ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 167 by ph10, Wed May 9 15:53:54 2007 UTC revision 1149 by zherczeg, Sat Oct 20 21:33:38 2012 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4  Version 7.2 01-May-07  Version 8.32
5    ------------
7    1.  Improved JIT compiler optimizations for first character search and single
8        character iterators.
10    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
11        Patch by Daniel Richard G.
13    3.  Single character iterator optimizations in the JIT compiler.
15    4.  Improved JIT compiler optimizations for character ranges.
17    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
18        Reported by Giuseppe D'Angelo.
20    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
21        of a line, was being set incorrectly in cases where .* appeared inside
22        atomic brackets at the start of a pattern, or where there was a subsequent
23        *PRUNE or *SKIP.
25    7.  Improved instruction cache flush for POWER/PowerPC.
26        Patch by Daniel Richard G.
28    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
29        grep:
31        (a) There is now no limit to the number of patterns to be matched.
33        (b) An error is given if a pattern is too long.
35        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
36            are now supported.
38        (d) --exclude-from and --include-from (multiple use) have been added.
40        (e) Exclusions and inclusions now apply to all files and directories, not
41            just to those obtained from scanning a directory recursively.
43        (f) Multiple uses of -f and --file-list are now supported.
45        (g) In a Windows environment, the default for -d has been changed from
46            "read" (the GNU grep default) to "skip", because otherwise the presence
47            of a directory in the file list provokes an error.
49        (h) The documentation has been revised and clarified in places.
51    9.  Improve the matching speed of capturing brackets.
53    10. Changed the meaning of \X so that it now matches a Unicode extended
54        grapheme cluster.
56    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
57        out POSIX threads when JIT support is configured.
59    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
61    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
62        structure could go wrong in environments where size_t is not the same size
63        as int.
65    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
66        set.
68    15. The EBCDIC support had decayed; later updates to the code had included
69        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
70        general tidy up of EBCDIC-related issues, and the documentation was also
71        not quite right. There is now a test that can be run on ASCII systems to
72        check some of the EBCDIC-related things (but is it not a full test).
74    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
75        in a small tidy to the code.
77    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
79    18. If the --only-matching (-o) option in pcregrep is specified multiple
80        times, each one causes appropriate output. For example, -o1 -o2 outputs the
81        substrings matched by the 1st and 2nd capturing parentheses. A separating
82        string can be specified by --om-separator (default empty).
84    19. Improving the first n character searches.
86    20. Turn case lists for horizontal and vertical white space into macros so that
87        they are defined only once.
89    21. This set of changes together give more compatible Unicode case-folding
90        behaviour for characters that have more than one other case when UCP
91        support is available.
93        (a) The Unicode property table now has offsets into a new table of sets of
94            three or more characters that are case-equivalent. The MultiStage2.py
95            script that generates these tables (the pcre_ucd.c file) now scans
96            CaseFolding.txt instead of UnicodeData.txt for character case
97            information.
99        (b) The code for adding characters or ranges of characters to a character
100            class has been abstracted into a generalized function that also handles
101            case-independence. In UTF-mode with UCP support, this uses the new data
102            to handle characters with more than one other case.
104        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
105            whose other case is greater than 256 are now correctly matched
106            caselessly. Previously, the high codepoint matched the low one, but not
107            vice versa.
109        (d) The processing of \h, \H, \v, and \ in character classes now makes use
110            of the new class addition function, using character lists defined as
111            macros alongside the case definitions of 20 above.
113        (e) Caseless back references now work with characters that have more than
114            one other case.
116        (f) General caseless matching of characters with more than one other case
117            is supported.
119    22. Unicode character properties were updated from Unicode 6.2.0
121    23. Improved CMake support under Windows. Patch by Daniel Richard G.
123    24. Add support for 32-bit character strings, and UTF-32
125    25. Major JIT compiler update (code refactoring and bugfixing).
126        Experimental Sparc 32 support is added.
129    Version 8.31 06-July-2012
130    -------------------------
132    1.  Fixing a wrong JIT test case and some compiler warnings.
134    2.  Removed a bashism from the RunTest script.
136    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
137        to unsigned type, result still unsigned" that was given by an MS compiler
138        on encountering the code "-sizeof(xxx)".
140    4.  Partial matching support is added to the JIT compiler.
142    5.  Fixed several bugs concerned with partial matching of items that consist
143        of more than one character:
145        (a) /^(..)\1/ did not partially match "aba" because checking references was
146            done on an "all or nothing" basis. This also applied to repeated
147            references.
149        (b) \R did not give a hard partial match if \r was found at the end of the
150            subject.
152        (c) \X did not give a hard partial match after matching one or more
153            characters at the end of the subject.
155        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
156            a partial match for the string "\r".
158        (e) When newline was set to CRLF, the metacharacter "." did not recognize
159            a partial match for a CR character at the end of the subject string.
161    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
162        running pcretest, the text "(JIT)" added to the output whenever JIT is
163        actually used to run the match.
165    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
166        or /S+[+] with a digit between 1 and 7.
168    8.  OP_NOT now supports any UTF character not just single-byte ones.
170    9.  (*MARK) control verb is now supported by the JIT compiler.
172    10. The command "./RunTest list" lists the available tests without actually
173        running any of them. (Because I keep forgetting what they all are.)
177    12. Applied a (slightly modified) user-supplied patch that improves performance
178        when the heap is used for recursion (compiled with --disable-stack-for-
179        recursion). Instead of malloc and free for each heap frame each time a
180        logical recursion happens, frames are retained on a chain and re-used where
181        possible. This sometimes gives as much as 30% improvement.
183    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
184        call.
186    14. As documented, (*COMMIT) is now confined to within a positive assertion.
188    15. It is now possible to link pcretest with libedit as an alternative to
189        libreadline.
191    16. (*COMMIT) control verb is now supported by the JIT compiler.
193    17. The Unicode data tables have been updated to Unicode 6.1.0.
195    18. Added --file-list option to pcregrep.
197    19. Added binary file support to pcregrep, including the -a, --binary-files,
198        -I, and --text options.
200    20. The madvise function is renamed for posix_madvise for QNX compatibility
201        reasons. Fixed by Giuseppe D'Angelo.
203    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
205    22. Changed the option for creating symbolic links for 16-bit man pages from
206        -s to -sf so that re-installing does not cause issues.
208    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
210    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
211        matches in certain environments (the workspace was not being correctly
212        retained). Also added to pcre_dfa_exec() a simple plausibility check on
213        some of the workspace data at the beginning of a restart.
215    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
216        was not doing so when it should - probably a typo introduced by SVN 528
217        (change 8.10/14).
219    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
220        \w+ when the character tables indicated that \x{c4} was a word character.
221        There were several related cases, all because the tests for doing a table
222        lookup were testing for characters less than 127 instead of 255.
224    27. If a pattern contains capturing parentheses that are not used in a match,
225        their slots in the ovector are set to -1. For those that are higher than
226        any matched groups, this happens at the end of processing. In the case when
227        there were back references that the ovector was too small to contain
228        (causing temporary malloc'd memory to be used during matching), and the
229        highest capturing number was not used, memory off the end of the ovector
230        was incorrectly being set to -1. (It was using the size of the temporary
231        memory instead of the true size.)
233    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
234        ovector size, it uses memory at the end of the block that it has got.
236    29. Check for an overlong MARK name and give an error at compile time. The
237        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
239    30. JIT compiler update.
241    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
242        Rill for the patch.
244    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
246    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
248    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
249        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
250        was enabled.
252    35. Improve JIT code generation for greedy plus quantifier.
254    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
255        1 to "aa" instead of to an empty string. The bug affected repeated groups
256        that could potentially match an empty string.
258    37. Optimizing single character iterators in JIT.
260    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
261        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
262        codepoints that are too big for the mode are faulted, and in a UTF mode,
263        disallowed codepoints are also faulted.
265    39. If PCRE was compiled with UTF support, in three places in the DFA
266        matcher there was code that should only have been obeyed in UTF mode, but
267        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
268        processing when bytes with values greater than 127 were present. In 16-bit
269        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
270        both cases the values are those that cannot be the first data item in a UTF
271        character. The three items that might have provoked this were recursions,
272        possessively repeated groups, and atomic groups.
274    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
275        and pcregrep, because some OS require shared objects to be explicitly
276        passed to ld, causing the link step to fail if they are not.
278    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
279        mode, patterns that started with \h* or \R* might be incorrectly matched.
282    Version 8.30 04-February-2012
283    -----------------------------
285    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
286        name is defined in ctype.h.
288    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
289        only in quite long subpatterns.
291    3.  Removed the function pcre_info(), which has been obsolete and deprecated
292        since it was replaced by pcre_fullinfo() in February 2000.
294    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
295        match a (*MARK), and the match failed at the start of the subject, a
296        reference to memory before the start of the subject could occur. This bug
297        was introduced by fix 17 of release 8.21.
299    5.  A reference to an unset group with zero minimum repetition was giving
300        totally wrong answers (in non-JavaScript-compatibility mode). For example,
301        /(another)?(\1?)test/ matched against "hello world test". This bug was
302        introduced in release 8.13.
