ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 91 by nigel, Sat Feb 24 21:41:34 2007 UTC revision 1238 by ph10, Sat Jan 5 16:27:59 2013 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.33 xx-xxxx-201x
5    -------------------------
7    1.  Added 'U' to some constants that are compared to unsigned integers, to
8        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
9        variables that are added to signed variables, to ensure the result is
10        signed and can be negated.
12    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
13        CMake config files.
15    3.  Revise the creation of config.h.generic so that all boolean macros are
16        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
17        overriding via -D on the command line possible.
19    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
20        to unsigned int is reported to make a quite noticeable speed difference in
21        a specific Windows environment. Testing on Linux did also appear to show
22        some benefit (and it is clearly not harmful). Also fixed the definition of
23        Xop which should be unsigned.
25    5.  Related to (4), changing the definition of the intermediate variable cc
26        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
27        improvements.
30    Version 8.32 30-November-2012
31    -----------------------------
33    1.  Improved JIT compiler optimizations for first character search and single
34        character iterators.
36    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
37        Patch by Daniel Richard G.
39    3.  Single character iterator optimizations in the JIT compiler.
41    4.  Improved JIT compiler optimizations for character ranges.
43    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
44        Reported by Giuseppe D'Angelo.
46    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
47        of a line, was being set incorrectly in cases where .* appeared inside
48        atomic brackets at the start of a pattern, or where there was a subsequent
49        *PRUNE or *SKIP.
51    7.  Improved instruction cache flush for POWER/PowerPC.
52        Patch by Daniel Richard G.
54    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
55        grep:
57        (a) There is now no limit to the number of patterns to be matched.
59        (b) An error is given if a pattern is too long.
61        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
62            are now supported.
64        (d) --exclude-from and --include-from (multiple use) have been added.
66        (e) Exclusions and inclusions now apply to all files and directories, not
67            just to those obtained from scanning a directory recursively.
69        (f) Multiple uses of -f and --file-list are now supported.
71        (g) In a Windows environment, the default for -d has been changed from
72            "read" (the GNU grep default) to "skip", because otherwise the presence
73            of a directory in the file list provokes an error.
75        (h) The documentation has been revised and clarified in places.
77    9.  Improve the matching speed of capturing brackets.
79    10. Changed the meaning of \X so that it now matches a Unicode extended
80        grapheme cluster.
82    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
83        out POSIX threads when JIT support is configured.
85    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
87    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
88        structure could go wrong in environments where size_t is not the same size
89        as int.
91    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
92        set.
94    15. The EBCDIC support had decayed; later updates to the code had included
95        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
96        general tidy up of EBCDIC-related issues, and the documentation was also
97        not quite right. There is now a test that can be run on ASCII systems to
98        check some of the EBCDIC-related things (but is it not a full test).
100    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
101        in a small tidy to the code.
103    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
105    18. If the --only-matching (-o) option in pcregrep is specified multiple
106        times, each one causes appropriate output. For example, -o1 -o2 outputs the
107        substrings matched by the 1st and 2nd capturing parentheses. A separating
108        string can be specified by --om-separator (default empty).
110    19. Improving the first n character searches.
112    20. Turn case lists for horizontal and vertical white space into macros so that
113        they are defined only once.
115    21. This set of changes together give more compatible Unicode case-folding
116        behaviour for characters that have more than one other case when UCP
117        support is available.
119        (a) The Unicode property table now has offsets into a new table of sets of
120            three or more characters that are case-equivalent. The MultiStage2.py
121            script that generates these tables (the pcre_ucd.c file) now scans
122            CaseFolding.txt instead of UnicodeData.txt for character case
123            information.
125        (b) The code for adding characters or ranges of characters to a character
126            class has been abstracted into a generalized function that also handles
127            case-independence. In UTF-mode with UCP support, this uses the new data
128            to handle characters with more than one other case.
130        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
131            whose other case is greater than 256 are now correctly matched
132            caselessly. Previously, the high codepoint matched the low one, but not
133            vice versa.
135        (d) The processing of \h, \H, \v, and \ in character classes now makes use
136            of the new class addition function, using character lists defined as
137            macros alongside the case definitions of 20 above.
139        (e) Caseless back references now work with characters that have more than
140            one other case.
142        (f) General caseless matching of characters with more than one other case
143            is supported.
145    22. Unicode character properties were updated from Unicode 6.2.0
147    23. Improved CMake support under Windows. Patch by Daniel Richard G.
149    24. Add support for 32-bit character strings, and UTF-32
151    25. Major JIT compiler update (code refactoring and bugfixing).
152        Experimental Sparc 32 support is added.
154    26. Applied a modified version of Daniel Richard G's patch to create
155        pcre.h.generic and config.h.generic by "make" instead of in the
156        PrepareRelease script.
158    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
159        pcre_compile.c when checking for a zero character.
161    28. Introducing a native interface for JIT. Through this interface, the compiled
162        machine code can be directly executed. The purpose of this interface is to
163        provide fast pattern matching, so several sanity checks are not performed.
164        However, feature tests are still performed. The new interface provides
165        1.4x speedup compared to the old one.
167    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
168        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
169        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
171    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
172        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
173        the "old" RFC 2279). Instead, it ended up passing a negative length to
174        pcre_exec().
176    31. Add support for GCC's visibility feature to hide internal functions.
178    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
179        "unknown -C option" after outputting 0 or 1.
181    33. There is now support for generating a code coverage report for the test
182        suite in environments where gcc is the compiler and lcov is installed. This
183        is mainly for the benefit of the developers.
185    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
186        unaddressable using valgrind annotations, allowing valgrind to detect
187        invalid memory accesses. This is mainly for the benefit of the developers.
189    25. (*UTF) can now be used to start a pattern in any of the three libraries.
191    26. Give configure error if --enable-cpp but no C++ compiler found.
194    Version 8.31 06-July-2012
195    -------------------------
197    1.  Fixing a wrong JIT test case and some compiler warnings.
199    2.  Removed a bashism from the RunTest script.
201    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
202        to unsigned type, result still unsigned" that was given by an MS compiler
203        on encountering the code "-sizeof(xxx)".
205    4.  Partial matching support is added to the JIT compiler.
207    5.  Fixed several bugs concerned with partial matching of items that consist
208        of more than one character:
210        (a) /^(..)\1/ did not partially match "aba" because checking references was
211            done on an "all or nothing" basis. This also applied to repeated
212            references.
214        (b) \R did not give a hard partial match if \r was found at the end of the
215            subject.
217        (c) \X did not give a hard partial match after matching one or more
218            characters at the end of the subject.
220        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
221            a partial match for the string "\r".
223        (e) When newline was set to CRLF, the metacharacter "." did not recognize
224            a partial match for a CR character at the end of the subject string.
226    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
227        running pcretest, the text "(JIT)" added to the output whenever JIT is
228        actually used to run the match.
230    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
231        or /S+[+] with a digit between 1 and 7.
233    8.  OP_NOT now supports any UTF character not just single-byte ones.
235    9.  (*MARK) control verb is now supported by the JIT compiler.
237    10. The command "./RunTest list" lists the available tests without actually
238        running any of them. (Because I keep forgetting what they all are.)
242    12. Applied a (slightly modified) user-supplied patch that improves performance
243        when the heap is used for recursion (compiled with --disable-stack-for-
244        recursion). Instead of malloc and free for each heap frame each time a
245        logical recursion happens, frames are retained on a chain and re-used where
246        possible. This sometimes gives as much as 30% improvement.
248    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
249        call.
251    14. As documented, (*COMMIT) is now confined to within a positive assertion.
253    15. It is now possible to link pcretest with libedit as an alternative to
254        libreadline.
256    16. (*COMMIT) control verb is now supported by the JIT compiler.
258    17. The Unicode data tables have been updated to Unicode 6.1.0.
