        explicit and implicit when PCRE_UCP is set.
888    
889    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
890        input, it could crash or give wrong results if characters with values
891        greater than 0xc0 were present in the subject string. (Detail: it assumed
892        UTF-8 input when processing these items.)
893    
894    16. Added a lot of (int) casts to avoid compiler warnings in systems where
895        size_t is 64-bit (#991).
896    
897    17. Added a check for running out of memory when PCRE is compiled with
898        --disable-stack-for-recursion (#990).
899    
900    18. If the last data line in a file for pcretest does not have a newline on
901        the end, a newline was missing in the output.
902    
903    19. The default pcre_chartables.c file recognizes only ASCII characters (values
904        less than 128) in its various bitmaps. However, there is a facility for
905        generating tables according to the current locale when PCRE is compiled. It
906        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
907        characters, are recognized by isspace() and therefore were getting set in
908        these tables, and indeed these tables seem to approximate to ISO 8859. This
909        caused a problem in UTF-8 mode when pcre_study() was used to create a list
910        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
911        which of course cannot start UTF-8 characters. I have changed the code so
912        that only real ASCII characters (less than 128) and the correct starting
913        bytes for UTF-8 encodings are set for characters greater than 127 when in
914        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
915        altogether.)
916    
917    20. Added the /T option to pcretest so as to be able to run tests with non-
918        standard character tables, thus making it possible to include the tests
919        used for 19 above in the standard set of tests.
920    
921    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
922        reference to a subpattern the other side of a comment that contains an
923        opening parenthesis caused either an internal compiling error, or a
924        reference to the wrong subpattern.
925    
926    
927    Version 8.02 19-Mar-2010
928    ------------------------
929    
930    1.  The Unicode data tables have been updated to Unicode 5.2.0.
931    
932    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
933        configured.
934    
935    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
936        original author of that file, following a query about its status.
937    
938    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
939        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
940    
941    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
942        quantifier applied to a forward-referencing subroutine call, could compile
943        incorrect code or give the error "internal error: previously-checked
944        referenced subpattern not found".
945    
946    6.  Both MS Visual Studio and Symbian OS have problems with initializing
947        variables to point to external functions. For these systems, therefore,
948        pcre_malloc etc. are now initialized to local functions that call the
949        relevant global functions.
950    
951    7.  There were two entries missing in the vectors called coptable and poptable
952        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
953        I've fixed the data, and added a kludgy way of testing at compile time that
954        the lengths are correct (equal to the number of opcodes).
955    
956    8.  Following on from 7, I added a similar kludge to check the length of the
957        eint vector in pcreposix.c.
958    
959    9.  Error texts for pcre_compile() are held as one long string to avoid too
960        much relocation at load time. To find a text, the string is searched,
961        counting zeros. There was no check for running off the end of the string,
962        which could happen if a new error number was added without updating the
963        string.
964    
965    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
966    
967    11. \K was not working if it appeared in an atomic group or in a group that
968        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
969        \K is "not well defined" if used in an assertion. PCRE now accepts it if
970        the assertion is positive, but not if it is negative.
971    
972    12. Change 11 fortuitously reduced the size of the stack frame used in the
973        "match()" function of pcre_exec.c by one pointer. Forthcoming
974        implementation of support for (*MARK) will need an extra pointer on the
975        stack; I have reserved it now, so that the stack frame size does not
976        decrease.
977    
978    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
979        item in branch that calls a recursion is a subroutine call - as in the
980        second branch in the above example - was incorrectly given the compile-
981        time error "recursive call could loop indefinitely" because pcre_compile()
982        was not correctly checking the subroutine for matching a non-empty string.
983    
984    14. The checks for overrunning compiling workspace could trigger after an
985        overrun had occurred. This is a "should never occur" error, but it can be
986        triggered by pathological patterns such as hundreds of nested parentheses.
987        The checks now trigger 100 bytes before the end of the workspace.
988    
989    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
990    
991    
992    Version 8.01 19-Jan-2010
993    ------------------------
994    
995    1.  If a pattern contained a conditional subpattern with only one branch (in
996        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
997        computed the wrong minimum data length (which is of course zero for such
998        subpatterns). This could cause incorrect "no match" results.
999    
1000    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1001        the pattern is reset in the first branch, pcre_compile() failed with
1002        "internal error: code overflow at offset...". This happened only when
1003        the reset was to the original external option setting. (An optimization
1004        abstracts leading options settings into an external setting, which was the
1005        cause of this.)
1006    
1007    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1008        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1009        assertion pattern did not match (meaning that the assertion was true), it
1010        was incorrectly treated as false if the SKIP had been reached during the
1011        matching. This also applied to assertions used as conditions.
1012    
1013    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1014        assertion subpattern, including such a pattern used as a condition,
1015        unpredictable results occurred, instead of the error return
1016        PCRE_ERROR_DFA_UITEM.
1017    
1018    5.  The C++ GlobalReplace function was not working like Perl for the special
1019        situation when an empty string is matched. It now does the fancy magic
1020        stuff that is necessary.
1021    
1022    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1023        removed. (These were left over from very, very early versions of PCRE.)