304    6.  Add support for 16-bit character strings (a large amount of work involving
305        many changes and refactorings).
307    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
308        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
309        from a file.
311    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
312        rounding is not applied in this particular case).
314    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
315        if they appear, or are escaped, in patterns.
317    10. Get rid of a number of -Wunused-but-set-variable warnings.
319    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
320        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
321        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
322        also returns the mark "x". This bug applied to capturing parentheses,
323        non-capturing parentheses, and atomic parentheses. It also applied to some
324        assertions.
326    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
327        information out of configure.ac instead of relying on pcre.h.generic, which
328        is not stored in the repository.
330    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
331        -lreadline.
333    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
335    15. Applied Graycode's patch to put the top-level frame on the stack rather
336        than the heap when not using the stack for recursion. This gives a
337        performance improvement in many cases when recursion is not deep.
339    16. Experimental code added to "pcretest -C" to output the stack frame size.
342    Version 8.21 12-Dec-2011
343    ------------------------
345    1.  Updating the JIT compiler.
347    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
348        are added as well.
350    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
351        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
352        calling _pcre_jit_exec. Some extra comments are added.
354    4.  (*MARK) settings inside atomic groups that do not contain any capturing
355        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
356        was introduced by change 18 for 8.20.
358    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
359        ECMA-262 standard.
361    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
362        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
363        This bug was probably introduced by change 9 of 8.13.
365    7.  While fixing 6 above, I noticed that a number of other items were being
366        incorrectly rejected as "not fixed length". This arose partly because newer
367        opcodes had not been added to the fixed-length checking code. I have (a)
368        corrected the bug and added tests for these items, and (b) arranged for an
369        error to occur if an unknown opcode is encountered while checking for fixed
370        length instead of just assuming "not fixed length". The items that were
371        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
372        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
373        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
375    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
376        being incorrectly compiled and would have given unpredicatble results.
378    9.  A possessively repeated subpattern with minimum repeat count greater than
379        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
380        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
381        the first (A) could occur when it should not.
383    10. Add a cast and remove a redundant test from the code.
385    11. JIT should use pcre_malloc/pcre_free for allocation.
387    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
388        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
389        is anything other than /usr, -L is still shown).
391    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
393    14. Perl does not support \N without a following name in a [] class; PCRE now
394        also gives an error.
396    15. If a forward reference was repeated with an upper limit of around 2000,
397        it caused the error "internal error: overran compiling workspace". The
398        maximum number of forward references (including repeats) was limited by the
399        internal workspace, and dependent on the LINK_SIZE. The code has been
400        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
401        the default depends on LINK_SIZE. There is a new upper limit (for safety)
402        of around 200,000 forward references. While doing this, I also speeded up
403        the filling in of repeated forward references.
405    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
406        incorrectly expecting the subject to contain another "a" after the start.
408    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
409        in the match, the SKIP should be ignored. This was not happening; instead
410        the SKIP was being treated as NOMATCH. For patterns such as
411        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
412        tested.
414    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
415        now much more compatible with Perl, in particular in cases where the result
416        is a non-match for a non-anchored pattern. For example, if
417        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
418        "m", where previously it did not return a name. A side effect of this
419        change is that for partial matches, the last encountered mark name is
420        returned, as for non matches. A number of tests that were previously not
421        Perl-compatible have been moved into the Perl-compatible test files. The
422        refactoring has had the pleasing side effect of removing one argument from
423        the match() function, thus reducing its stack requirements.
425    19. If the /S+ option was used in pcretest to study a pattern using JIT,
426        subsequent uses of /S (without +) incorrectly behaved like /S+.
428    21. Retrieve executable code size support for the JIT compiler and fixing
429        some warnings.
431    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
432        not work when the shorter character appeared right at the end of the
433        subject string.
435    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
436        systems.
438    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
439        output it when the /M option is used in pcretest.
441    25. The CheckMan script was not being included in the distribution. Also, added
442        an explicit "perl" to run Perl scripts from the PrepareRelease script
443        because this is reportedly needed in Windows.
445    26. If study data was being save in a file and studying had not found a set of
446        "starts with" bytes for the pattern, the data written to the file (though
447        never used) was taken from uninitialized memory and so caused valgrind to
448        complain.
450    27. Updated RunTest.bat as provided by Sheri Pierce.
452    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
454    29. Computation of memory usage for the table of capturing group names was
455        giving an unnecessarily large value.
458    Version 8.20 21-Oct-2011
459    ------------------------
461    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
462        a POSIX class. After further experiments with Perl, which convinced me that
463        Perl has bugs and confusions, a closing square bracket is no longer allowed
464        in a POSIX name. This bug also affected patterns with classes that started
465        with full stops.
467    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
468        captured substring, but while checking the failing first alternative,
469        substring 1 is temporarily captured. If the output vector supplied to
470        pcre_exec() was not big enough for this capture, the yield of the function
471        was still zero ("insufficient space for captured substrings"). This cannot
472        be totally fixed without adding another stack variable, which seems a lot
473        of expense for a edge case. However, I have improved the situation in cases
474        such as /(a)(b)x|abc/ matched against "abc", where the return code
475        indicates that fewer than the maximum number of slots in the ovector have
476        been set.
478    3.  Related to (2) above: when there are more back references in a pattern than
479        slots in the output vector, pcre_exec() uses temporary memory during
480        matching, and copies in the captures as far as possible afterwards. It was
481        using the entire output vector, but this conflicts with the specification
482        that only 2/3 is used for passing back captured substrings. Now it uses
483        only the first 2/3, for compatibility. This is, of course, another edge
484        case.
486    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
487        main code base, and can be used by building with --enable-jit. When this is
488        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
489        runtime --no-jit option is given.
491    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
492        ovector, the return from the function was zero, implying that there were
493        other matches that did not fit. The correct "exactly full" value is now
494        returned.
496    6.  If a subpattern that was called recursively or as a subroutine contained
497        (*PRUNE) or any other control that caused it to give a non-standard return,
498        invalid errors such as "Error -26 (nested recursion at the same subject
499        position)" or even infinite loops could occur.
501    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
502        computing the minimum length on reaching *ACCEPT, and so ended up with the
503        wrong value of 1 rather than 0. Further investigation indicates that
504        computing a minimum subject length in the presence of *ACCEPT is difficult
505        (think back references, subroutine calls), and so I have changed the code
506        so that no minimum is registered for a pattern that contains *ACCEPT.
508    8.  If (*THEN) was present in the first (true) branch of a conditional group,
509        it was not handled as intended. [But see 16 below.]
511    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
512        Sheri Pierce.
514    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
515        the first byte in a match must be "a".
517    11. Change 17 for 8.13 increased the recursion depth for patterns like
518        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
519        pattern contains any instances of (*THEN). If it does not, the old
520        optimizations are restored. It would be nice to do this on a per-group
521        basis, but at the moment that is not feasible.
523    12. In some environments, the output of pcretest -C is CRLF terminated. This
524        broke RunTest's code that checks for the link size. A single white space
525        character after the value is now allowed for.
527    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
528        For "fr", it uses the Windows-specific input and output files.
530    14. If (*THEN) appeared in a group that was called recursively or as a
531        subroutine, it did not work as intended. [But see next item.]
533    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
534        pattern fragments (but not containing any | characters). If A and B are
535        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
536        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
537        D. In other words, Perl considers parentheses that do not contain any |
538        characters to be part of a surrounding alternative, whereas PCRE was
539        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
540        differently. PCRE now behaves in the same way as Perl, except in the case
541        of subroutine/recursion calls such as (?1) which have in any case always
542        been different (but PCRE had them first :-).
544    16. Related to 15 above: Perl does not treat the | in a conditional group as
545        creating alternatives. Such a group is treated in the same way as an
546        ordinary group without any | characters when processing (*THEN). PCRE has
547        been changed to match Perl's behaviour.
549    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
550        RunGrepTest script failed.
552    18. Change 22 for version 13 caused atomic groups to use more stack. This is
553        inevitable for groups that contain captures, but it can lead to a lot of
554        stack use in large patterns. The old behaviour has been restored for atomic
555        groups that do not contain any capturing parentheses.
557    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
558        suppress the check for a minimum subject length at run time. (If it was
559        given to pcre_exec() or pcre_dfa_exec() it did work.)
561    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
562        fail to work when decoding hex characters in data strings in EBCDIC
563        environments.
565    21. It appears that in at least one Mac OS environment, the isxdigit() function
566        is implemented as a macro that evaluates to its argument more than once,
567        contravening the C 90 Standard (I haven't checked a later standard). There
568        was an instance in pcretest which caused it to go wrong when processing
569        \x{...} escapes in subject strings. The has been rewritten to avoid using
570        things like p++ in the argument of isxdigit().
573    Version 8.13 16-Aug-2011
574    ------------------------
576    1.  The Unicode data tables have been updated to Unicode 6.0.0.
578    2.  Two minor typos in pcre_internal.h have been fixed.