260    18. Added --file-list option to pcregrep.
262    19. Added binary file support to pcregrep, including the -a, --binary-files,
263        -I, and --text options.
265    20. The madvise function is renamed for posix_madvise for QNX compatibility
266        reasons. Fixed by Giuseppe D'Angelo.
268    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
270    22. Changed the option for creating symbolic links for 16-bit man pages from
271        -s to -sf so that re-installing does not cause issues.
273    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
275    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
276        matches in certain environments (the workspace was not being correctly
277        retained). Also added to pcre_dfa_exec() a simple plausibility check on
278        some of the workspace data at the beginning of a restart.
280    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
281        was not doing so when it should - probably a typo introduced by SVN 528
282        (change 8.10/14).
284    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
285        \w+ when the character tables indicated that \x{c4} was a word character.
286        There were several related cases, all because the tests for doing a table
287        lookup were testing for characters less than 127 instead of 255.
289    27. If a pattern contains capturing parentheses that are not used in a match,
290        their slots in the ovector are set to -1. For those that are higher than
291        any matched groups, this happens at the end of processing. In the case when
292        there were back references that the ovector was too small to contain
293        (causing temporary malloc'd memory to be used during matching), and the
294        highest capturing number was not used, memory off the end of the ovector
295        was incorrectly being set to -1. (It was using the size of the temporary
296        memory instead of the true size.)
298    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
299        ovector size, it uses memory at the end of the block that it has got.
301    29. Check for an overlong MARK name and give an error at compile time. The
302        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
304    30. JIT compiler update.
306    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
307        Rill for the patch.
309    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
311    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
313    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
314        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
315        was enabled.
317    35. Improve JIT code generation for greedy plus quantifier.
319    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
320        1 to "aa" instead of to an empty string. The bug affected repeated groups
321        that could potentially match an empty string.
323    37. Optimizing single character iterators in JIT.
325    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
326        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
327        codepoints that are too big for the mode are faulted, and in a UTF mode,
328        disallowed codepoints are also faulted.
330    39. If PCRE was compiled with UTF support, in three places in the DFA
331        matcher there was code that should only have been obeyed in UTF mode, but
332        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
333        processing when bytes with values greater than 127 were present. In 16-bit
334        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
335        both cases the values are those that cannot be the first data item in a UTF
336        character. The three items that might have provoked this were recursions,
337        possessively repeated groups, and atomic groups.
339    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
340        and pcregrep, because some OS require shared objects to be explicitly
341        passed to ld, causing the link step to fail if they are not.
343    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
344        mode, patterns that started with \h* or \R* might be incorrectly matched.
347    Version 8.30 04-February-2012
348    -----------------------------
350    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
351        name is defined in ctype.h.
353    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
354        only in quite long subpatterns.
356    3.  Removed the function pcre_info(), which has been obsolete and deprecated
357        since it was replaced by pcre_fullinfo() in February 2000.
359    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
360        match a (*MARK), and the match failed at the start of the subject, a
361        reference to memory before the start of the subject could occur. This bug
362        was introduced by fix 17 of release 8.21.
364    5.  A reference to an unset group with zero minimum repetition was giving
365        totally wrong answers (in non-JavaScript-compatibility mode). For example,
366        /(another)?(\1?)test/ matched against "hello world test". This bug was
367        introduced in release 8.13.
369    6.  Add support for 16-bit character strings (a large amount of work involving
370        many changes and refactorings).
372    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
373        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
374        from a file.
376    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
377        rounding is not applied in this particular case).
379    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
380        if they appear, or are escaped, in patterns.
382    10. Get rid of a number of -Wunused-but-set-variable warnings.
384    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
385        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
386        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
387        also returns the mark "x". This bug applied to capturing parentheses,
388        non-capturing parentheses, and atomic parentheses. It also applied to some
389        assertions.
391    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
392        information out of configure.ac instead of relying on pcre.h.generic, which
393        is not stored in the repository.
395    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
396        -lreadline.
398    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
400    15. Applied Graycode's patch to put the top-level frame on the stack rather
401        than the heap when not using the stack for recursion. This gives a
402        performance improvement in many cases when recursion is not deep.
404    16. Experimental code added to "pcretest -C" to output the stack frame size.
407    Version 8.21 12-Dec-2011
408    ------------------------
410    1.  Updating the JIT compiler.
412    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
413        are added as well.
415    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
416        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
417        calling _pcre_jit_exec. Some extra comments are added.
419    4.  (*MARK) settings inside atomic groups that do not contain any capturing
420        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
421        was introduced by change 18 for 8.20.
423    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
424        ECMA-262 standard.
426    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
427        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
428        This bug was probably introduced by change 9 of 8.13.
430    7.  While fixing 6 above, I noticed that a number of other items were being
431        incorrectly rejected as "not fixed length". This arose partly because newer
432        opcodes had not been added to the fixed-length checking code. I have (a)
433        corrected the bug and added tests for these items, and (b) arranged for an
434        error to occur if an unknown opcode is encountered while checking for fixed
435        length instead of just assuming "not fixed length". The items that were
436        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
437        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
438        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
440    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
441        being incorrectly compiled and would have given unpredicatble results.
443    9.  A possessively repeated subpattern with minimum repeat count greater than
444        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
445        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
446        the first (A) could occur when it should not.
448    10. Add a cast and remove a redundant test from the code.
450    11. JIT should use pcre_malloc/pcre_free for allocation.
452    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
453        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
454        is anything other than /usr, -L is still shown).
456    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
458    14. Perl does not support \N without a following name in a [] class; PCRE now
459        also gives an error.
461    15. If a forward reference was repeated with an upper limit of around 2000,
462        it caused the error "internal error: overran compiling workspace". The
463        maximum number of forward references (including repeats) was limited by the
464        internal workspace, and dependent on the LINK_SIZE. The code has been
465        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
466        the default depends on LINK_SIZE. There is a new upper limit (for safety)
467        of around 200,000 forward references. While doing this, I also speeded up
468        the filling in of repeated forward references.
470    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
471        incorrectly expecting the subject to contain another "a" after the start.
473    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
474        in the match, the SKIP should be ignored. This was not happening; instead
475        the SKIP was being treated as NOMATCH. For patterns such as
476        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
477        tested.
479    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
480        now much more compatible with Perl, in particular in cases where the result
481        is a non-match for a non-anchored pattern. For example, if
482        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
483        "m", where previously it did not return a name. A side effect of this
484        change is that for partial matches, the last encountered mark name is
485        returned, as for non matches. A number of tests that were previously not
486        Perl-compatible have been moved into the Perl-compatible test files. The
487        refactoring has had the pleasing side effect of removing one argument from
488        the match() function, thus reducing its stack requirements.
490    19. If the /S+ option was used in pcretest to study a pattern using JIT,
491        subsequent uses of /S (without +) incorrectly behaved like /S+.
493    21. Retrieve executable code size support for the JIT compiler and fixing
494        some warnings.
496    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
497        not work when the shorter character appeared right at the end of the
498        subject string.
500    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
501        systems.
503    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
504        output it when the /M option is used in pcretest.
506    25. The CheckMan script was not being included in the distribution. Also, added
507        an explicit "perl" to run Perl scripts from the PrepareRelease script
508        because this is reportedly needed in Windows.
510    26. If study data was being save in a file and studying had not found a set of
511        "starts with" bytes for the pattern, the data written to the file (though
512        never used) was taken from uninitialized memory and so caused valgrind to
513        complain.
515    27. Updated RunTest.bat as provided by Sheri Pierce.
517    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
519    29. Computation of memory usage for the table of capturing group names was
520        giving an unnecessarily large value.