1024    
1025    7.  Some cosmetic changes to the code to make life easier when compiling it
1026        as part of something else:
1027    
1028        (a) Change DEBUG to PCRE_DEBUG.
1029    
1030        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1031            called "current" as "current_branch", to prevent a collision with the
1032            Linux macro when compiled as a kernel module.
1033    
1034        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1035            prevent a collision with the Linux macro when compiled as a kernel
1036            module.
1037    
1038    8.  In pcre_compile() there are some checks for integer overflows that used to
1039        cast potentially large values to (double). This has been changed to that
1040        when building, a check for int64_t is made, and if it is found, it is used
1041        instead, thus avoiding the use of floating point arithmetic. (There is no
1042        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1043        double.
1044    
1045    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1046        2005 (difference between two addresses compared to an unsigned value).
1047    
1048    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1049        custom one, because of the following reported problem in Windows:
1050    
1051          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1052              under Win32.
1053          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1054              therefore missing the function definition.
1055          - The compiler thus generates a "C" signature for the test function.
1056          - The linker fails to find the "C" function.
1057          - PCRE fails to configure if asked to do so against libbz2.
1058    
1059    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1060        messages were output:
1061    
1062          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1063          rerunning libtoolize, to keep the correct libtool macros in-tree.
1064          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1065    
1066        I have done both of these things.
1067    
1068    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1069        most of the time, it *can* run out if it is given a pattern that contains a
1070        runaway infinite recursion. I updated the discussion in the pcrestack man
1071        page.
1072    
1073    13. Now that we have gone to the x.xx style of version numbers, the minor
1074        version may start with zero. Using 08 or 09 is a bad idea because users
1075        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1076        interpreted as invalid octal numbers. I've updated the previous comment in
1077        configure.ac, and also added a check that gives an error if 08 or 09 are
1078        used.
1079    
1080    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1081        causing partial matching to fail when the end of the subject matched \W
1082        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1083    
1084    15. There were some discrepancies between the declarations in pcre_internal.h
1085        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1086        their definitions. The declarations used "const uschar *" and the
1087        definitions used USPTR. Even though USPTR is normally defined as "const
1088        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1089        reported that: "This difference in casting confuses some C++ compilers, for
1090        example, SunCC recognizes above declarations as different functions and
1091        generates broken code for hbpcre." I have changed the declarations to use
1092        USPTR.
1093    
1094    16. GNU libtool is named differently on some systems. The autogen.sh script now
1095        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1096        (FreeBSD).
1097    
1098    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1099        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1100        comment: "Figure out how to create a longlong from a string: strtoll and
1101        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1102        instance, but it only takes 2 args instead of 3!"
1103    
1104    18. A subtle bug concerned with back references has been fixed by a change of
1105        specification, with a corresponding code fix. A pattern such as
1106        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1107        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1108        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1109        same bug. Such groups have to be quantified to be useful, or contained
1110        inside another quantified group. (If there's no repetition, the reference
1111        can never match.) The problem arises because, having left the group and
1112        moved on to the rest of the pattern, a later failure that backtracks into
1113        the group uses the captured value from the final iteration of the group
1114        rather than the correct earlier one. I have fixed this in PCRE by forcing
1115        any group that contains a reference to itself to be an atomic group; that
1116        is, there cannot be any backtracking into it once it has completed. This is
1117        similar to recursive and subroutine calls.
1118    
1119    
1120    Version 8.00 19-Oct-09
1121    ----------------------
1122    
1123    1.  The table for translating pcre_compile() error codes into POSIX error codes
1124        was out-of-date, and there was no check on the pcre_compile() error code
1125        being within the table. This could lead to an OK return being given in
1126        error.
1127    
1128    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1129        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1130        in a Windows environment.
1131    
1132    3.  The pcregrep --count option prints the count for each file even when it is
1133        zero, as does GNU grep. However, pcregrep was also printing all files when
1134        --files-with-matches was added. Now, when both options are given, it prints
1135        counts only for those files that have at least one match. (GNU grep just
1136        prints the file name in this circumstance, but including the count seems
1137        more useful - otherwise, why use --count?) Also ensured that the
1138        combination -clh just lists non-zero counts, with no names.
1139    
1140    4.  The long form of the pcregrep -F option was incorrectly implemented as
1141        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1142        but it seems right to fix it, and I didn't think it was worth preserving
1143        the old behaviour.
1144    
1145    5.  The command line items --regex=pattern and --regexp=pattern were not
1146        recognized by pcregrep, which required --regex pattern or --regexp pattern
1147        (with a space rather than an '='). The man page documented the '=' forms,
1148        which are compatible with GNU grep; these now work.
1149    
1150    6.  No libpcreposix.pc file was created for pkg-config; there was just
1151        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1152    
1153    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1154        when UCP support is not needed, by modifying the Python script that
1155        generates it from Unicode data files. This should not matter if the module
1156        is correctly used as a library, but I received one complaint about 50K of
1157        unwanted data. My guess is that the person linked everything into his
1158        program rather than using a library. Anyway, it does no harm.