580    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
581        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
582        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
584    4.  There were a number of related bugs in the code for matching backrefences
585        caselessly in UTF-8 mode when codes for the characters concerned were
586        different numbers of bytes. For example, U+023A and U+2C65 are an upper
587        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
588        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
589        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
590        2-byte code at the end of the subject (it thought there wasn't enough data
591        left).
593    5.  Comprehensive information about what went wrong is now returned by
594        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
595        as the output vector has at least 2 elements. The offset of the start of
596        the failing character and a reason code are placed in the vector.
598    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
599        now returned is for the first byte of the failing character, instead of the
600        last byte inspected. This is an incompatible change, but I hope it is small
601        enough not to be a problem. It makes the returned offset consistent with
602        pcre_exec() and pcre_dfa_exec().
604    7.  pcretest now gives a text phrase as well as the error number when
605        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
606        failure, the offset and reason code are output.
608    8.  When \R was used with a maximizing quantifier it failed to skip backwards
609        over a \r\n pair if the subsequent match failed. Instead, it just skipped
610        back over a single character (\n). This seems wrong (because it treated the
611        two characters as a single entity when going forwards), conflicts with the
612        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
613        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
614        has been changed.
616    9.  Some internal refactoring has changed the processing so that the handling
617        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
618        time (the PCRE_DOTALL option was changed this way some time ago: version
619        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
620        which was always a bit of a fudge. It also means that there is one less
621        argument for the match() function, which reduces its stack requirements
622        slightly. This change also fixes an incompatibility with Perl: the pattern
623        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
625    10. More internal refactoring has drastically reduced the number of recursive
626        calls to match() for possessively repeated groups such as (abc)++ when
627        using pcre_exec().
629    11. While implementing 10, a number of bugs in the handling of groups were
630        discovered and fixed:
632        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
633        (a|)*(?1) gave a compile-time internal error.
634        ((a|)+)+  did not notice that the outer group could match an empty string.
635        (^a|^)+   was not marked as anchored.
636        (.*a|.*)+ was not marked as matching at start or after a newline.
638    12. Yet more internal refactoring has removed another argument from the match()
639        function. Special calls to this function are now indicated by setting a
640        value in a variable in the "match data" data block.
642    13. Be more explicit in pcre_study() instead of relying on "default" for
643        opcodes that mean there is no starting character; this means that when new
644        ones are added and accidentally left out of pcre_study(), testing should
645        pick them up.
647    14. The -s option of pcretest has been documented for ages as being an old
648        synonym of -m (show memory usage). I have changed it to mean "force study
649        for every regex", that is, assume /S for every regex. This is similar to -i
650        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
651        using it. It makes it easier to run collections of tests with and without
652        study enabled, and thereby test pcre_study() more easily. All the standard
653        tests are now run with and without -s (but some patterns can be marked as
654        "never study" - see 20 below).
656    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
657        restoration of the capturing data to the outer values was not happening
658        correctly.
660    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
661        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
662        pattern had matched an empty string, and so incorrectly returned a no
663        match.
665    17. There was optimizing code for the last branch of non-capturing parentheses,
666        and also for the obeyed branch of a conditional subexpression, which used
667        tail recursion to cut down on stack usage. Unfortunately, now that there is
668        the possibility of (*THEN) occurring in these branches, tail recursion is
669        no longer possible because the return has to be checked for (*THEN). These
670        two optimizations have therefore been removed. [But see 8.20/11 above.]
672    18. If a pattern containing \R was studied, it was assumed that \R always
673        matched two bytes, thus causing the minimum subject length to be
674        incorrectly computed because \R can also match just one byte.
676    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
677        was incorrectly computed.
679    20. If /S is present twice on a test pattern in pcretest input, it now
680        *disables* studying, thereby overriding the use of -s on the command line
681        (see 14 above). This is necessary for one or two tests to keep the output
682        identical in both cases.
684    21. When (*ACCEPT) was used in an assertion that matched an empty string and
685        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
687    22. When an atomic group that contained a capturing parenthesis was
688        successfully matched, but the branch in which it appeared failed, the
689        capturing was not being forgotten if a higher numbered group was later
690        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
691        group 1 to "a", when in fact it should be unset. This applied to multi-
692        branched capturing and non-capturing groups, repeated or not, and also to
693        positive assertions (capturing in negative assertions does not happen
694        in PCRE) and also to nested atomic groups.
696    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
697        subject after a captured substring, to make it easier to tell which of a
698        number of identical substrings has been captured.
700    24. The way atomic groups are processed by pcre_exec() has been changed so that
701        if they are repeated, backtracking one repetition now resets captured
702        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
703        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
704        "aaa". Previously, it would have been "a". As part of this code
705        refactoring, the way recursive calls are handled has also been changed.
707    25. If an assertion condition captured any substrings, they were not passed
708        back unless some other capturing happened later. For example, if
709        (?(?=(a))a) was matched against "a", no capturing was returned.
711    26. When studying a pattern that contained subroutine calls or assertions,
712        the code for finding the minimum length of a possible match was handling
713        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
714        group 1 called group 2 while simultaneously a separate group 2 called group
715        1). A stack overflow occurred in this case. I have fixed this by limiting
716        the recursion depth to 10.
718    27. Updated RunTest.bat in the distribution to the version supplied by Tom
719        Fortmann. This supports explicit test numbers on the command line, and has
720        argument validation and error reporting.
722    28. An instance of \X with an unlimited repeat could fail if at any point the
723        first character it looked at was a mark character.
725    29. Some minor code refactoring concerning Unicode properties and scripts
726        should reduce the stack requirement of match() slightly.
728    30. Added the '=' option to pcretest to check the setting of unused capturing
729        slots at the end of the pattern, which are documented as being -1, but are
730        not included in the return count.
732    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
733        compiled something random. Now it gives a compile-time error (as does
734        Perl).
736    32. A *MARK encountered during the processing of a positive assertion is now
737        recorded and passed back (compatible with Perl).
739    33. If --only-matching or --colour was set on a pcregrep call whose pattern
740        had alternative anchored branches, the search for a second match in a line
741        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
742        matched the line "0102" twice. The same bug affected patterns that started
743        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
744        twice.
746    34. Previously, PCRE did not allow quantification of assertions. However, Perl
747        does, and because of capturing effects, quantifying parenthesized
748        assertions may at times be useful. Quantifiers are now allowed for
749        parenthesized assertions.
751    35. A minor code tidy in pcre_compile() when checking options for \R usage.
753    36. \g was being checked for fancy things in a character class, when it should
754        just be a literal "g".
756    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
757        appearance of a nested POSIX class supersedes an apparent external class.
758        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
759        unescaped square brackets may also appear as part of class names. For
760        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
761        more like Perl. (But see 8.20/1 above.)
763    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
764        was because it thought it was \N{name}, which is not supported).
766    39. Add minix to OS list not supporting the -S option in pcretest.
768    40. PCRE tries to detect cases of infinite recursion at compile time, but it
769        cannot analyze patterns in sufficient detail to catch mutual recursions
770        such as ((?1))((?2)). There is now a runtime test that gives an error if a
771        subgroup is called recursively as a subpattern for a second time at the
772        same position in the subject string. In previous releases this might have
773        been caught by the recursion limit, or it might have run out of stack.
775    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
776        happen only once. PCRE was, however incorrectly giving a compile time error
777        "recursive call could loop indefinitely" because it cannot analyze the
778        pattern in sufficient detail. The compile time test no longer happens when
779        PCRE is compiling a conditional subpattern, but actual runaway loops are
780        now caught at runtime (see 40 above).
782    42. It seems that Perl allows any characters other than a closing parenthesis
783        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
784        has been changed to be the same.
786    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
787        as not to get warnings when autogen.sh is called. Also changed
788        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
790    44. To help people who use pcregrep to scan files containing exceedingly long
791        lines, the following changes have been made:
793        (a) The default value of the buffer size parameter has been increased from
794            8K to 20K. (The actual buffer used is three times this size.)
796        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
797            PCRE is built.
799        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
800            to be set at run time.
802        (d) Numerical values in pcregrep options can be followed by K or M, for
803            example --buffer-size=50K.
805        (e) If a line being scanned overflows pcregrep's buffer, an error is now
806            given and the return code is set to 2.
808    45. Add a pointer to the latest mark to the callout data block.
810    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
811        partial match of an empty string instead of no match. This was specific to
812        the use of ".".
814    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
815        complete match instead of a partial match. This bug was dependent on both
816        the PCRE_UTF8 and PCRE_DOTALL options being set.
818    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
819        starting byte set, because \b was not being ignored.
822    Version 8.12 15-Jan-2011
823    ------------------------
825    1.  Fixed some typos in the markup of the man pages, and wrote a script that
826        checks for such things as part of the documentation building process.
828    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
829        --match-limit and --recursion-limit options (added for 8.11). In
830        particular, this made one of the standard tests fail. (The integer value
831        went into the wrong half of a long int.)
833    3.  If the --colour option was given to pcregrep with -v (invert match), it
834        did strange things, either producing crazy output, or crashing. It should,
835        of course, ignore a request for colour when reporting lines that do not
836        match.