523    Version 8.20 21-Oct-2011
524    ------------------------
526    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
527        a POSIX class. After further experiments with Perl, which convinced me that
528        Perl has bugs and confusions, a closing square bracket is no longer allowed
529        in a POSIX name. This bug also affected patterns with classes that started
530        with full stops.
532    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
533        captured substring, but while checking the failing first alternative,
534        substring 1 is temporarily captured. If the output vector supplied to
535        pcre_exec() was not big enough for this capture, the yield of the function
536        was still zero ("insufficient space for captured substrings"). This cannot
537        be totally fixed without adding another stack variable, which seems a lot
538        of expense for a edge case. However, I have improved the situation in cases
539        such as /(a)(b)x|abc/ matched against "abc", where the return code
540        indicates that fewer than the maximum number of slots in the ovector have
541        been set.
543    3.  Related to (2) above: when there are more back references in a pattern than
544        slots in the output vector, pcre_exec() uses temporary memory during
545        matching, and copies in the captures as far as possible afterwards. It was
546        using the entire output vector, but this conflicts with the specification
547        that only 2/3 is used for passing back captured substrings. Now it uses
548        only the first 2/3, for compatibility. This is, of course, another edge
549        case.
551    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
552        main code base, and can be used by building with --enable-jit. When this is
553        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
554        runtime --no-jit option is given.
556    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
557        ovector, the return from the function was zero, implying that there were
558        other matches that did not fit. The correct "exactly full" value is now
559        returned.
561    6.  If a subpattern that was called recursively or as a subroutine contained
562        (*PRUNE) or any other control that caused it to give a non-standard return,
563        invalid errors such as "Error -26 (nested recursion at the same subject
564        position)" or even infinite loops could occur.
566    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
567        computing the minimum length on reaching *ACCEPT, and so ended up with the
568        wrong value of 1 rather than 0. Further investigation indicates that
569        computing a minimum subject length in the presence of *ACCEPT is difficult
570        (think back references, subroutine calls), and so I have changed the code
571        so that no minimum is registered for a pattern that contains *ACCEPT.
573    8.  If (*THEN) was present in the first (true) branch of a conditional group,
574        it was not handled as intended. [But see 16 below.]
576    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
577        Sheri Pierce.
579    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
580        the first byte in a match must be "a".
582    11. Change 17 for 8.13 increased the recursion depth for patterns like
583        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
584        pattern contains any instances of (*THEN). If it does not, the old
585        optimizations are restored. It would be nice to do this on a per-group
586        basis, but at the moment that is not feasible.
588    12. In some environments, the output of pcretest -C is CRLF terminated. This
589        broke RunTest's code that checks for the link size. A single white space
590        character after the value is now allowed for.
592    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
593        For "fr", it uses the Windows-specific input and output files.
595    14. If (*THEN) appeared in a group that was called recursively or as a
596        subroutine, it did not work as intended. [But see next item.]
598    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
599        pattern fragments (but not containing any | characters). If A and B are
600        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
601        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
602        D. In other words, Perl considers parentheses that do not contain any |
603        characters to be part of a surrounding alternative, whereas PCRE was
604        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
605        differently. PCRE now behaves in the same way as Perl, except in the case
606        of subroutine/recursion calls such as (?1) which have in any case always
607        been different (but PCRE had them first :-).
609    16. Related to 15 above: Perl does not treat the | in a conditional group as
610        creating alternatives. Such a group is treated in the same way as an
611        ordinary group without any | characters when processing (*THEN). PCRE has
612        been changed to match Perl's behaviour.
614    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
615        RunGrepTest script failed.
617    18. Change 22 for version 13 caused atomic groups to use more stack. This is
618        inevitable for groups that contain captures, but it can lead to a lot of
619        stack use in large patterns. The old behaviour has been restored for atomic
620        groups that do not contain any capturing parentheses.
622    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
623        suppress the check for a minimum subject length at run time. (If it was
624        given to pcre_exec() or pcre_dfa_exec() it did work.)
626    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
627        fail to work when decoding hex characters in data strings in EBCDIC
628        environments.
630    21. It appears that in at least one Mac OS environment, the isxdigit() function
631        is implemented as a macro that evaluates to its argument more than once,
632        contravening the C 90 Standard (I haven't checked a later standard). There
633        was an instance in pcretest which caused it to go wrong when processing
634        \x{...} escapes in subject strings. The has been rewritten to avoid using
635        things like p++ in the argument of isxdigit().
638    Version 8.13 16-Aug-2011
639    ------------------------
641    1.  The Unicode data tables have been updated to Unicode 6.0.0.
643    2.  Two minor typos in pcre_internal.h have been fixed.
645    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
646        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
647        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
649    4.  There were a number of related bugs in the code for matching backrefences
650        caselessly in UTF-8 mode when codes for the characters concerned were
651        different numbers of bytes. For example, U+023A and U+2C65 are an upper
652        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
653        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
654        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
655        2-byte code at the end of the subject (it thought there wasn't enough data
656        left).
658    5.  Comprehensive information about what went wrong is now returned by
659        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
660        as the output vector has at least 2 elements. The offset of the start of
661        the failing character and a reason code are placed in the vector.
663    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
664        now returned is for the first byte of the failing character, instead of the
665        last byte inspected. This is an incompatible change, but I hope it is small
666        enough not to be a problem. It makes the returned offset consistent with
667        pcre_exec() and pcre_dfa_exec().
669    7.  pcretest now gives a text phrase as well as the error number when
670        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
671        failure, the offset and reason code are output.
673    8.  When \R was used with a maximizing quantifier it failed to skip backwards
674        over a \r\n pair if the subsequent match failed. Instead, it just skipped
675        back over a single character (\n). This seems wrong (because it treated the
676        two characters as a single entity when going forwards), conflicts with the
677        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
678        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
679        has been changed.
681    9.  Some internal refactoring has changed the processing so that the handling
682        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
683        time (the PCRE_DOTALL option was changed this way some time ago: version
684        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
685        which was always a bit of a fudge. It also means that there is one less
686        argument for the match() function, which reduces its stack requirements
687        slightly. This change also fixes an incompatibility with Perl: the pattern
688        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
690    10. More internal refactoring has drastically reduced the number of recursive
691        calls to match() for possessively repeated groups such as (abc)++ when
692        using pcre_exec().
694    11. While implementing 10, a number of bugs in the handling of groups were
695        discovered and fixed:
697        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
698        (a|)*(?1) gave a compile-time internal error.
699        ((a|)+)+  did not notice that the outer group could match an empty string.
700        (^a|^)+   was not marked as anchored.
701        (.*a|.*)+ was not marked as matching at start or after a newline.
703    12. Yet more internal refactoring has removed another argument from the match()
704        function. Special calls to this function are now indicated by setting a
705        value in a variable in the "match data" data block.
707    13. Be more explicit in pcre_study() instead of relying on "default" for
708        opcodes that mean there is no starting character; this means that when new
709        ones are added and accidentally left out of pcre_study(), testing should
710        pick them up.
712    14. The -s option of pcretest has been documented for ages as being an old
713        synonym of -m (show memory usage). I have changed it to mean "force study
714        for every regex", that is, assume /S for every regex. This is similar to -i
715        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
716        using it. It makes it easier to run collections of tests with and without
717        study enabled, and thereby test pcre_study() more easily. All the standard
718        tests are now run with and without -s (but some patterns can be marked as
719        "never study" - see 20 below).
721    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
722        restoration of the capturing data to the outer values was not happening
723        correctly.
725    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
726        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
727        pattern had matched an empty string, and so incorrectly returned a no
728        match.
730    17. There was optimizing code for the last branch of non-capturing parentheses,
731        and also for the obeyed branch of a conditional subexpression, which used
732        tail recursion to cut down on stack usage. Unfortunately, now that there is
733        the possibility of (*THEN) occurring in these branches, tail recursion is
734        no longer possible because the return has to be checked for (*THEN). These
735        two optimizations have therefore been removed. [But see 8.20/11 above.]