1159    
1160    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1161        was a minimum greater than 1 for a wide character in a possessive
1162        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1163        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1164        character. Chaos in the form of incorrect output or a compiling loop could
1165        result.
1166    
1167    9.  The restrictions on what a pattern can contain when partial matching is
1168        requested for pcre_exec() have been removed. All patterns can now be
1169        partially matched by this function. In addition, if there are at least two
1170        slots in the offset vector, the offset of the earliest inspected character
1171        for the match and the offset of the end of the subject are set in them when
1172        PCRE_ERROR_PARTIAL is returned.
1173    
1174    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1175        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1176        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1177        and may be more useful for multi-segment matching.
1178    
1179    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1180        used to be given if ever the end of the subject was reached; now it is
1181        given only if matching could not proceed because another character was
1182        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1183        string "Z", which now yields "no match" instead of "partial match". In the
1184        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1185        final character ended with (*FAIL).
1186    
1187    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1188        if the pattern had a "must contain" character that was already found in the
1189        earlier partial match, unless partial matching was again requested. For
1190        example, with the pattern /dog.(body)?/, the "must contain" character is
1191        "g". If the first part-match was for the string "dog", restarting with
1192        "sbody" failed. This bug has been fixed.
1193    
1194    13. The string returned by pcre_dfa_exec() after a partial match has been
1195        changed so that it starts at the first inspected character rather than the
1196        first character of the match. This makes a difference only if the pattern
1197        starts with a lookbehind assertion or \b or \B (\K is not supported by
1198        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1199        matching functions compatible, and I think it's the right thing to do.
1200    
1201    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1202        so that the demonstration program is easily available in environments where
1203        PCRE has not been installed from source.
1204    
1205    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1206        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1207        library.
1208    
1209    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1210        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1211        is not the first non-POSIX option to be added. Clearly some people find
1212        these options useful.
1213    
1214    17. If a caller to the POSIX matching function regexec() passes a non-zero
1215        value for nmatch with a NULL value for pmatch, the value of
1216        nmatch is forced to zero.
1217    
1218    18. RunGrepTest did not have a test for the availability of the -u option of
1219        the diff command, as RunTest does. It now checks in the same way as
1220        RunTest, and also checks for the -b option.
1221    
1222    19. If an odd number of negated classes containing just a single character
1223        interposed, within parentheses, between a forward reference to a named
1224        subpattern and the definition of the subpattern, compilation crashed with
1225        an internal error, complaining that it could not find the referenced
1226        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1227        [The bug was that it was starting one character too far in when skipping
1228        over the character class, thus treating the ] as data rather than
1229        terminating the class. This meant it could skip too much.]
1230    
1231    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1232        /g option in pcretest when the pattern contains \K, which makes it possible
1233        to have an empty string match not at the start, even when the pattern is
1234        anchored. Updated pcretest and pcredemo to use this option.
1235    
1236    21. If the maximum number of capturing subpatterns in a recursion was greater
1237        than the maximum at the outer level, the higher number was returned, but
1238        with unset values at the outer level. The correct (outer level) value is
1239        now given.
1240    
1241    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1242        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1243        make it do so. The string so far is captured, making this feature
1244        compatible with Perl.
1245    
1246    23. The tests have been re-organized, adding tests 11 and 12, to make it
1247        possible to check the Perl 5.10 features against Perl 5.10.
1248    
1249    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1250        pattern matches a fixed length string. PCRE did not allow this; now it
1251        does. Neither allows recursion.
1252    
1253    25. I finally figured out how to implement a request to provide the minimum
1254        length of subject string that was needed in order to match a given pattern.
1255        (It was back references and recursion that I had previously got hung up
1256        on.) This code has now been added to pcre_study(); it finds a lower bound
1257        to the length of subject needed. It is not necessarily the greatest lower
1258        bound, but using it to avoid searching strings that are too short does give
1259        some useful speed-ups. The value is available to calling programs via
1260        pcre_fullinfo().
1261    
1262    26. While implementing 25, I discovered to my embarrassment that pcretest had
1263        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1264        study optimizations had never been tested with that matching function.
1265        Oops. What is worse, even when it was passed study data, there was a bug in
1266        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1267        were also very few tests of studied patterns with pcre_dfa_exec().
1268    
1269    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1270        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1271        on the other side of the coin, they are no longer allowed to have different
1272        names, because these cannot be distinguished in PCRE, and this has caused
1273        confusion. (This is a difference from Perl.)
1274    
1275    28. When duplicate subpattern names are present (necessarily with different
1276        numbers, as required by 27 above), and a test is made by name in a
1277        conditional pattern, either for a subpattern having been matched, or for
1278        recursion in such a pattern, all the associated numbered subpatterns are
1279        tested, and the overall condition is true if the condition is true for any
1280        one of them. This is the way Perl works, and is also more like the way
1281        testing by number works.
1282    
1283    
1284    Version 7.9 11-Apr-09
1285    ---------------------
1286    
1287    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1288        (pcretest), all targets were linked against these libraries. This included
1289        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1290        libraries. This caused unwanted dependencies to be created. This problem
1291        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1292        pcretest is linked with readline.
1293    
1294    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1295        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1296        moved outside it again, because FALSE and TRUE are already defined in AIX,
1297        but BOOL is not.