838    4.  Another pcregrep bug caused similar problems if --colour was specified with
839        -M (multiline) and the pattern match finished with a line ending.
841    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
842        matched in multiline mode, the following line was shown as part of the
843        match. This seems wrong, so I have changed it.
845    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
846        the check for further matches in the same line (so they could be coloured)
847        to overrun the end of the current line. If another match was found, it was
848        incorrectly shown (and then shown again when found in the next line).
850    7.  If pcregrep was compiled under Windows, there was a reference to the
851        function pcregrep_exit() before it was defined. I am assuming this was
852        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
853        reported by a user. I've moved the definition above the reference.
856    Version 8.11 10-Dec-2010
857    ------------------------
859    1.  (*THEN) was not working properly if there were untried alternatives prior
860        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
861        backtracked to try for "b" instead of moving to the next alternative branch
862        at the same level (in this case, to look for "c"). The Perl documentation
863        is clear that when (*THEN) is backtracked onto, it goes to the "next
864        alternative in the innermost enclosing group".
866    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
867        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
868        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
869        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
870        (*THEN).
872    3.  If \s appeared in a character class, it removed the VT character from
873        the class, even if it had been included by some previous item, for example
874        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
875        of \s, but is part of the POSIX "space" class.)
877    4.  A partial match never returns an empty string (because you can always
878        match an empty string at the end of the subject); however the checking for
879        an empty string was starting at the "start of match" point. This has been
880        changed to the "earliest inspected character" point, because the returned
881        data for a partial match starts at this character. This means that, for
882        example, /(?<=abc)def/ gives a partial match for the subject "abc"
883        (previously it gave "no match").
885    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
886        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
887        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
888        has an implication that the given string is incomplete (because a partial
889        match is preferred over a full match). For this reason, these items now
890        give a partial match in this situation. [Aside: previously, the one case
891        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
892        match rather than a full match, which was wrong by the old rules, but is
893        now correct.]
895    6.  There was a bug in the handling of #-introduced comments, recognized when
896        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
897        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
898        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
899        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
900        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
901        places in pcre_compile().
903    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
904        comments when looking ahead for named forward references to subpatterns,
905        the only newline sequence it recognized was NL. It now handles newlines
906        according to the set newline convention.
908    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
909        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
910        cater for a lack of strerror(). These oversights have been fixed.
912    9.  Added --match-limit and --recursion-limit to pcregrep.
914    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
916    11. When the -o option was used, pcregrep was setting a return code of 1, even
917        when matches were found, and --line-buffered was not being honoured.
919    12. Added an optional parentheses number to the -o and --only-matching options
920        of pcregrep.
922    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
923        can match an empty string. The code to do it in pcretest and pcredemo
924        needed fixing:
926        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
927            only one byte after an empty string match just before CRLF (this case
928            just got forgotten; "any" and "anycrlf" were OK).
930        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
931            mode when an empty string match preceded an ASCII character followed by
932            a non-ASCII character. (The code for advancing by one character rather
933            than one byte was nonsense.)
935        (c) The pcredemo.c sample program did not have any code at all to handle
936            the cases when CRLF is a valid newline sequence.
938    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
939        as a starting offset was within the subject string. There is now a new
940        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
941        negative or greater than the length of the string. In order to test this,
942        pcretest is extended to allow the setting of negative starting offsets.
944    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
945        starting offset points to the beginning of a UTF-8 character was
946        unnecessarily clumsy. I tidied it up.
948    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
949        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
951    17. Nobody had reported that the --include_dir option, which was added in
952        release 7.7 should have been called --include-dir (hyphen, not underscore)
953        for compatibility with GNU grep. I have changed it to --include-dir, but
954        left --include_dir as an undocumented synonym, and the same for
955        --exclude-dir, though that is not available in GNU grep, at least as of
956        release 2.5.4.
958    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
959        characters from a string of bytes) have been redefined so as not to use
960        loops, in order to improve performance in some environments. At the same
961        time, I abstracted some of the common code into auxiliary macros to save
962        repetition (this should not affect the compiled code).
964    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
965        compile-time error is now given if \c is not followed by an ASCII
966        character, that is, a byte less than 128. (In EBCDIC mode, the code is
967        different, and any byte value is allowed.)
969    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
970        START_OPTIMIZE option, which is now allowed at compile time - but just
971        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
972        to pcregrep and other applications that have no direct access to PCRE
973        options. The new /Y option in pcretest sets this option when calling
974        pcre_compile().
976    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
977        back references. Groups containing recursive back references were forced to
978        be atomic by that change, but in the case of named groups, the amount of
979        memory required was incorrectly computed, leading to "Failed: internal
980        error: code overflow". This has been fixed.
982    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
983        pcretest.c, to avoid build problems in some Borland environments.
986    Version 8.10 25-Jun-2010
987    ------------------------
989    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
990        THEN.
992    2.  (*ACCEPT) was not working when inside an atomic group.
994    3.  Inside a character class, \B is treated as a literal by default, but
995        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
996        causes the error). The code is unchanged, but I tidied the documentation.
998    4.  Inside a character class, PCRE always treated \R and \X as literals,
999        whereas Perl faults them if its -w option is set. I have changed PCRE so
1000        that it faults them when PCRE_EXTRA is set.
1002    5.  Added support for \N, which always matches any character other than
1003        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1005    6.  When compiling pcregrep with newer versions of gcc which may have
1006        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1007        declared with attribute warn_unused_result" were given. Just casting the
1008        result to (void) does not stop the warnings; a more elaborate fudge is
1009        needed. I've used a macro to implement this.
1011    7.  Minor change to pcretest.c to avoid a compiler warning.
1013    8.  Added four artifical Unicode properties to help with an option to make
1014        \s etc use properties (see next item). The new properties are: Xan
1015        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1017    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1018        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1019        this option. Modified pcretest to add /W to test this facility. Added
1020        REG_UCP to make it available via the POSIX interface.
1022    10. Added --line-buffered to pcregrep.
1024    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1025        studied, and the match started with a letter with a code point greater than
1026        127 whose first byte was different to the first byte of the other case of
1027        the letter, the other case of this starting letter was not recognized
1028        (#976).
1030    12. If a pattern that was studied started with a repeated Unicode property
1031        test, for example, \p{Nd}+, there was the theoretical possibility of
1032        setting up an incorrect bitmap of starting bytes, but fortunately it could
1033        not have actually happened in practice until change 8 above was made (it
1034        added property types that matched character-matching opcodes).
1036    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1037        possible starting bytes for non-anchored patterns.
1039    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1040        \R, and also a number of cases that involve Unicode properties, both
1041        explicit and implicit when PCRE_UCP is set.
1043    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1044        input, it could crash or give wrong results if characters with values
1045        greater than 0xc0 were present in the subject string. (Detail: it assumed
1046        UTF-8 input when processing these items.)
1048    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1049        size_t is 64-bit (#991).
1051    17. Added a check for running out of memory when PCRE is compiled with
1052        --disable-stack-for-recursion (#990).
1054    18. If the last data line in a file for pcretest does not have a newline on
1055        the end, a newline was missing in the output.
1057    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1058        less than 128) in its various bitmaps. However, there is a facility for
1059        generating tables according to the current locale when PCRE is compiled. It
1060        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1061        characters, are recognized by isspace() and therefore were getting set in
1062        these tables, and indeed these tables seem to approximate to ISO 8859. This
1063        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1064        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1065        which of course cannot start UTF-8 characters. I have changed the code so
1066        that only real ASCII characters (less than 128) and the correct starting
1067        bytes for UTF-8 encodings are set for characters greater than 127 when in
1068        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1069        altogether.)
1071    20. Added the /T option to pcretest so as to be able to run tests with non-
1072        standard character tables, thus making it possible to include the tests
1073        used for 19 above in the standard set of tests.
1075    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1076        reference to a subpattern the other side of a comment that contains an
1077        opening parenthesis caused either an internal compiling error, or a
1078        reference to the wrong subpattern.
1081    Version 8.02 19-Mar-2010
1082    ------------------------
1084    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1086    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1087        configured.
1089    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1090        original author of that file, following a query about its status.
1092    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1093        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1095    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1096        quantifier applied to a forward-referencing subroutine call, could compile
1097        incorrect code or give the error "internal error: previously-checked
1098        referenced subpattern not found".
1100    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1101        variables to point to external functions. For these systems, therefore,
1102        pcre_malloc etc. are now initialized to local functions that call the
1103        relevant global functions.
1105    7.  There were two entries missing in the vectors called coptable and poptable
1106        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1107        I've fixed the data, and added a kludgy way of testing at compile time that
1108        the lengths are correct (equal to the number of opcodes).
1110    8.  Following on from 7, I added a similar kludge to check the length of the
1111        eint vector in pcreposix.c.
1113    9.  Error texts for pcre_compile() are held as one long string to avoid too
1114        much relocation at load time. To find a text, the string is searched,
1115        counting zeros. There was no check for running off the end of the string,
1116        which could happen if a new error number was added without updating the
1117        string.