737    18. If a pattern containing \R was studied, it was assumed that \R always
738        matched two bytes, thus causing the minimum subject length to be
739        incorrectly computed because \R can also match just one byte.
741    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
742        was incorrectly computed.
744    20. If /S is present twice on a test pattern in pcretest input, it now
745        *disables* studying, thereby overriding the use of -s on the command line
746        (see 14 above). This is necessary for one or two tests to keep the output
747        identical in both cases.
749    21. When (*ACCEPT) was used in an assertion that matched an empty string and
750        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
752    22. When an atomic group that contained a capturing parenthesis was
753        successfully matched, but the branch in which it appeared failed, the
754        capturing was not being forgotten if a higher numbered group was later
755        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
756        group 1 to "a", when in fact it should be unset. This applied to multi-
757        branched capturing and non-capturing groups, repeated or not, and also to
758        positive assertions (capturing in negative assertions does not happen
759        in PCRE) and also to nested atomic groups.
761    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
762        subject after a captured substring, to make it easier to tell which of a
763        number of identical substrings has been captured.
765    24. The way atomic groups are processed by pcre_exec() has been changed so that
766        if they are repeated, backtracking one repetition now resets captured
767        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
768        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
769        "aaa". Previously, it would have been "a". As part of this code
770        refactoring, the way recursive calls are handled has also been changed.
772    25. If an assertion condition captured any substrings, they were not passed
773        back unless some other capturing happened later. For example, if
774        (?(?=(a))a) was matched against "a", no capturing was returned.
776    26. When studying a pattern that contained subroutine calls or assertions,
777        the code for finding the minimum length of a possible match was handling
778        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
779        group 1 called group 2 while simultaneously a separate group 2 called group
780        1). A stack overflow occurred in this case. I have fixed this by limiting
781        the recursion depth to 10.
783    27. Updated RunTest.bat in the distribution to the version supplied by Tom
784        Fortmann. This supports explicit test numbers on the command line, and has
785        argument validation and error reporting.
787    28. An instance of \X with an unlimited repeat could fail if at any point the
788        first character it looked at was a mark character.
790    29. Some minor code refactoring concerning Unicode properties and scripts
791        should reduce the stack requirement of match() slightly.
793    30. Added the '=' option to pcretest to check the setting of unused capturing
794        slots at the end of the pattern, which are documented as being -1, but are
795        not included in the return count.
797    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
798        compiled something random. Now it gives a compile-time error (as does
799        Perl).
801    32. A *MARK encountered during the processing of a positive assertion is now
802        recorded and passed back (compatible with Perl).
804    33. If --only-matching or --colour was set on a pcregrep call whose pattern
805        had alternative anchored branches, the search for a second match in a line
806        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
807        matched the line "0102" twice. The same bug affected patterns that started
808        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
809        twice.
811    34. Previously, PCRE did not allow quantification of assertions. However, Perl
812        does, and because of capturing effects, quantifying parenthesized
813        assertions may at times be useful. Quantifiers are now allowed for
814        parenthesized assertions.
816    35. A minor code tidy in pcre_compile() when checking options for \R usage.
818    36. \g was being checked for fancy things in a character class, when it should
819        just be a literal "g".
821    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
822        appearance of a nested POSIX class supersedes an apparent external class.
823        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
824        unescaped square brackets may also appear as part of class names. For
825        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
826        more like Perl. (But see 8.20/1 above.)
828    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
829        was because it thought it was \N{name}, which is not supported).
831    39. Add minix to OS list not supporting the -S option in pcretest.
833    40. PCRE tries to detect cases of infinite recursion at compile time, but it
834        cannot analyze patterns in sufficient detail to catch mutual recursions
835        such as ((?1))((?2)). There is now a runtime test that gives an error if a
836        subgroup is called recursively as a subpattern for a second time at the
837        same position in the subject string. In previous releases this might have
838        been caught by the recursion limit, or it might have run out of stack.
840    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
841        happen only once. PCRE was, however incorrectly giving a compile time error
842        "recursive call could loop indefinitely" because it cannot analyze the
843        pattern in sufficient detail. The compile time test no longer happens when
844        PCRE is compiling a conditional subpattern, but actual runaway loops are
845        now caught at runtime (see 40 above).
847    42. It seems that Perl allows any characters other than a closing parenthesis
848        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
849        has been changed to be the same.
851    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
852        as not to get warnings when autogen.sh is called. Also changed
853        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
855    44. To help people who use pcregrep to scan files containing exceedingly long
856        lines, the following changes have been made:
858        (a) The default value of the buffer size parameter has been increased from
859            8K to 20K. (The actual buffer used is three times this size.)
861        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
862            PCRE is built.
864        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
865            to be set at run time.
867        (d) Numerical values in pcregrep options can be followed by K or M, for
868            example --buffer-size=50K.
870        (e) If a line being scanned overflows pcregrep's buffer, an error is now
871            given and the return code is set to 2.
873    45. Add a pointer to the latest mark to the callout data block.
875    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
876        partial match of an empty string instead of no match. This was specific to
877        the use of ".".
879    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
880        complete match instead of a partial match. This bug was dependent on both
881        the PCRE_UTF8 and PCRE_DOTALL options being set.
883    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
884        starting byte set, because \b was not being ignored.
887    Version 8.12 15-Jan-2011
888    ------------------------
890    1.  Fixed some typos in the markup of the man pages, and wrote a script that
891        checks for such things as part of the documentation building process.
893    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
894        --match-limit and --recursion-limit options (added for 8.11). In
895        particular, this made one of the standard tests fail. (The integer value
896        went into the wrong half of a long int.)
898    3.  If the --colour option was given to pcregrep with -v (invert match), it
899        did strange things, either producing crazy output, or crashing. It should,
900        of course, ignore a request for colour when reporting lines that do not
901        match.
903    4.  Another pcregrep bug caused similar problems if --colour was specified with
904        -M (multiline) and the pattern match finished with a line ending.
906    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
907        matched in multiline mode, the following line was shown as part of the
908        match. This seems wrong, so I have changed it.
910    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
911        the check for further matches in the same line (so they could be coloured)
912        to overrun the end of the current line. If another match was found, it was
913        incorrectly shown (and then shown again when found in the next line).
915    7.  If pcregrep was compiled under Windows, there was a reference to the
916        function pcregrep_exit() before it was defined. I am assuming this was
917        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
918        reported by a user. I've moved the definition above the reference.
921    Version 8.11 10-Dec-2010
922    ------------------------
924    1.  (*THEN) was not working properly if there were untried alternatives prior
925        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
926        backtracked to try for "b" instead of moving to the next alternative branch
927        at the same level (in this case, to look for "c"). The Perl documentation
928        is clear that when (*THEN) is backtracked onto, it goes to the "next
929        alternative in the innermost enclosing group".
931    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
932        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
933        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
934        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
935        (*THEN).
937    3.  If \s appeared in a character class, it removed the VT character from
938        the class, even if it had been included by some previous item, for example
939        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
940        of \s, but is part of the POSIX "space" class.)
942    4.  A partial match never returns an empty string (because you can always
943        match an empty string at the end of the subject); however the checking for
944        an empty string was starting at the "start of match" point. This has been
945        changed to the "earliest inspected character" point, because the returned
946        data for a partial match starts at this character. This means that, for
947        example, /(?<=abc)def/ gives a partial match for the subject "abc"
948        (previously it gave "no match").
950    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
951        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
952        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
953        has an implication that the given string is incomplete (because a partial
954        match is preferred over a full match). For this reason, these items now
955        give a partial match in this situation. [Aside: previously, the one case
956        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
957        match rather than a full match, which was wrong by the old rules, but is
958        now correct.]