1298    
1299    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1300        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1301    
1302    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1303        hyphens) following file names and line numbers when outputting matching
1304        lines. This is not true; no spaces are inserted. I have also clarified the
1305        wording for the --colour (or --color) option.
1306    
1307    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1308        was not coloured; this is different to GNU grep, so I have changed it to be
1309        the same.
1310    
1311    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1312        each matching line was coloured. Now it goes on to look for further matches
1313        of any of the test patterns, which is the same behaviour as GNU grep.
1314    
1315    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1316        doesn't make sense to accept an empty string match in pcregrep, so I have
1317        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1318        seems to be how GNU grep behaves.
1319    
1320    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1321        start or after a newline", because the conditional assertion was not being
1322        correctly handled. The rule now is that both the assertion and what follows
1323        in the first alternative must satisfy the test.
1324    
1325    9.  If auto-callout was enabled in a pattern with a conditional group whose
1326        condition was an assertion, PCRE could crash during matching, both with
1327        pcre_exec() and pcre_dfa_exec().
1328    
1329    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1330        used for matching.
1331    
1332    11. Unicode property support in character classes was not working for
1333        characters (bytes) greater than 127 when not in UTF-8 mode.
1334    
1335    12. Added the -M command line option to pcretest.
1336    
1337    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1338    
1339    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1340    
1341    16. Added comments and documentation about mis-use of no_arg in the C++
1342        wrapper.
1343    
1344    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1345        from Martin Jerabek that uses macro names for all relevant character and
1346        string constants.
1347    
1348    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1349        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1350        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1351        these, but not everybody uses configure.
1352    
1353    19. A conditional group that had only one branch was not being correctly
1354        recognized as an item that could match an empty string. This meant that an
1355        enclosing group might also not be so recognized, causing infinite looping
1356        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1357        with the subject "ab", where knowledge that the repeated group can match
1358        nothing is needed in order to break the loop.
1359    
1360    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1361        exec(), but without supplying a callout function, matching went wrong.
1362    
1363    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1364        leak if the size of the offset vector was greater than 30. When the vector
1365        is smaller, the saved offsets during recursion go onto a local stack
1366        vector, but for larger vectors malloc() is used. It was failing to free
1367        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1368        error, in fact).
1369    
1370    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1371        heapframe that is used only when UTF-8 support is enabled. This caused no
1372        problem, but was untidy.
1373    
1374    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1375        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1376        included within another project.
1377    
1378    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1379        slightly modified by me:
1380    
1381          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1382              not building pcregrep.
1383    
1384          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1385              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1386    
1387    25. Forward references, both numeric and by name, in patterns that made use of
1388        duplicate group numbers, could behave incorrectly or give incorrect errors,
1389        because when scanning forward to find the reference group, PCRE was not
1390        taking into account the duplicate group numbers. A pattern such as
1391        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1392    
1393    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1394        the feature of a custom pointer more persuasive (as requested by a user).
1395    
1396    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1397        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1398        pre-defined.
1399    
1400    28. Added support for (*UTF8) at the start of a pattern.
1401    
1402    29. Arrange for flags added by the "release type" setting in CMake to be shown
1403        in the configuration summary.
1404    
1405    
1406    Version 7.8 05-Sep-08
1407    ---------------------
1408    
1409    1.  Replaced UCP searching code with optimized version as implemented for Ad
1410        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1411        stage table and inline lookup instead of a function, giving speed ups of 2
1412        to 5 times on some simple patterns that I tested. Permission was given to
1413        distribute the MultiStage2.py script that generates the tables (it's not in
1414        the tarball, but is in the Subversion repository).
1415    
1416    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1417        scripts.
1418    
1419    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1420        a group with a zero qualifier. The result of the study could be incorrect,
1421        or the function might crash, depending on the pattern.
1422    
1423    4.  Caseless matching was not working for non-ASCII characters in back
1424        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1425        It now works when Unicode Property Support is available.
1426    
1427    5.  In pcretest, an escape such as \x{de} in the data was always generating
1428        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1429        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1430        truncation.
1431    
1432    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1433    
1434    7.  Added two (int) casts to pcregrep when printing the difference of two
1435        pointers, in case they are 64-bit values.
1436    
1437    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1438        test 2 if it fails.
1439    
1440    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1441        and a #define of that name to empty if it is not externally set. This is to
1442        allow users of MSVC to set it if necessary.
1443    
1444    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1445        the convenience functions in the pcre_get.c source file.
1446    
1447    11. An option change at the start of a pattern that had top-level alternatives
1448        could cause overwriting and/or a crash. This command provoked a crash in
1449        some environments:
1450    
1451          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1452    
1453        This potential security problem was recorded as CVE-2008-2371.
1454    
1455    12. For a pattern where the match had to start at the beginning or immediately
1456        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1457        pcre_dfa_exec() could read past the end of the passed subject if there was
1458        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1459        pcretest so that it places the subject at the end of its malloc-ed buffer.
1460    
1461    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1462        exec() might read past the end of the data buffer in UTF-8 mode.