1119    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1121    11. \K was not working if it appeared in an atomic group or in a group that
1122        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1123        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1124        the assertion is positive, but not if it is negative.
1126    12. Change 11 fortuitously reduced the size of the stack frame used in the
1127        "match()" function of pcre_exec.c by one pointer. Forthcoming
1128        implementation of support for (*MARK) will need an extra pointer on the
1129        stack; I have reserved it now, so that the stack frame size does not
1130        decrease.
1132    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1133        item in branch that calls a recursion is a subroutine call - as in the
1134        second branch in the above example - was incorrectly given the compile-
1135        time error "recursive call could loop indefinitely" because pcre_compile()
1136        was not correctly checking the subroutine for matching a non-empty string.
1138    14. The checks for overrunning compiling workspace could trigger after an
1139        overrun had occurred. This is a "should never occur" error, but it can be
1140        triggered by pathological patterns such as hundreds of nested parentheses.
1141        The checks now trigger 100 bytes before the end of the workspace.
1143    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1146    Version 8.01 19-Jan-2010
1147    ------------------------
1149    1.  If a pattern contained a conditional subpattern with only one branch (in
1150        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1151        computed the wrong minimum data length (which is of course zero for such
1152        subpatterns). This could cause incorrect "no match" results.
1154    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1155        the pattern is reset in the first branch, pcre_compile() failed with
1156        "internal error: code overflow at offset...". This happened only when
1157        the reset was to the original external option setting. (An optimization
1158        abstracts leading options settings into an external setting, which was the
1159        cause of this.)
1161    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1162        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1163        assertion pattern did not match (meaning that the assertion was true), it
1164        was incorrectly treated as false if the SKIP had been reached during the
1165        matching. This also applied to assertions used as conditions.
1167    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1168        assertion subpattern, including such a pattern used as a condition,
1169        unpredictable results occurred, instead of the error return
1172    5.  The C++ GlobalReplace function was not working like Perl for the special
1173        situation when an empty string is matched. It now does the fancy magic
1174        stuff that is necessary.
1176    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1177        removed. (These were left over from very, very early versions of PCRE.)
1179    7.  Some cosmetic changes to the code to make life easier when compiling it
1180        as part of something else:
1182        (a) Change DEBUG to PCRE_DEBUG.
1184        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1185            called "current" as "current_branch", to prevent a collision with the
1186            Linux macro when compiled as a kernel module.
1188        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1189            prevent a collision with the Linux macro when compiled as a kernel
1190            module.
1192    8.  In pcre_compile() there are some checks for integer overflows that used to
1193        cast potentially large values to (double). This has been changed to that
1194        when building, a check for int64_t is made, and if it is found, it is used
1195        instead, thus avoiding the use of floating point arithmetic. (There is no
1196        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1197        double.
1199    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1200        2005 (difference between two addresses compared to an unsigned value).
1202    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1203        custom one, because of the following reported problem in Windows:
1205          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1206              under Win32.
1207          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1208              therefore missing the function definition.
1209          - The compiler thus generates a "C" signature for the test function.
1210          - The linker fails to find the "C" function.
1211          - PCRE fails to configure if asked to do so against libbz2.
1213    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1214        messages were output:
1216          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1217          rerunning libtoolize, to keep the correct libtool macros in-tree.
1218          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1220        I have done both of these things.
1222    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1223        most of the time, it *can* run out if it is given a pattern that contains a
1224        runaway infinite recursion. I updated the discussion in the pcrestack man
1225        page.
1227    13. Now that we have gone to the x.xx style of version numbers, the minor
1228        version may start with zero. Using 08 or 09 is a bad idea because users
1229        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1230        interpreted as invalid octal numbers. I've updated the previous comment in
1231        configure.ac, and also added a check that gives an error if 08 or 09 are
1232        used.
1234    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1235        causing partial matching to fail when the end of the subject matched \W
1236        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1238    15. There were some discrepancies between the declarations in pcre_internal.h
1239        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1240        their definitions. The declarations used "const uschar *" and the
1241        definitions used USPTR. Even though USPTR is normally defined as "const
1242        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1243        reported that: "This difference in casting confuses some C++ compilers, for
1244        example, SunCC recognizes above declarations as different functions and
1245        generates broken code for hbpcre." I have changed the declarations to use
1246        USPTR.
1248    16. GNU libtool is named differently on some systems. The autogen.sh script now
1249        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1250        (FreeBSD).
1252    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1253        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1254        comment: "Figure out how to create a longlong from a string: strtoll and
1255        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1256        instance, but it only takes 2 args instead of 3!"
1258    18. A subtle bug concerned with back references has been fixed by a change of
1259        specification, with a corresponding code fix. A pattern such as
1260        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1261        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1262        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1263        same bug. Such groups have to be quantified to be useful, or contained
1264        inside another quantified group. (If there's no repetition, the reference
1265        can never match.) The problem arises because, having left the group and
1266        moved on to the rest of the pattern, a later failure that backtracks into
1267        the group uses the captured value from the final iteration of the group
1268        rather than the correct earlier one. I have fixed this in PCRE by forcing
1269        any group that contains a reference to itself to be an atomic group; that
1270        is, there cannot be any backtracking into it once it has completed. This is
1271        similar to recursive and subroutine calls.
1274    Version 8.00 19-Oct-09
1275    ----------------------
1277    1.  The table for translating pcre_compile() error codes into POSIX error codes
1278        was out-of-date, and there was no check on the pcre_compile() error code
1279        being within the table. This could lead to an OK return being given in
1280        error.
1282    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1283        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1284        in a Windows environment.
1286    3.  The pcregrep --count option prints the count for each file even when it is
1287        zero, as does GNU grep. However, pcregrep was also printing all files when
1288        --files-with-matches was added. Now, when both options are given, it prints
1289        counts only for those files that have at least one match. (GNU grep just
1290        prints the file name in this circumstance, but including the count seems
1291        more useful - otherwise, why use --count?) Also ensured that the
1292        combination -clh just lists non-zero counts, with no names.
1294    4.  The long form of the pcregrep -F option was incorrectly implemented as
1295        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1296        but it seems right to fix it, and I didn't think it was worth preserving
1297        the old behaviour.
1299    5.  The command line items --regex=pattern and --regexp=pattern were not
1300        recognized by pcregrep, which required --regex pattern or --regexp pattern
1301        (with a space rather than an '='). The man page documented the '=' forms,
1302        which are compatible with GNU grep; these now work.
1304    6.  No libpcreposix.pc file was created for pkg-config; there was just
1305        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1307    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1308        when UCP support is not needed, by modifying the Python script that
1309        generates it from Unicode data files. This should not matter if the module
1310        is correctly used as a library, but I received one complaint about 50K of
1311        unwanted data. My guess is that the person linked everything into his
1312        program rather than using a library. Anyway, it does no harm.
1314    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1315        was a minimum greater than 1 for a wide character in a possessive
1316        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1317        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1318        character. Chaos in the form of incorrect output or a compiling loop could
1319        result.
1321    9.  The restrictions on what a pattern can contain when partial matching is
1322        requested for pcre_exec() have been removed. All patterns can now be
1323        partially matched by this function. In addition, if there are at least two
1324        slots in the offset vector, the offset of the earliest inspected character
1325        for the match and the offset of the end of the subject are set in them when
1326        PCRE_ERROR_PARTIAL is returned.
1328    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1329        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1330        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1331        and may be more useful for multi-segment matching.
1333    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1334        used to be given if ever the end of the subject was reached; now it is
1335        given only if matching could not proceed because another character was
1336        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1337        string "Z", which now yields "no match" instead of "partial match". In the
1338        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1339        final character ended with (*FAIL).
1341    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1342        if the pattern had a "must contain" character that was already found in the
1343        earlier partial match, unless partial matching was again requested. For
1344        example, with the pattern /dog.(body)?/, the "must contain" character is
1345        "g". If the first part-match was for the string "dog", restarting with
1346        "sbody" failed. This bug has been fixed.
1348    13. The string returned by pcre_dfa_exec() after a partial match has been
1349        changed so that it starts at the first inspected character rather than the
1350        first character of the match. This makes a difference only if the pattern
1351        starts with a lookbehind assertion or \b or \B (\K is not supported by
1352        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1353        matching functions compatible, and I think it's the right thing to do.
1355    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1356        so that the demonstration program is easily available in environments where
1357        PCRE has not been installed from source.
1359    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1360        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1361        library.
1363    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1364        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1365        is not the first non-POSIX option to be added. Clearly some people find
1366        these options useful.
1368    17. If a caller to the POSIX matching function regexec() passes a non-zero
1369        value for nmatch with a NULL value for pmatch, the value of
1370        nmatch is forced to zero.
1372    18. RunGrepTest did not have a test for the availability of the -u option of
1373        the diff command, as RunTest does. It now checks in the same way as
1374        RunTest, and also checks for the -b option.
1376    19. If an odd number of negated classes containing just a single character
1377        interposed, within parentheses, between a forward reference to a named
1378        subpattern and the definition of the subpattern, compilation crashed with
1379        an internal error, complaining that it could not find the referenced
1380        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1381        [The bug was that it was starting one character too far in when skipping
1382        over the character class, thus treating the ] as data rather than
1383        terminating the class. This meant it could skip too much.]