960    6.  There was a bug in the handling of #-introduced comments, recognized when
961        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
962        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
963        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
964        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
965        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
966        places in pcre_compile().
968    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
969        comments when looking ahead for named forward references to subpatterns,
970        the only newline sequence it recognized was NL. It now handles newlines
971        according to the set newline convention.
973    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
974        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
975        cater for a lack of strerror(). These oversights have been fixed.
977    9.  Added --match-limit and --recursion-limit to pcregrep.
979    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
981    11. When the -o option was used, pcregrep was setting a return code of 1, even
982        when matches were found, and --line-buffered was not being honoured.
984    12. Added an optional parentheses number to the -o and --only-matching options
985        of pcregrep.
987    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
988        can match an empty string. The code to do it in pcretest and pcredemo
989        needed fixing:
991        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
992            only one byte after an empty string match just before CRLF (this case
993            just got forgotten; "any" and "anycrlf" were OK).
995        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
996            mode when an empty string match preceded an ASCII character followed by
997            a non-ASCII character. (The code for advancing by one character rather
998            than one byte was nonsense.)
1000        (c) The pcredemo.c sample program did not have any code at all to handle
1001            the cases when CRLF is a valid newline sequence.
1003    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1004        as a starting offset was within the subject string. There is now a new
1005        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1006        negative or greater than the length of the string. In order to test this,
1007        pcretest is extended to allow the setting of negative starting offsets.
1009    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1010        starting offset points to the beginning of a UTF-8 character was
1011        unnecessarily clumsy. I tidied it up.
1013    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1014        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1016    17. Nobody had reported that the --include_dir option, which was added in
1017        release 7.7 should have been called --include-dir (hyphen, not underscore)
1018        for compatibility with GNU grep. I have changed it to --include-dir, but
1019        left --include_dir as an undocumented synonym, and the same for
1020        --exclude-dir, though that is not available in GNU grep, at least as of
1021        release 2.5.4.
1023    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1024        characters from a string of bytes) have been redefined so as not to use
1025        loops, in order to improve performance in some environments. At the same
1026        time, I abstracted some of the common code into auxiliary macros to save
1027        repetition (this should not affect the compiled code).
1029    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1030        compile-time error is now given if \c is not followed by an ASCII
1031        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1032        different, and any byte value is allowed.)
1034    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1035        START_OPTIMIZE option, which is now allowed at compile time - but just
1036        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1037        to pcregrep and other applications that have no direct access to PCRE
1038        options. The new /Y option in pcretest sets this option when calling
1039        pcre_compile().
1041    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1042        back references. Groups containing recursive back references were forced to
1043        be atomic by that change, but in the case of named groups, the amount of
1044        memory required was incorrectly computed, leading to "Failed: internal
1045        error: code overflow". This has been fixed.
1047    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1048        pcretest.c, to avoid build problems in some Borland environments.
1051    Version 8.10 25-Jun-2010
1052    ------------------------
1054    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1055        THEN.
1057    2.  (*ACCEPT) was not working when inside an atomic group.
1059    3.  Inside a character class, \B is treated as a literal by default, but
1060        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1061        causes the error). The code is unchanged, but I tidied the documentation.
1063    4.  Inside a character class, PCRE always treated \R and \X as literals,
1064        whereas Perl faults them if its -w option is set. I have changed PCRE so
1065        that it faults them when PCRE_EXTRA is set.
1067    5.  Added support for \N, which always matches any character other than
1068        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1070    6.  When compiling pcregrep with newer versions of gcc which may have
1071        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1072        declared with attribute warn_unused_result" were given. Just casting the
1073        result to (void) does not stop the warnings; a more elaborate fudge is
1074        needed. I've used a macro to implement this.
1076    7.  Minor change to pcretest.c to avoid a compiler warning.
1078    8.  Added four artifical Unicode properties to help with an option to make
1079        \s etc use properties (see next item). The new properties are: Xan
1080        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1082    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1083        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1084        this option. Modified pcretest to add /W to test this facility. Added
1085        REG_UCP to make it available via the POSIX interface.
1087    10. Added --line-buffered to pcregrep.
1089    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1090        studied, and the match started with a letter with a code point greater than
1091        127 whose first byte was different to the first byte of the other case of
1092        the letter, the other case of this starting letter was not recognized
1093        (#976).
1095    12. If a pattern that was studied started with a repeated Unicode property
1096        test, for example, \p{Nd}+, there was the theoretical possibility of
1097        setting up an incorrect bitmap of starting bytes, but fortunately it could
1098        not have actually happened in practice until change 8 above was made (it
1099        added property types that matched character-matching opcodes).
1101    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1102        possible starting bytes for non-anchored patterns.
1104    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1105        \R, and also a number of cases that involve Unicode properties, both
1106        explicit and implicit when PCRE_UCP is set.
1108    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1109        input, it could crash or give wrong results if characters with values
1110        greater than 0xc0 were present in the subject string. (Detail: it assumed
1111        UTF-8 input when processing these items.)
1113    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1114        size_t is 64-bit (#991).
1116    17. Added a check for running out of memory when PCRE is compiled with
1117        --disable-stack-for-recursion (#990).
1119    18. If the last data line in a file for pcretest does not have a newline on
1120        the end, a newline was missing in the output.
1122    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1123        less than 128) in its various bitmaps. However, there is a facility for
1124        generating tables according to the current locale when PCRE is compiled. It
1125        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1126        characters, are recognized by isspace() and therefore were getting set in
1127        these tables, and indeed these tables seem to approximate to ISO 8859. This
1128        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1129        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1130        which of course cannot start UTF-8 characters. I have changed the code so
1131        that only real ASCII characters (less than 128) and the correct starting
1132        bytes for UTF-8 encodings are set for characters greater than 127 when in
1133        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1134        altogether.)
1136    20. Added the /T option to pcretest so as to be able to run tests with non-
1137        standard character tables, thus making it possible to include the tests
1138        used for 19 above in the standard set of tests.
1140    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1141        reference to a subpattern the other side of a comment that contains an
1142        opening parenthesis caused either an internal compiling error, or a
1143        reference to the wrong subpattern.
1146    Version 8.02 19-Mar-2010
1147    ------------------------
1149    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1151    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1152        configured.
1154    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1155        original author of that file, following a query about its status.
1157    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1158        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1160    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1161        quantifier applied to a forward-referencing subroutine call, could compile
1162        incorrect code or give the error "internal error: previously-checked
1163        referenced subpattern not found".
1165    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1166        variables to point to external functions. For these systems, therefore,
1167        pcre_malloc etc. are now initialized to local functions that call the
1168        relevant global functions.
1170    7.  There were two entries missing in the vectors called coptable and poptable
1171        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1172        I've fixed the data, and added a kludgy way of testing at compile time that
1173        the lengths are correct (equal to the number of opcodes).
1175    8.  Following on from 7, I added a similar kludge to check the length of the
1176        eint vector in pcreposix.c.
1178    9.  Error texts for pcre_compile() are held as one long string to avoid too
1179        much relocation at load time. To find a text, the string is searched,
1180        counting zeros. There was no check for running off the end of the string,
1181        which could happen if a new error number was added without updating the
1182        string.
1184    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1186    11. \K was not working if it appeared in an atomic group or in a group that
1187        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1188        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1189        the assertion is positive, but not if it is negative.
1191    12. Change 11 fortuitously reduced the size of the stack frame used in the
1192        "match()" function of pcre_exec.c by one pointer. Forthcoming
1193        implementation of support for (*MARK) will need an extra pointer on the
1194        stack; I have reserved it now, so that the stack frame size does not
1195        decrease.
1197    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1198        item in branch that calls a recursion is a subroutine call - as in the
1199        second branch in the above example - was incorrectly given the compile-
1200        time error "recursive call could loop indefinitely" because pcre_compile()
1201        was not correctly checking the subroutine for matching a non-empty string.