1463    
1464    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1465        the data contained the byte 0x85 as part of a UTF-8 character within its
1466        first line. This applied both to normal and DFA matching.
1467    
1468    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1469        /^[^d]*?$/8 failed to match "abc".
1470    
1471    16. Added a missing copyright notice to pcrecpp_internal.h.
1472    
1473    17. Make it more clear in the documentation that values returned from
1474        pcre_exec() in ovector are byte offsets, not character counts.
1475    
1476    18. Tidied a few places to stop certain compilers from issuing warnings.
1477    
1478    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1479        supplied by Stefan Weber. I made a further small update for 7.8 because
1480        there is a change of source arrangements: the pcre_searchfuncs.c module is
1481        replaced by pcre_ucd.c.
1482    
1483    
1484    Version 7.7 07-May-08
1485    ---------------------
1486    
1487    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1488        a string to a long long, pretend we don't even have a long long." This is
1489        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1490    
1491    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1492        pre-7.6 versions, which defined a global no_arg variable instead of putting
1493        it in the RE class. (See also #8 below.)
1494    
1495    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1496        Lopes.
1497    
1498    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1499    
1500        (1) The include/exclude patterns were being applied to the whole pathnames
1501            of files, instead of just to the final components.
1502    
1503        (2) If there was more than one level of directory, the subdirectories were
1504            skipped unless they satisfied the include/exclude conditions. This is
1505            inconsistent with GNU grep (and could even be seen as contrary to the
1506            pcregrep specification - which I improved to make it absolutely clear).
1507            The action now is always to scan all levels of directory, and just
1508            apply the include/exclude patterns to regular files.
1509    
1510    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1511        --exclude_dir in the tests to avoid scanning .svn directories.
1512    
1513    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1514        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1515        doesn't support NULs in patterns.
1516    
1517    7.  Added some missing "const"s to declarations of static tables in
1518        pcre_compile.c and pcre_dfa_exec.c.
1519    
1520    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1521        caused by fix #2  above. (Subsequently also a second patch to fix the
1522        first patch. And a third patch - this was a messy problem.)
1523    
1524    9.  Applied Craig's patch to remove the use of push_back().
1525    
1526    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1527        matching function regexec().
1528    
1529    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1530        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1531        references. PCRE supports relative numbers with this syntax (I don't think
1532        Oniguruma does).
1533    
1534    12. Previously, a group with a zero repeat such as (...){0} was completely
1535        omitted from the compiled regex. However, this means that if the group
1536        was called as a subroutine from elsewhere in the pattern, things went wrong
1537        (an internal error was given). Such groups are now left in the compiled
1538        pattern, with a new opcode that causes them to be skipped at execution
1539        time.
1540    
1541    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1542        to the way PCRE behaves:
1543    
1544        (a) A lone ] character is dis-allowed (Perl treats it as data).
1545    
1546        (b) A back reference to an unmatched subpattern matches an empty string
1547            (Perl fails the current match path).
1548    
1549        (c) A data ] in a character class must be notated as \] because if the
1550            first data character in a class is ], it defines an empty class. (In
1551            Perl it is not possible to have an empty class.) The empty class []
1552            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1553            The negative empty class [^] matches any one character, independently
1554            of the DOTALL setting.
1555    
1556    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1557        non-existent subpattern following a character class starting with ']' and
1558        containing () gave an internal compiling error instead of "reference to
1559        non-existent subpattern". Fortunately, when the pattern did exist, the
1560        compiled code was correct. (When scanning forwards to check for the
1561        existencd of the subpattern, it was treating the data ']' as terminating
1562        the class, so got the count wrong. When actually compiling, the reference
1563        was subsequently set up correctly.)
1564    
1565    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1566        it was being rejected as not supported by pcre_dfa_exec(), even though
1567        other assertions are supported. I have made pcre_dfa_exec() support
1568        (*FAIL).
1569    
1570    16. The implementation of 13c above involved the invention of a new opcode,
1571        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1572        cannot be changed at match time, I realized I could make a small
1573        improvement to matching performance by compiling OP_ALLANY instead of
1574        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1575        on the OP_ANY path.
1576    
1577    17. Compiling pcretest on Windows with readline support failed without the
1578        following two fixes: (1) Make the unistd.h include conditional on
1579        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1580    
1581    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1582        ncurses library to be included for pcretest when ReadLine support is
1583        requested, but also to allow for it to be overridden. This patch came from
1584        Daniel Bergström.
1585    
1586    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1587        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1588        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1589        spotting this.
1590    
1591    
1592    Version 7.6 28-Jan-08
1593    ---------------------
1594    
1595    1.  A character class containing a very large number of characters with
1596        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1597        overflow.
1598    
1599    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1600        HAVE_LONG_LONG is not defined.
1601    
1602    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1603        bring them up to date and include new features. This patch includes:
1604    
1605        - Fixed PH's badly added libz and libbz2 support.
1606        - Fixed a problem with static linking.
1607        - Added pcredemo. [But later removed - see 7 below.]
1608        - Fixed dftables problem and added an option.
1609        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1610            HAVE_LONG_LONG.
1611        - Added readline support for pcretest.
1612        - Added an listing of the option settings after cmake has run.