1385    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1386        /g option in pcretest when the pattern contains \K, which makes it possible
1387        to have an empty string match not at the start, even when the pattern is
1388        anchored. Updated pcretest and pcredemo to use this option.
1390    21. If the maximum number of capturing subpatterns in a recursion was greater
1391        than the maximum at the outer level, the higher number was returned, but
1392        with unset values at the outer level. The correct (outer level) value is
1393        now given.
1395    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1396        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1397        make it do so. The string so far is captured, making this feature
1398        compatible with Perl.
1400    23. The tests have been re-organized, adding tests 11 and 12, to make it
1401        possible to check the Perl 5.10 features against Perl 5.10.
1403    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1404        pattern matches a fixed length string. PCRE did not allow this; now it
1405        does. Neither allows recursion.
1407    25. I finally figured out how to implement a request to provide the minimum
1408        length of subject string that was needed in order to match a given pattern.
1409        (It was back references and recursion that I had previously got hung up
1410        on.) This code has now been added to pcre_study(); it finds a lower bound
1411        to the length of subject needed. It is not necessarily the greatest lower
1412        bound, but using it to avoid searching strings that are too short does give
1413        some useful speed-ups. The value is available to calling programs via
1414        pcre_fullinfo().
1416    26. While implementing 25, I discovered to my embarrassment that pcretest had
1417        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1418        study optimizations had never been tested with that matching function.
1419        Oops. What is worse, even when it was passed study data, there was a bug in
1420        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1421        were also very few tests of studied patterns with pcre_dfa_exec().
1423    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1424        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1425        on the other side of the coin, they are no longer allowed to have different
1426        names, because these cannot be distinguished in PCRE, and this has caused
1427        confusion. (This is a difference from Perl.)
1429    28. When duplicate subpattern names are present (necessarily with different
1430        numbers, as required by 27 above), and a test is made by name in a
1431        conditional pattern, either for a subpattern having been matched, or for
1432        recursion in such a pattern, all the associated numbered subpatterns are
1433        tested, and the overall condition is true if the condition is true for any
1434        one of them. This is the way Perl works, and is also more like the way
1435        testing by number works.
1438    Version 7.9 11-Apr-09
1439    ---------------------
1441    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1442        (pcretest), all targets were linked against these libraries. This included
1443        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1444        libraries. This caused unwanted dependencies to be created. This problem
1445        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1446        pcretest is linked with readline.
1448    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1449        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1450        moved outside it again, because FALSE and TRUE are already defined in AIX,
1451        but BOOL is not.
1453    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1454        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1456    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1457        hyphens) following file names and line numbers when outputting matching
1458        lines. This is not true; no spaces are inserted. I have also clarified the
1459        wording for the --colour (or --color) option.
1461    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1462        was not coloured; this is different to GNU grep, so I have changed it to be
1463        the same.
1465    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1466        each matching line was coloured. Now it goes on to look for further matches
1467        of any of the test patterns, which is the same behaviour as GNU grep.
1469    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1470        doesn't make sense to accept an empty string match in pcregrep, so I have
1471        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1472        seems to be how GNU grep behaves.
1474    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1475        start or after a newline", because the conditional assertion was not being
1476        correctly handled. The rule now is that both the assertion and what follows
1477        in the first alternative must satisfy the test.
1479    9.  If auto-callout was enabled in a pattern with a conditional group whose
1480        condition was an assertion, PCRE could crash during matching, both with
1481        pcre_exec() and pcre_dfa_exec().
1483    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1484        used for matching.
1486    11. Unicode property support in character classes was not working for
1487        characters (bytes) greater than 127 when not in UTF-8 mode.
1489    12. Added the -M command line option to pcretest.
1491    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1493    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1495    16. Added comments and documentation about mis-use of no_arg in the C++
1496        wrapper.
1498    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1499        from Martin Jerabek that uses macro names for all relevant character and
1500        string constants.
1502    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1503        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1504        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1505        these, but not everybody uses configure.
1507    19. A conditional group that had only one branch was not being correctly
1508        recognized as an item that could match an empty string. This meant that an
1509        enclosing group might also not be so recognized, causing infinite looping
1510        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1511        with the subject "ab", where knowledge that the repeated group can match
1512        nothing is needed in order to break the loop.
1514    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1515        exec(), but without supplying a callout function, matching went wrong.
1517    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1518        leak if the size of the offset vector was greater than 30. When the vector
1519        is smaller, the saved offsets during recursion go onto a local stack
1520        vector, but for larger vectors malloc() is used. It was failing to free
1521        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1522        error, in fact).
1524    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1525        heapframe that is used only when UTF-8 support is enabled. This caused no
1526        problem, but was untidy.
1528    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1529        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1530        included within another project.
1532    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1533        slightly modified by me:
1535          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1536              not building pcregrep.
1538          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1539              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1541    25. Forward references, both numeric and by name, in patterns that made use of
1542        duplicate group numbers, could behave incorrectly or give incorrect errors,
1543        because when scanning forward to find the reference group, PCRE was not
1544        taking into account the duplicate group numbers. A pattern such as
1545        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1547    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1548        the feature of a custom pointer more persuasive (as requested by a user).
1550    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1551        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1552        pre-defined.
1554    28. Added support for (*UTF8) at the start of a pattern.
1556    29. Arrange for flags added by the "release type" setting in CMake to be shown
1557        in the configuration summary.
1560    Version 7.8 05-Sep-08
1561    ---------------------
1563    1.  Replaced UCP searching code with optimized version as implemented for Ad
1564        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1565        stage table and inline lookup instead of a function, giving speed ups of 2
1566        to 5 times on some simple patterns that I tested. Permission was given to
1567        distribute the MultiStage2.py script that generates the tables (it's not in
1568        the tarball, but is in the Subversion repository).
1570    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1571        scripts.
1573    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1574        a group with a zero qualifier. The result of the study could be incorrect,
1575        or the function might crash, depending on the pattern.
1577    4.  Caseless matching was not working for non-ASCII characters in back
1578        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1579        It now works when Unicode Property Support is available.
1581    5.  In pcretest, an escape such as \x{de} in the data was always generating
1582        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1583        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1584        truncation.
1586    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1588    7.  Added two (int) casts to pcregrep when printing the difference of two
1589        pointers, in case they are 64-bit values.
1591    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1592        test 2 if it fails.
1594    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1595        and a #define of that name to empty if it is not externally set. This is to
1596        allow users of MSVC to set it if necessary.
1598    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1599        the convenience functions in the pcre_get.c source file.
1601    11. An option change at the start of a pattern that had top-level alternatives
1602        could cause overwriting and/or a crash. This command provoked a crash in
1603        some environments:
1605          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1607        This potential security problem was recorded as CVE-2008-2371.
1609    12. For a pattern where the match had to start at the beginning or immediately
1610        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1611        pcre_dfa_exec() could read past the end of the passed subject if there was
1612        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1613        pcretest so that it places the subject at the end of its malloc-ed buffer.
1615    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1616        exec() might read past the end of the data buffer in UTF-8 mode.
1618    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1619        the data contained the byte 0x85 as part of a UTF-8 character within its
1620        first line. This applied both to normal and DFA matching.
1622    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1623        /^[^d]*?$/8 failed to match "abc".
1625    16. Added a missing copyright notice to pcrecpp_internal.h.
1627    17. Make it more clear in the documentation that values returned from
1628        pcre_exec() in ovector are byte offsets, not character counts.
1630    18. Tidied a few places to stop certain compilers from issuing warnings.
1632    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1633        supplied by Stefan Weber. I made a further small update for 7.8 because
1634        there is a change of source arrangements: the pcre_searchfuncs.c module is
1635        replaced by pcre_ucd.c.
1638    Version 7.7 07-May-08
1639    ---------------------
1641    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1642        a string to a long long, pretend we don't even have a long long." This is
1643        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1645    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1646        pre-7.6 versions, which defined a global no_arg variable instead of putting
1647        it in the RE class. (See also #8 below.)
1649    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1650        Lopes.
1652    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1654        (1) The include/exclude patterns were being applied to the whole pathnames
1655            of files, instead of just to the final components.
1657        (2) If there was more than one level of directory, the subdirectories were
1658            skipped unless they satisfied the include/exclude conditions. This is
1659            inconsistent with GNU grep (and could even be seen as contrary to the
1660            pcregrep specification - which I improved to make it absolutely clear).
1661            The action now is always to scan all levels of directory, and just
1662            apply the include/exclude patterns to regular files.
1664    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1665        --exclude_dir in the tests to avoid scanning .svn directories.
1667    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1668        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1669        doesn't support NULs in patterns.
1671    7.  Added some missing "const"s to declarations of static tables in
1672        pcre_compile.c and pcre_dfa_exec.c.
1674    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1675        caused by fix #2  above. (Subsequently also a second patch to fix the
1676        first patch. And a third patch - this was a messy problem.)