1203    14. The checks for overrunning compiling workspace could trigger after an
1204        overrun had occurred. This is a "should never occur" error, but it can be
1205        triggered by pathological patterns such as hundreds of nested parentheses.
1206        The checks now trigger 100 bytes before the end of the workspace.
1208    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1211    Version 8.01 19-Jan-2010
1212    ------------------------
1214    1.  If a pattern contained a conditional subpattern with only one branch (in
1215        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1216        computed the wrong minimum data length (which is of course zero for such
1217        subpatterns). This could cause incorrect "no match" results.
1219    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1220        the pattern is reset in the first branch, pcre_compile() failed with
1221        "internal error: code overflow at offset...". This happened only when
1222        the reset was to the original external option setting. (An optimization
1223        abstracts leading options settings into an external setting, which was the
1224        cause of this.)
1226    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1227        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1228        assertion pattern did not match (meaning that the assertion was true), it
1229        was incorrectly treated as false if the SKIP had been reached during the
1230        matching. This also applied to assertions used as conditions.
1232    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1233        assertion subpattern, including such a pattern used as a condition,
1234        unpredictable results occurred, instead of the error return
1237    5.  The C++ GlobalReplace function was not working like Perl for the special
1238        situation when an empty string is matched. It now does the fancy magic
1239        stuff that is necessary.
1241    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1242        removed. (These were left over from very, very early versions of PCRE.)
1244    7.  Some cosmetic changes to the code to make life easier when compiling it
1245        as part of something else:
1247        (a) Change DEBUG to PCRE_DEBUG.
1249        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1250            called "current" as "current_branch", to prevent a collision with the
1251            Linux macro when compiled as a kernel module.
1253        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1254            prevent a collision with the Linux macro when compiled as a kernel
1255            module.
1257    8.  In pcre_compile() there are some checks for integer overflows that used to
1258        cast potentially large values to (double). This has been changed to that
1259        when building, a check for int64_t is made, and if it is found, it is used
1260        instead, thus avoiding the use of floating point arithmetic. (There is no
1261        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1262        double.
1264    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1265        2005 (difference between two addresses compared to an unsigned value).
1267    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1268        custom one, because of the following reported problem in Windows:
1270          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1271              under Win32.
1272          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1273              therefore missing the function definition.
1274          - The compiler thus generates a "C" signature for the test function.
1275          - The linker fails to find the "C" function.
1276          - PCRE fails to configure if asked to do so against libbz2.
1278    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1279        messages were output:
1281          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1282          rerunning libtoolize, to keep the correct libtool macros in-tree.
1283          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1285        I have done both of these things.
1287    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1288        most of the time, it *can* run out if it is given a pattern that contains a
1289        runaway infinite recursion. I updated the discussion in the pcrestack man
1290        page.
1292    13. Now that we have gone to the x.xx style of version numbers, the minor
1293        version may start with zero. Using 08 or 09 is a bad idea because users
1294        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1295        interpreted as invalid octal numbers. I've updated the previous comment in
1296        configure.ac, and also added a check that gives an error if 08 or 09 are
1297        used.
1299    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1300        causing partial matching to fail when the end of the subject matched \W
1301        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1303    15. There were some discrepancies between the declarations in pcre_internal.h
1304        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1305        their definitions. The declarations used "const uschar *" and the
1306        definitions used USPTR. Even though USPTR is normally defined as "const
1307        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1308        reported that: "This difference in casting confuses some C++ compilers, for
1309        example, SunCC recognizes above declarations as different functions and
1310        generates broken code for hbpcre." I have changed the declarations to use
1311        USPTR.
1313    16. GNU libtool is named differently on some systems. The autogen.sh script now
1314        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1315        (FreeBSD).
1317    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1318        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1319        comment: "Figure out how to create a longlong from a string: strtoll and
1320        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1321        instance, but it only takes 2 args instead of 3!"
1323    18. A subtle bug concerned with back references has been fixed by a change of
1324        specification, with a corresponding code fix. A pattern such as
1325        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1326        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1327        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1328        same bug. Such groups have to be quantified to be useful, or contained
1329        inside another quantified group. (If there's no repetition, the reference
1330        can never match.) The problem arises because, having left the group and
1331        moved on to the rest of the pattern, a later failure that backtracks into
1332        the group uses the captured value from the final iteration of the group
1333        rather than the correct earlier one. I have fixed this in PCRE by forcing
1334        any group that contains a reference to itself to be an atomic group; that
1335        is, there cannot be any backtracking into it once it has completed. This is
1336        similar to recursive and subroutine calls.
1339    Version 8.00 19-Oct-09
1340    ----------------------
1342    1.  The table for translating pcre_compile() error codes into POSIX error codes
1343        was out-of-date, and there was no check on the pcre_compile() error code
1344        being within the table. This could lead to an OK return being given in
1345        error.
1347    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1348        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1349        in a Windows environment.
1351    3.  The pcregrep --count option prints the count for each file even when it is
1352        zero, as does GNU grep. However, pcregrep was also printing all files when
1353        --files-with-matches was added. Now, when both options are given, it prints
1354        counts only for those files that have at least one match. (GNU grep just
1355        prints the file name in this circumstance, but including the count seems
1356        more useful - otherwise, why use --count?) Also ensured that the
1357        combination -clh just lists non-zero counts, with no names.
1359    4.  The long form of the pcregrep -F option was incorrectly implemented as
1360        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1361        but it seems right to fix it, and I didn't think it was worth preserving
1362        the old behaviour.
1364    5.  The command line items --regex=pattern and --regexp=pattern were not
1365        recognized by pcregrep, which required --regex pattern or --regexp pattern
1366        (with a space rather than an '='). The man page documented the '=' forms,
1367        which are compatible with GNU grep; these now work.
1369    6.  No libpcreposix.pc file was created for pkg-config; there was just
1370        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1372    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1373        when UCP support is not needed, by modifying the Python script that
1374        generates it from Unicode data files. This should not matter if the module
1375        is correctly used as a library, but I received one complaint about 50K of
1376        unwanted data. My guess is that the person linked everything into his
1377        program rather than using a library. Anyway, it does no harm.
1379    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1380        was a minimum greater than 1 for a wide character in a possessive
1381        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1382        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1383        character. Chaos in the form of incorrect output or a compiling loop could
1384        result.
1386    9.  The restrictions on what a pattern can contain when partial matching is
1387        requested for pcre_exec() have been removed. All patterns can now be
1388        partially matched by this function. In addition, if there are at least two
1389        slots in the offset vector, the offset of the earliest inspected character
1390        for the match and the offset of the end of the subject are set in them when
1391        PCRE_ERROR_PARTIAL is returned.
1393    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1394        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1395        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1396        and may be more useful for multi-segment matching.
1398    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1399        used to be given if ever the end of the subject was reached; now it is
1400        given only if matching could not proceed because another character was
1401        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1402        string "Z", which now yields "no match" instead of "partial match". In the
1403        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1404        final character ended with (*FAIL).
1406    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1407        if the pattern had a "must contain" character that was already found in the
1408        earlier partial match, unless partial matching was again requested. For
1409        example, with the pattern /dog.(body)?/, the "must contain" character is
1410        "g". If the first part-match was for the string "dog", restarting with
1411        "sbody" failed. This bug has been fixed.
1413    13. The string returned by pcre_dfa_exec() after a partial match has been
1414        changed so that it starts at the first inspected character rather than the
1415        first character of the match. This makes a difference only if the pattern
1416        starts with a lookbehind assertion or \b or \B (\K is not supported by
1417        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1418        matching functions compatible, and I think it's the right thing to do.
1420    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1421        so that the demonstration program is easily available in environments where
1422        PCRE has not been installed from source.
1424    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1425        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1426        library.