1613    
1614    4.  A user submitted a patch to Makefile that makes it easy to create
1615        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1616        Makefile.am that cause it to include this special target, without
1617        affecting anything else. Note that the same mingw target plus all
1618        the other distribution libraries and programs are now supported
1619        when configuring with CMake (see 6 below) instead of with
1620        Configure/Make.
1621    
1622    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1623        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1624        exported in the Windows port". It has not yet been confirmed that the patch
1625        solves the problem, but it does no harm.
1626    
1627    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1628        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1629        with CMake, and also correct the comment about stack recursion.
1630    
1631    7.  Remove the automatic building of pcredemo from the ./configure system and
1632        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1633        of a program that users should build themselves after PCRE is installed, so
1634        building it automatically is not really right. What is more, it gave
1635        trouble in some build environments.
1636    
1637    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1638    
1639    
1640    Version 7.5 10-Jan-08
1641    ---------------------
1642    
1643    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1644        values in parens when parsing an RE using the C++ wrapper."
1645    
1646    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1647        Characters greater than 255 were excluded from the class instead of being
1648        included.
1649    
1650    3.  The same bug as (2) above applied to negated POSIX classes such as
1651        [:^space:].
1652    
1653    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1654        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1655        I have changed it.
1656    
1657    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1658        first named subpattern) and a construct such as (?&a) would reference the
1659        first named subpattern whose name started with "a" (in other words, the
1660        length check was missing). Both these problems are fixed. "Subpattern name
1661        expected" is now given for (?&) (a zero-length name), and this patch also
1662        makes it give the same error for \k'' (previously it complained that that
1663        was a reference to a non-existent subpattern).
1664    
1665    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1666        this is right because (?- can be followed by option settings as well as by
1667        digits. I have, however, made the messages clearer.
1668    
1669    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1670        than the number used in the conditional) now cause a compile-time error.
1671        This is actually not compatible with Perl, which accepts such patterns, but
1672        treats the conditional as always being FALSE (as PCRE used to), but it
1673        seems to me that giving a diagnostic is better.
1674    
1675    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1676        and messages.
1677    
1678    9.  Fix two occurrences of "backslash" in comments that should have been
1679        "backspace".
1680    
1681    10. Remove two redundant lines of code that can never be obeyed (their function
1682        was moved elsewhere).
1683    
1684    11. The program that makes PCRE's Unicode character property table had a bug
1685        which caused it to generate incorrect table entries for sequences of
1686        characters that have the same character type, but are in different scripts.
1687        It amalgamated them into a single range, with the script of the first of
1688        them. In other words, some characters were in the wrong script. There were
1689        thirteen such cases, affecting characters in the following ranges:
1690    
1691          U+002b0 - U+002c1
1692          U+0060c - U+0060d
1693          U+0061e - U+00612
1694          U+0064b - U+0065e
1695          U+0074d - U+0076d
1696          U+01800 - U+01805
1697          U+01d00 - U+01d77
1698          U+01d9b - U+01dbf
1699          U+0200b - U+0200f
1700          U+030fc - U+030fe
1701          U+03260 - U+0327f
1702          U+0fb46 - U+0fbb1
1703          U+10450 - U+1049d
1704    
1705    12. The -o option (show only the matching part of a line) for pcregrep was not
1706        compatible with GNU grep in that, if there was more than one match in a
1707        line, it showed only the first of them. It now behaves in the same way as
1708        GNU grep.
1709    
1710    13. If the -o and -v options were combined for pcregrep, it printed a blank
1711        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1712        does the same. The return code can be used to tell if there were any
1713        non-matching lines.
1714    
1715    14. Added --file-offsets and --line-offsets to pcregrep.
1716    
1717    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1718        infinitely looping recursion. The bug was that positive lookaheads were not
1719        being skipped when checking for a possible empty match (negative lookaheads
1720        and both kinds of lookbehind were skipped).
1721    
1722    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1723        inclusion of <windows.h> to before rather than after the definition of
1724        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1725    
1726    17. Specifying a possessive quantifier with a specific limit for a Unicode
1727        character property caused pcre_compile() to compile bad code, which led at
1728        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1729        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1730        caused the error; without that there was no problem.
1731    
1732    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1733    
1734    19. Added --enable-pcretest-libreadline.
1735    
1736    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1737        RE::GlobalReplace(). As a result, the number of replacements returned was
1738        double what it should be. I removed one of the increments, but Craig sent a
1739        later patch that removed the other one (the right fix) and added unit tests
1740        that check the return values (which was not done before).
1741    
1742    21. Several CMake things:
1743    
1744        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1745            the names libpcre and libpcreposix, not just pcre and pcreposix.
1746    
1747        (2) The above change means that pcretest and pcregrep are now correctly
1748            linked with the newly-built libraries, not previously installed ones.
1749    
1750        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1751    
1752    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1753        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1754        UTF-8 newline character). The key issue is that the pattern starts .*;
1755        this means that the match must be either at the beginning, or after a
1756        newline. The bug was in the code for advancing after a failed match and
1757        checking that the new position followed a newline. It was not taking
1758        account of UTF-8 characters correctly.