1678    9.  Applied Craig's patch to remove the use of push_back().
1680    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1681        matching function regexec().
1683    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1684        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1685        references. PCRE supports relative numbers with this syntax (I don't think
1686        Oniguruma does).
1688    12. Previously, a group with a zero repeat such as (...){0} was completely
1689        omitted from the compiled regex. However, this means that if the group
1690        was called as a subroutine from elsewhere in the pattern, things went wrong
1691        (an internal error was given). Such groups are now left in the compiled
1692        pattern, with a new opcode that causes them to be skipped at execution
1693        time.
1695    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1696        to the way PCRE behaves:
1698        (a) A lone ] character is dis-allowed (Perl treats it as data).
1700        (b) A back reference to an unmatched subpattern matches an empty string
1701            (Perl fails the current match path).
1703        (c) A data ] in a character class must be notated as \] because if the
1704            first data character in a class is ], it defines an empty class. (In
1705            Perl it is not possible to have an empty class.) The empty class []
1706            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1707            The negative empty class [^] matches any one character, independently
1708            of the DOTALL setting.
1710    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1711        non-existent subpattern following a character class starting with ']' and
1712        containing () gave an internal compiling error instead of "reference to
1713        non-existent subpattern". Fortunately, when the pattern did exist, the
1714        compiled code was correct. (When scanning forwards to check for the
1715        existencd of the subpattern, it was treating the data ']' as terminating
1716        the class, so got the count wrong. When actually compiling, the reference
1717        was subsequently set up correctly.)
1719    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1720        it was being rejected as not supported by pcre_dfa_exec(), even though
1721        other assertions are supported. I have made pcre_dfa_exec() support
1722        (*FAIL).
1724    16. The implementation of 13c above involved the invention of a new opcode,
1725        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1726        cannot be changed at match time, I realized I could make a small
1727        improvement to matching performance by compiling OP_ALLANY instead of
1728        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1729        on the OP_ANY path.
1731    17. Compiling pcretest on Windows with readline support failed without the
1732        following two fixes: (1) Make the unistd.h include conditional on
1733        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1735    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1736        ncurses library to be included for pcretest when ReadLine support is
1737        requested, but also to allow for it to be overridden. This patch came from
1738        Daniel Bergström.
1740    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1741        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1742        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1743        spotting this.
1746    Version 7.6 28-Jan-08
1747    ---------------------
1749    1.  A character class containing a very large number of characters with
1750        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1751        overflow.
1753    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1754        HAVE_LONG_LONG is not defined.
1756    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1757        bring them up to date and include new features. This patch includes:
1759        - Fixed PH's badly added libz and libbz2 support.
1760        - Fixed a problem with static linking.
1761        - Added pcredemo. [But later removed - see 7 below.]
1762        - Fixed dftables problem and added an option.
1763        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1764            HAVE_LONG_LONG.
1765        - Added readline support for pcretest.
1766        - Added an listing of the option settings after cmake has run.
1768    4.  A user submitted a patch to Makefile that makes it easy to create
1769        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1770        Makefile.am that cause it to include this special target, without
1771        affecting anything else. Note that the same mingw target plus all
1772        the other distribution libraries and programs are now supported
1773        when configuring with CMake (see 6 below) instead of with
1774        Configure/Make.
1776    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1777        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1778        exported in the Windows port". It has not yet been confirmed that the patch
1779        solves the problem, but it does no harm.
1781    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1782        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1783        with CMake, and also correct the comment about stack recursion.
1785    7.  Remove the automatic building of pcredemo from the ./configure system and
1786        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1787        of a program that users should build themselves after PCRE is installed, so
1788        building it automatically is not really right. What is more, it gave
1789        trouble in some build environments.
1791    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1794    Version 7.5 10-Jan-08
1795    ---------------------
1797    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1798        values in parens when parsing an RE using the C++ wrapper."
1800    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1801        Characters greater than 255 were excluded from the class instead of being
1802        included.
1804    3.  The same bug as (2) above applied to negated POSIX classes such as
1805        [:^space:].
1807    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1808        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1809        I have changed it.
1811    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1812        first named subpattern) and a construct such as (?&a) would reference the
1813        first named subpattern whose name started with "a" (in other words, the
1814        length check was missing). Both these problems are fixed. "Subpattern name
1815        expected" is now given for (?&) (a zero-length name), and this patch also
1816        makes it give the same error for \k'' (previously it complained that that
1817        was a reference to a non-existent subpattern).
1819    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1820        this is right because (?- can be followed by option settings as well as by
1821        digits. I have, however, made the messages clearer.
1823    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1824        than the number used in the conditional) now cause a compile-time error.
1825        This is actually not compatible with Perl, which accepts such patterns, but
1826        treats the conditional as always being FALSE (as PCRE used to), but it
1827        seems to me that giving a diagnostic is better.
1829    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1830        and messages.
1832    9.  Fix two occurrences of "backslash" in comments that should have been
1833        "backspace".
1835    10. Remove two redundant lines of code that can never be obeyed (their function
1836        was moved elsewhere).
1838    11. The program that makes PCRE's Unicode character property table had a bug
1839        which caused it to generate incorrect table entries for sequences of
1840        characters that have the same character type, but are in different scripts.
1841        It amalgamated them into a single range, with the script of the first of
1842        them. In other words, some characters were in the wrong script. There were
1843        thirteen such cases, affecting characters in the following ranges:
1845          U+002b0 - U+002c1
1846          U+0060c - U+0060d
1847          U+0061e - U+00612
1848          U+0064b - U+0065e
1849          U+0074d - U+0076d
1850          U+01800 - U+01805
1851          U+01d00 - U+01d77
1852          U+01d9b - U+01dbf
1853          U+0200b - U+0200f
1854          U+030fc - U+030fe
1855          U+03260 - U+0327f
1856          U+0fb46 - U+0fbb1
1857          U+10450 - U+1049d
1859    12. The -o option (show only the matching part of a line) for pcregrep was not
1860        compatible with GNU grep in that, if there was more than one match in a
1861        line, it showed only the first of them. It now behaves in the same way as
1862        GNU grep.
1864    13. If the -o and -v options were combined for pcregrep, it printed a blank
1865        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1866        does the same. The return code can be used to tell if there were any
1867        non-matching lines.
1869    14. Added --file-offsets and --line-offsets to pcregrep.
1871    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1872        infinitely looping recursion. The bug was that positive lookaheads were not
1873        being skipped when checking for a possible empty match (negative lookaheads
1874        and both kinds of lookbehind were skipped).
1876    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1877        inclusion of <windows.h> to before rather than after the definition of
1878        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1880    17. Specifying a possessive quantifier with a specific limit for a Unicode
1881        character property caused pcre_compile() to compile bad code, which led at
1882        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1883        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1884        caused the error; without that there was no problem.
1886    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1888    19. Added --enable-pcretest-libreadline.
1890    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1891        RE::GlobalReplace(). As a result, the number of replacements returned was
1892        double what it should be. I removed one of the increments, but Craig sent a
1893        later patch that removed the other one (the right fix) and added unit tests
1894        that check the return values (which was not done before).
1896    21. Several CMake things:
1898        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1899            the names libpcre and libpcreposix, not just pcre and pcreposix.
1901        (2) The above change means that pcretest and pcregrep are now correctly
1902            linked with the newly-built libraries, not previously installed ones.
1906    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1907        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1908        UTF-8 newline character). The key issue is that the pattern starts .*;
1909        this means that the match must be either at the beginning, or after a
1910        newline. The bug was in the code for advancing after a failed match and
1911        checking that the new position followed a newline. It was not taking
1912        account of UTF-8 characters correctly.
1914    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1915        character classes. PCRE was not treating the sequence [:...:] as a
1916        character class unless the ... were all letters. Perl, however, seems to
1917        allow any characters between [: and :], though of course it rejects as
1918        unknown any "names" that contain non-letters, because all the known class
1919        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1920        for example, whereas PCRE did not - it did not recognize a POSIX character
1921        class. This seemed a bit dangerous, so the code has been changed to be
1922        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1923        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1924        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1925        Perl does, and where it didn't before.
1927    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1928        Windows environments %n is disabled by default.
1931    Version 7.4 21-Sep-07
1932    ---------------------
1934    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1935        means that a class such as [\s] counted as "explicit reference to CR or
1936        LF". That isn't really right - the whole point of the change was to try to
1937        help when there was an actual mention of one of the two characters. So now
1938        the change happens only if \r or \n (or a literal CR or LF) character is
1939        encountered.
1941    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1942        of both had grown to the point where there were only 3 bits left.
1943        Fortunately, there was spare space in the data structure, and so I have
1944        moved the internal flags into a new 16-bit field to free up more option
1945        bits.
1947    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1948        but did not set the internal JCHANGED flag - either of these is enough to
1949        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1950        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1951        start sets both bits.
1953    4.  Added options (at build time, compile time, exec time) to change \R from
1954        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1956    5.  doc/pcresyntax.html was missing from the distribution.
1958    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1959        compatibility, even though it is no longer used.