1428    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1429        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1430        is not the first non-POSIX option to be added. Clearly some people find
1431        these options useful.
1433    17. If a caller to the POSIX matching function regexec() passes a non-zero
1434        value for nmatch with a NULL value for pmatch, the value of
1435        nmatch is forced to zero.
1437    18. RunGrepTest did not have a test for the availability of the -u option of
1438        the diff command, as RunTest does. It now checks in the same way as
1439        RunTest, and also checks for the -b option.
1441    19. If an odd number of negated classes containing just a single character
1442        interposed, within parentheses, between a forward reference to a named
1443        subpattern and the definition of the subpattern, compilation crashed with
1444        an internal error, complaining that it could not find the referenced
1445        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1446        [The bug was that it was starting one character too far in when skipping
1447        over the character class, thus treating the ] as data rather than
1448        terminating the class. This meant it could skip too much.]
1450    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1451        /g option in pcretest when the pattern contains \K, which makes it possible
1452        to have an empty string match not at the start, even when the pattern is
1453        anchored. Updated pcretest and pcredemo to use this option.
1455    21. If the maximum number of capturing subpatterns in a recursion was greater
1456        than the maximum at the outer level, the higher number was returned, but
1457        with unset values at the outer level. The correct (outer level) value is
1458        now given.
1460    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1461        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1462        make it do so. The string so far is captured, making this feature
1463        compatible with Perl.
1465    23. The tests have been re-organized, adding tests 11 and 12, to make it
1466        possible to check the Perl 5.10 features against Perl 5.10.
1468    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1469        pattern matches a fixed length string. PCRE did not allow this; now it
1470        does. Neither allows recursion.
1472    25. I finally figured out how to implement a request to provide the minimum
1473        length of subject string that was needed in order to match a given pattern.
1474        (It was back references and recursion that I had previously got hung up
1475        on.) This code has now been added to pcre_study(); it finds a lower bound
1476        to the length of subject needed. It is not necessarily the greatest lower
1477        bound, but using it to avoid searching strings that are too short does give
1478        some useful speed-ups. The value is available to calling programs via
1479        pcre_fullinfo().
1481    26. While implementing 25, I discovered to my embarrassment that pcretest had
1482        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1483        study optimizations had never been tested with that matching function.
1484        Oops. What is worse, even when it was passed study data, there was a bug in
1485        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1486        were also very few tests of studied patterns with pcre_dfa_exec().
1488    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1489        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1490        on the other side of the coin, they are no longer allowed to have different
1491        names, because these cannot be distinguished in PCRE, and this has caused
1492        confusion. (This is a difference from Perl.)
1494    28. When duplicate subpattern names are present (necessarily with different
1495        numbers, as required by 27 above), and a test is made by name in a
1496        conditional pattern, either for a subpattern having been matched, or for
1497        recursion in such a pattern, all the associated numbered subpatterns are
1498        tested, and the overall condition is true if the condition is true for any
1499        one of them. This is the way Perl works, and is also more like the way
1500        testing by number works.
1503    Version 7.9 11-Apr-09
1504    ---------------------
1506    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1507        (pcretest), all targets were linked against these libraries. This included
1508        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1509        libraries. This caused unwanted dependencies to be created. This problem
1510        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1511        pcretest is linked with readline.
1513    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1514        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1515        moved outside it again, because FALSE and TRUE are already defined in AIX,
1516        but BOOL is not.
1518    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1519        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1521    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1522        hyphens) following file names and line numbers when outputting matching
1523        lines. This is not true; no spaces are inserted. I have also clarified the
1524        wording for the --colour (or --color) option.
1526    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1527        was not coloured; this is different to GNU grep, so I have changed it to be
1528        the same.
1530    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1531        each matching line was coloured. Now it goes on to look for further matches
1532        of any of the test patterns, which is the same behaviour as GNU grep.
1534    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1535        doesn't make sense to accept an empty string match in pcregrep, so I have
1536        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1537        seems to be how GNU grep behaves.
1539    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1540        start or after a newline", because the conditional assertion was not being
1541        correctly handled. The rule now is that both the assertion and what follows
1542        in the first alternative must satisfy the test.
1544    9.  If auto-callout was enabled in a pattern with a conditional group whose
1545        condition was an assertion, PCRE could crash during matching, both with
1546        pcre_exec() and pcre_dfa_exec().
1548    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1549        used for matching.
1551    11. Unicode property support in character classes was not working for
1552        characters (bytes) greater than 127 when not in UTF-8 mode.
1554    12. Added the -M command line option to pcretest.
1556    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1558    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1560    16. Added comments and documentation about mis-use of no_arg in the C++
1561        wrapper.
1563    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1564        from Martin Jerabek that uses macro names for all relevant character and
1565        string constants.
1567    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1568        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1569        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1570        these, but not everybody uses configure.
1572    19. A conditional group that had only one branch was not being correctly
1573        recognized as an item that could match an empty string. This meant that an
1574        enclosing group might also not be so recognized, causing infinite looping
1575        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1576        with the subject "ab", where knowledge that the repeated group can match
1577        nothing is needed in order to break the loop.
1579    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1580        exec(), but without supplying a callout function, matching went wrong.
1582    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1583        leak if the size of the offset vector was greater than 30. When the vector
1584        is smaller, the saved offsets during recursion go onto a local stack
1585        vector, but for larger vectors malloc() is used. It was failing to free
1586        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1587        error, in fact).
1589    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1590        heapframe that is used only when UTF-8 support is enabled. This caused no
1591        problem, but was untidy.
1593    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1594        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1595        included within another project.
1597    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1598        slightly modified by me:
1600          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1601              not building pcregrep.
1603          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1604              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1606    25. Forward references, both numeric and by name, in patterns that made use of
1607        duplicate group numbers, could behave incorrectly or give incorrect errors,
1608        because when scanning forward to find the reference group, PCRE was not
1609        taking into account the duplicate group numbers. A pattern such as
1610        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1612    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1613        the feature of a custom pointer more persuasive (as requested by a user).
1615    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1616        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1617        pre-defined.
1619    28. Added support for (*UTF8) at the start of a pattern.
1621    29. Arrange for flags added by the "release type" setting in CMake to be shown
1622        in the configuration summary.
1625    Version 7.8 05-Sep-08
1626    ---------------------
1628    1.  Replaced UCP searching code with optimized version as implemented for Ad
1629        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1630        stage table and inline lookup instead of a function, giving speed ups of 2
1631        to 5 times on some simple patterns that I tested. Permission was given to
1632        distribute the MultiStage2.py script that generates the tables (it's not in
1633        the tarball, but is in the Subversion repository).
1635    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1636        scripts.
1638    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1639        a group with a zero qualifier. The result of the study could be incorrect,
1640        or the function might crash, depending on the pattern.
1642    4.  Caseless matching was not working for non-ASCII characters in back
1643        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1644        It now works when Unicode Property Support is available.
1646    5.  In pcretest, an escape such as \x{de} in the data was always generating
1647        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1648        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1649        truncation.
1651    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1653    7.  Added two (int) casts to pcregrep when printing the difference of two
1654        pointers, in case they are 64-bit values.
1656    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1657        test 2 if it fails.
1659    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1660        and a #define of that name to empty if it is not externally set. This is to
1661        allow users of MSVC to set it if necessary.
1663    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1664        the convenience functions in the pcre_get.c source file.
1666    11. An option change at the start of a pattern that had top-level alternatives
1667        could cause overwriting and/or a crash. This command provoked a crash in
1668        some environments:
1670          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1672        This potential security problem was recorded as CVE-2008-2371.
1674    12. For a pattern where the match had to start at the beginning or immediately
1675        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1676        pcre_dfa_exec() could read past the end of the passed subject if there was
1677        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1678        pcretest so that it places the subject at the end of its malloc-ed buffer.