1759    
1760    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1761        character classes. PCRE was not treating the sequence [:...:] as a
1762        character class unless the ... were all letters. Perl, however, seems to
1763        allow any characters between [: and :], though of course it rejects as
1764        unknown any "names" that contain non-letters, because all the known class
1765        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1766        for example, whereas PCRE did not - it did not recognize a POSIX character
1767        class. This seemed a bit dangerous, so the code has been changed to be
1768        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1769        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1770        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1771        Perl does, and where it didn't before.
1772    
1773    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1774        Windows environments %n is disabled by default.
1775    
1776    
1777    Version 7.4 21-Sep-07
1778    ---------------------
1779    
1780    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1781        means that a class such as [\s] counted as "explicit reference to CR or
1782        LF". That isn't really right - the whole point of the change was to try to
1783        help when there was an actual mention of one of the two characters. So now
1784        the change happens only if \r or \n (or a literal CR or LF) character is
1785        encountered.
1786    
1787    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1788        of both had grown to the point where there were only 3 bits left.
1789        Fortunately, there was spare space in the data structure, and so I have
1790        moved the internal flags into a new 16-bit field to free up more option
1791        bits.
1792    
1793    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1794        but did not set the internal JCHANGED flag - either of these is enough to
1795        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1796        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1797        start sets both bits.
1798    
1799    4.  Added options (at build time, compile time, exec time) to change \R from
1800        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1801    
1802    5.  doc/pcresyntax.html was missing from the distribution.
1803    
1804    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1805        compatibility, even though it is no longer used.
1806    
1807    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1808        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1809        windows.h file is present (where different names are used). [This was
1810        reversed later after testing - see 16 below.]
1811    
1812    8.  Changed all #include <config.h> to #include "config.h". There were also
1813        some further <pcre.h> cases that I changed to "pcre.h".
1814    
1815    9.  When pcregrep was used with the --colour option, it missed the line ending
1816        sequence off the lines that it output.
1817    
1818    10. It was pointed out to me that arrays of string pointers cause lots of
1819        relocations when a shared library is dynamically loaded. A technique of
1820        using a single long string with a table of offsets can drastically reduce
1821        these. I have refactored PCRE in four places to do this. The result is
1822        dramatic:
1823    
1824          Originally:                          290
1825          After changing UCP table:            187
1826          After changing error message table:   43
1827          After changing table of "verbs"       36
1828          After changing table of Posix names   22
1829    
1830        Thanks to the folks working on Gregex for glib for this insight.
1831    
1832    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1833        unicode-properties was also set.
1834    
1835    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1836    
1837    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1838        checked only for CRLF.
1839    
1840    14. Added casts to pcretest.c to avoid compiler warnings.
1841    
1842    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1843    
1844    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1845        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1846        entirely. This removes changes made in 7 above.
1847    
1848    17. The CMake files have been updated, and there is now more information about
1849        building with CMake in the NON-UNIX-USE document.
1850    
1851    
1852    Version 7.3 28-Aug-07
1853    ---------------------
1854    
1855     1. In the rejigging of the build system that eventually resulted in 7.1, the
1856        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1857        brackets there is not right, since it causes compilers to look for an
1858        installed pcre.h, not the version that is in the source that is being
1859        compiled (which of course may be different). I have changed it back to:
1860    
1861          #include "pcre.h"
1862    
1863        I have a vague recollection that the change was concerned with compiling in
1864        different directories, but in the new build system, that is taken care of
1865        by the VPATH setting the Makefile.
1866    
1867     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1868        when the subject happened to end in the byte 0x85 (e.g. if the last
1869        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1870        characters but of course it shouldn't be taken as a newline when it is part
1871        of another character. The bug was that, for an unlimited repeat of . in
1872        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1873        characters when looking for a newline.
1874    
1875     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1876    
1877     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1878        in debug output.
1879    
1880     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1881        long printing in the pcrecpp unittest when running under MinGW.
1882    
1883     6. ESC_K was left out of the EBCDIC table.
1884    
1885     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1886        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1887        limit also applies to "virtual nesting" when a pattern is recursive, and in
1888        this case 1000 isn't so big. I have been able to remove this limit at the
1889        expense of backing off one optimization in certain circumstances. Normally,
1890        when pcre_exec() would call its internal match() function recursively and
1891        immediately return the result unconditionally, it uses a "tail recursion"
1892        feature to save stack. However, when a subpattern that can match an empty
1893        string has an unlimited repetition quantifier, it no longer makes this
1894        optimization. That gives it a stack frame in which to save the data for
1895        checking that an empty string has been matched. Previously this was taken
1896        from the 1000-entry workspace that had been reserved. So now there is no
1897        explicit limit, but more stack is used.
1898    
1899     8. Applied Daniel's patches to solve problems with the import/export magic
1900        syntax that is required for Windows, and which was going wrong for the
1901        pcreposix and pcrecpp parts of the library. These were overlooked when this
1902        problem was solved for the main library.