1961    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1962        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1963        windows.h file is present (where different names are used). [This was
1964        reversed later after testing - see 16 below.]
1966    8.  Changed all #include <config.h> to #include "config.h". There were also
1967        some further <pcre.h> cases that I changed to "pcre.h".
1969    9.  When pcregrep was used with the --colour option, it missed the line ending
1970        sequence off the lines that it output.
1972    10. It was pointed out to me that arrays of string pointers cause lots of
1973        relocations when a shared library is dynamically loaded. A technique of
1974        using a single long string with a table of offsets can drastically reduce
1975        these. I have refactored PCRE in four places to do this. The result is
1976        dramatic:
1978          Originally:                          290
1979          After changing UCP table:            187
1980          After changing error message table:   43
1981          After changing table of "verbs"       36
1982          After changing table of Posix names   22
1984        Thanks to the folks working on Gregex for glib for this insight.
1986    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1987        unicode-properties was also set.
1989    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1991    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1992        checked only for CRLF.
1994    14. Added casts to pcretest.c to avoid compiler warnings.
1996    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1998    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1999        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2000        entirely. This removes changes made in 7 above.
2002    17. The CMake files have been updated, and there is now more information about
2003        building with CMake in the NON-UNIX-USE document.
2006    Version 7.3 28-Aug-07
2007    ---------------------
2009     1. In the rejigging of the build system that eventually resulted in 7.1, the
2010        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
2011        brackets there is not right, since it causes compilers to look for an
2012        installed pcre.h, not the version that is in the source that is being
2013        compiled (which of course may be different). I have changed it back to:
2015          #include "pcre.h"
2017        I have a vague recollection that the change was concerned with compiling in
2018        different directories, but in the new build system, that is taken care of
2019        by the VPATH setting the Makefile.
2021     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
2022        when the subject happened to end in the byte 0x85 (e.g. if the last
2023        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
2024        characters but of course it shouldn't be taken as a newline when it is part
2025        of another character. The bug was that, for an unlimited repeat of . in
2026        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
2027        characters when looking for a newline.
2029     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
2031     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
2032        in debug output.
2034     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
2035        long printing in the pcrecpp unittest when running under MinGW.
2037     6. ESC_K was left out of the EBCDIC table.
2039     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
2040        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
2041        limit also applies to "virtual nesting" when a pattern is recursive, and in
2042        this case 1000 isn't so big. I have been able to remove this limit at the
2043        expense of backing off one optimization in certain circumstances. Normally,
2044        when pcre_exec() would call its internal match() function recursively and
2045        immediately return the result unconditionally, it uses a "tail recursion"
2046        feature to save stack. However, when a subpattern that can match an empty
2047        string has an unlimited repetition quantifier, it no longer makes this
2048        optimization. That gives it a stack frame in which to save the data for
2049        checking that an empty string has been matched. Previously this was taken
2050        from the 1000-entry workspace that had been reserved. So now there is no
2051        explicit limit, but more stack is used.
2053     8. Applied Daniel's patches to solve problems with the import/export magic
2054        syntax that is required for Windows, and which was going wrong for the
2055        pcreposix and pcrecpp parts of the library. These were overlooked when this
2056        problem was solved for the main library.
2058     9. There were some crude static tests to avoid integer overflow when computing
2059        the size of patterns that contain repeated groups with explicit upper
2060        limits. As the maximum quantifier is 65535, the maximum group length was
2061        set at 30,000 so that the product of these two numbers did not overflow a
2062        32-bit integer. However, it turns out that people want to use groups that
2063        are longer than 30,000 bytes (though not repeat them that many times).
2064        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
2065        made it possible to implement the integer overflow checks in a much more
2066        dynamic way, which I have now done. The artificial limitation on group
2067        length has been removed - we now have only the limit on the total length of
2068        the compiled pattern, which depends on the LINK_SIZE setting.
2070    10. Fixed a bug in the documentation for get/copy named substring when
2071        duplicate names are permitted. If none of the named substrings are set, the
2072        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
2073        empty string.
2075    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
2076        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
2077        because the ] is interpreted as the first data character and the
2078        terminating ] is not found. PCRE has been made compatible with Perl in this
2079        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
2080        cause memory overwriting.
2082    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
2083        string has been matched (to stop an infinite loop). It was not recognizing
2084        a conditional subpattern that could match an empty string if that
2085        subpattern was within another subpattern. For example, it looped when
2086        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
2087        condition was not nested. This bug has been fixed.
2089    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
2090        past the start of the subject in the presence of bytes with the top bit
2091        set, for example "\x8aBCD".
2093    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
2094        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
2096    14. Optimized (?!) to (*FAIL).
2098    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
2099        This restricts code points to be within the range 0 to 0x10FFFF, excluding
2100        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
2101        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
2102        does: it's just the validity check that is more restrictive.
2104    16. Inserted checks for integer overflows during escape sequence (backslash)
2105        processing, and also fixed erroneous offset values for syntax errors during
2106        backslash processing.
2108    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
2109        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
2111    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
2112        caused an overrun.
2114    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
2115        something other than just ASCII characters) inside a group that had an
2116        unlimited repeat caused a loop at compile time (while checking to see
2117        whether the group could match an empty string).
2119    20. Debugging a pattern containing \p or \P could cause a crash. For example,
2120        [\P{Any}] did so. (Error in the code for printing property names.)
2122    21. An orphan \E inside a character class could cause a crash.
2124    22. A repeated capturing bracket such as (A)? could cause a wild memory
2125        reference during compilation.
2127    23. There are several functions in pcre_compile() that scan along a compiled
2128        expression for various reasons (e.g. to see if it's fixed length for look
2129        behind). There were bugs in these functions when a repeated \p or \P was
2130        present in the pattern. These operators have additional parameters compared
2131        with \d, etc, and these were not being taken into account when moving along
2132        the compiled data. Specifically:
2134        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2135            length.
2137        (b) An item such as \pL+ within a repeated group could cause crashes or
2138            loops.
2140        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2141            "reference to non-existent subpattern" error.
2143        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2145    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2146        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2148    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2150    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2151        character were causing crashes (broken optimization).
2153    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2154        \p or \P) caused a compile-time loop.
2156    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2157        break. For example, the unstudied pattern [\r\n]A does not match the string
2158        "\r\nA" because change 7.0/46 below moves the current point on by two
2159        characters after failing to match at the start. However, the pattern \nA
2160        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2161        the same is true. There doesn't seem any very clean way out of this, but
2162        what I have chosen to do makes the common cases work: PCRE now takes note
2163        of whether there can be an explicit match for \r or \n anywhere in the
2164        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2165        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2166        pattern has explicit CR or LF references.
2168    29. Added (*CR) etc for changing newline setting at start of pattern.
2171    Version 7.2 19-Jun-07
2172  ---------------------  ---------------------
2174   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 21  Version 7.2 01-May-07 Line 2188  Version 7.2 01-May-07
2188      stack recursion. This gives a massive performance boost under BSD, but just      stack recursion. This gives a massive performance boost under BSD, but just
2189      a small improvement under Linux. However, it saves one field in the frame      a small improvement under Linux. However, it saves one field in the frame
2190      in all cases.      in all cases.
2192   6. Added more features from the forthcoming Perl 5.10:   6. Added more features from the forthcoming Perl 5.10:
2194      (a) (?-n) (where n is a string of digits) is a relative subroutine or      (a) (?-n) (where n is a string of digits) is a relative subroutine or
2195          recursion call. It refers to the nth most recently opened parentheses.          recursion call. It refers to the nth most recently opened parentheses.
2197      (b) (?+n) is also a relative subroutine call; it refers to the nth next      (b) (?+n) is also a relative subroutine call; it refers to the nth next
2198          to be opened parentheses.          to be opened parentheses.
2200      (c) Conditions that refer to capturing parentheses can be specified      (c) Conditions that refer to capturing parentheses can be specified
2201          relatively, for example, (?(-2)... or (?(+3)...          relatively, for example, (?(-2)... or (?(+3)...
2203        (d) \K resets the start of the current match so that everything before
2204            is not part of it.
2206        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2208        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2209            reference syntax.
2211        (g) (?| introduces a group in which the numbering of parentheses in each
2212            alternative starts with the same number.
2214        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2216     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2219     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2220        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2221        for detecting groups that can match an empty string.
2223     9. A pattern with a very large number of alternatives (more than several
2224        hundred) was running out of internal workspace during the pre-compile
2225        phase, where pcre_compile() figures out how much memory will be needed. A
2226        bit of new cunning has reduced the workspace needed for groups with
2227        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2228        workspace instead of running out of the 4096 that are available.
2230    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2232    11. Applied patch from Google to remove an optimization that didn't quite work.
2233        The report of the bug said:
2235          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2236          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2237          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2239    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2240        it matched the wrong number of bytes.
2243  Version 7.1 24-Apr-07  Version 7.1 24-Apr-07
2244  ---------------------  ---------------------

Removed from v.167  
changed lines
  Added in v.1149

  ViewVC Help
Powered by ViewVC 1.1.5