1903    
1904     9. There were some crude static tests to avoid integer overflow when computing
1905        the size of patterns that contain repeated groups with explicit upper
1906        limits. As the maximum quantifier is 65535, the maximum group length was
1907        set at 30,000 so that the product of these two numbers did not overflow a
1908        32-bit integer. However, it turns out that people want to use groups that
1909        are longer than 30,000 bytes (though not repeat them that many times).
1910        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1911        made it possible to implement the integer overflow checks in a much more
1912        dynamic way, which I have now done. The artificial limitation on group
1913        length has been removed - we now have only the limit on the total length of
1914        the compiled pattern, which depends on the LINK_SIZE setting.
1915    
1916    10. Fixed a bug in the documentation for get/copy named substring when
1917        duplicate names are permitted. If none of the named substrings are set, the
1918        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1919        empty string.
1920    
1921    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1922        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1923        because the ] is interpreted as the first data character and the
1924        terminating ] is not found. PCRE has been made compatible with Perl in this
1925        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1926        cause memory overwriting.
1927    
1928    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1929        string has been matched (to stop an infinite loop). It was not recognizing
1930        a conditional subpattern that could match an empty string if that
1931        subpattern was within another subpattern. For example, it looped when
1932        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1933        condition was not nested. This bug has been fixed.
1934    
1935    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1936        past the start of the subject in the presence of bytes with the top bit
1937        set, for example "\x8aBCD".
1938    
1939    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1940        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1941    
1942    14. Optimized (?!) to (*FAIL).
1943    
1944    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1945        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1946        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1947        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1948        does: it's just the validity check that is more restrictive.
1949    
1950    16. Inserted checks for integer overflows during escape sequence (backslash)
1951        processing, and also fixed erroneous offset values for syntax errors during
1952        backslash processing.
1953    
1954    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1955        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1956    
1957    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1958        caused an overrun.
1959    
1960    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1961        something other than just ASCII characters) inside a group that had an
1962        unlimited repeat caused a loop at compile time (while checking to see
1963        whether the group could match an empty string).
1964    
1965    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1966        [\P{Any}] did so. (Error in the code for printing property names.)
1967    
1968    21. An orphan \E inside a character class could cause a crash.
1969    
1970    22. A repeated capturing bracket such as (A)? could cause a wild memory
1971        reference during compilation.
1972    
1973    23. There are several functions in pcre_compile() that scan along a compiled
1974        expression for various reasons (e.g. to see if it's fixed length for look
1975        behind). There were bugs in these functions when a repeated \p or \P was
1976        present in the pattern. These operators have additional parameters compared
1977        with \d, etc, and these were not being taken into account when moving along
1978        the compiled data. Specifically:
1979    
1980        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1981            length.
1982    
1983        (b) An item such as \pL+ within a repeated group could cause crashes or
1984            loops.
1985    
1986        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1987            "reference to non-existent subpattern" error.
1988    
1989        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1990    
1991    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1992        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1993    
1994    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1995    
1996    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1997        character were causing crashes (broken optimization).
1998    
1999    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2000        \p or \P) caused a compile-time loop.
2001    
2002    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2003        break. For example, the unstudied pattern [\r\n]A does not match the string
2004        "\r\nA" because change 7.0/46 below moves the current point on by two
2005        characters after failing to match at the start. However, the pattern \nA
2006        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2007        the same is true. There doesn't seem any very clean way out of this, but
2008        what I have chosen to do makes the common cases work: PCRE now takes note
2009        of whether there can be an explicit match for \r or \n anywhere in the
2010        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2011        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2012        pattern has explicit CR or LF references.
2013    
2014    29. Added (*CR) etc for changing newline setting at start of pattern.
2015    
2016    
2017    Version 7.2 19-Jun-07
2018  ---------------------  ---------------------
2019    
2020   1. If the fr_FR locale cannot be found for test 3, try the "french" locale,   1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
# Line 41  Version 7.2 05-June-07 Line 2054  Version 7.2 05-June-07
2054      (f) \g{name} is another synonym - part of Perl 5.10's unification of      (f) \g{name} is another synonym - part of Perl 5.10's unification of
2055          reference syntax.          reference syntax.
2056    
2057        (g) (?| introduces a group in which the numbering of parentheses in each
2058            alternative starts with the same number.
2059    
2060        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2061    
2062   7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and   7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2063      PCRE_INFO_JCHANGED.      PCRE_INFO_JCHANGED.
2064    
# Line 55  Version 7.2 05-June-07 Line 2073  Version 7.2 05-June-07
2073      alternatives. The 1000-alternative test pattern now uses 12 bytes of      alternatives. The 1000-alternative test pattern now uses 12 bytes of
2074      workspace instead of running out of the 4096 that are available.      workspace instead of running out of the 4096 that are available.
2075    
2076    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2077    
2078    11. Applied patch from Google to remove an optimization that didn't quite work.
2079        The report of the bug said:
2080    
2081          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2082          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2083          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2084    
2085    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2086        it matched the wrong number of bytes.
2087    
2088    
2089  Version 7.1 24-Apr-07  Version 7.1 24-Apr-07
2090  ---------------------  ---------------------

Legend:
Removed from v.172  
changed lines
  Added in v.971

  ViewVC Help
Powered by ViewVC 1.1.5