1680    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1681        exec() might read past the end of the data buffer in UTF-8 mode.
1683    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1684        the data contained the byte 0x85 as part of a UTF-8 character within its
1685        first line. This applied both to normal and DFA matching.
1687    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1688        /^[^d]*?$/8 failed to match "abc".
1690    16. Added a missing copyright notice to pcrecpp_internal.h.
1692    17. Make it more clear in the documentation that values returned from
1693        pcre_exec() in ovector are byte offsets, not character counts.
1695    18. Tidied a few places to stop certain compilers from issuing warnings.
1697    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1698        supplied by Stefan Weber. I made a further small update for 7.8 because
1699        there is a change of source arrangements: the pcre_searchfuncs.c module is
1700        replaced by pcre_ucd.c.
1703    Version 7.7 07-May-08
1704    ---------------------
1706    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1707        a string to a long long, pretend we don't even have a long long." This is
1708        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1710    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1711        pre-7.6 versions, which defined a global no_arg variable instead of putting
1712        it in the RE class. (See also #8 below.)
1714    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1715        Lopes.
1717    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1719        (1) The include/exclude patterns were being applied to the whole pathnames
1720            of files, instead of just to the final components.
1722        (2) If there was more than one level of directory, the subdirectories were
1723            skipped unless they satisfied the include/exclude conditions. This is
1724            inconsistent with GNU grep (and could even be seen as contrary to the
1725            pcregrep specification - which I improved to make it absolutely clear).
1726            The action now is always to scan all levels of directory, and just
1727            apply the include/exclude patterns to regular files.
1729    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1730        --exclude_dir in the tests to avoid scanning .svn directories.
1732    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1733        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1734        doesn't support NULs in patterns.
1736    7.  Added some missing "const"s to declarations of static tables in
1737        pcre_compile.c and pcre_dfa_exec.c.
1739    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1740        caused by fix #2  above. (Subsequently also a second patch to fix the
1741        first patch. And a third patch - this was a messy problem.)
1743    9.  Applied Craig's patch to remove the use of push_back().
1745    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1746        matching function regexec().
1748    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1749        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1750        references. PCRE supports relative numbers with this syntax (I don't think
1751        Oniguruma does).
1753    12. Previously, a group with a zero repeat such as (...){0} was completely
1754        omitted from the compiled regex. However, this means that if the group
1755        was called as a subroutine from elsewhere in the pattern, things went wrong
1756        (an internal error was given). Such groups are now left in the compiled
1757        pattern, with a new opcode that causes them to be skipped at execution
1758        time.
1760    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1761        to the way PCRE behaves:
1763        (a) A lone ] character is dis-allowed (Perl treats it as data).
1765        (b) A back reference to an unmatched subpattern matches an empty string
1766            (Perl fails the current match path).
1768        (c) A data ] in a character class must be notated as \] because if the
1769            first data character in a class is ], it defines an empty class. (In
1770            Perl it is not possible to have an empty class.) The empty class []
1771            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1772            The negative empty class [^] matches any one character, independently
1773            of the DOTALL setting.
1775    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1776        non-existent subpattern following a character class starting with ']' and
1777        containing () gave an internal compiling error instead of "reference to
1778        non-existent subpattern". Fortunately, when the pattern did exist, the
1779        compiled code was correct. (When scanning forwards to check for the
1780        existencd of the subpattern, it was treating the data ']' as terminating
1781        the class, so got the count wrong. When actually compiling, the reference
1782        was subsequently set up correctly.)
1784    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1785        it was being rejected as not supported by pcre_dfa_exec(), even though
1786        other assertions are supported. I have made pcre_dfa_exec() support
1787        (*FAIL).
1789    16. The implementation of 13c above involved the invention of a new opcode,
1790        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1791        cannot be changed at match time, I realized I could make a small
1792        improvement to matching performance by compiling OP_ALLANY instead of
1793        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1794        on the OP_ANY path.
1796    17. Compiling pcretest on Windows with readline support failed without the
1797        following two fixes: (1) Make the unistd.h include conditional on
1798        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1800    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1801        ncurses library to be included for pcretest when ReadLine support is
1802        requested, but also to allow for it to be overridden. This patch came from
1803        Daniel Bergström.
1805    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1806        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1807        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1808        spotting this.
1811    Version 7.6 28-Jan-08
1812    ---------------------
1814    1.  A character class containing a very large number of characters with
1815        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1816        overflow.
1818    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1819        HAVE_LONG_LONG is not defined.
1821    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1822        bring them up to date and include new features. This patch includes:
1824        - Fixed PH's badly added libz and libbz2 support.
1825        - Fixed a problem with static linking.
1826        - Added pcredemo. [But later removed - see 7 below.]
1827        - Fixed dftables problem and added an option.
1828        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1829            HAVE_LONG_LONG.
1830        - Added readline support for pcretest.
1831        - Added an listing of the option settings after cmake has run.
1833    4.  A user submitted a patch to Makefile that makes it easy to create
1834        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1835        Makefile.am that cause it to include this special target, without
1836        affecting anything else. Note that the same mingw target plus all
1837        the other distribution libraries and programs are now supported
1838        when configuring with CMake (see 6 below) instead of with
1839        Configure/Make.
1841    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1842        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1843        exported in the Windows port". It has not yet been confirmed that the patch
1844        solves the problem, but it does no harm.
1846    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1847        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1848        with CMake, and also correct the comment about stack recursion.
1850    7.  Remove the automatic building of pcredemo from the ./configure system and
1851        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1852        of a program that users should build themselves after PCRE is installed, so
1853        building it automatically is not really right. What is more, it gave
1854        trouble in some build environments.
1856    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1859    Version 7.5 10-Jan-08
1860    ---------------------
1862    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1863        values in parens when parsing an RE using the C++ wrapper."
1865    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1866        Characters greater than 255 were excluded from the class instead of being
1867        included.
1869    3.  The same bug as (2) above applied to negated POSIX classes such as
1870        [:^space:].
1872    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1873        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1874        I have changed it.
1876    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1877        first named subpattern) and a construct such as (?&a) would reference the
1878        first named subpattern whose name started with "a" (in other words, the
1879        length check was missing). Both these problems are fixed. "Subpattern name
1880        expected" is now given for (?&) (a zero-length name), and this patch also
1881        makes it give the same error for \k'' (previously it complained that that
1882        was a reference to a non-existent subpattern).
1884    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1885        this is right because (?- can be followed by option settings as well as by
1886        digits. I have, however, made the messages clearer.
1888    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1889        than the number used in the conditional) now cause a compile-time error.
1890        This is actually not compatible with Perl, which accepts such patterns, but
1891        treats the conditional as always being FALSE (as PCRE used to), but it
1892        seems to me that giving a diagnostic is better.
1894    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1895        and messages.
1897    9.  Fix two occurrences of "backslash" in comments that should have been
1898        "backspace".
1900    10. Remove two redundant lines of code that can never be obeyed (their function
1901        was moved elsewhere).
1903    11. The program that makes PCRE's Unicode character property table had a bug
1904        which caused it to generate incorrect table entries for sequences of
1905        characters that have the same character type, but are in different scripts.
1906        It amalgamated them into a single range, with the script of the first of
1907        them. In other words, some characters were in the wrong script. There were
1908        thirteen such cases, affecting characters in the following ranges:
1910          U+002b0 - U+002c1
1911          U+0060c - U+0060d
1912          U+0061e - U+00612
1913          U+0064b - U+0065e
1914          U+0074d - U+0076d
1915          U+01800 - U+01805
1916          U+01d00 - U+01d77
1917          U+01d9b - U+01dbf
1918          U+0200b - U+0200f
1919          U+030fc - U+030fe
1920          U+03260 - U+0327f
1921          U+0fb46 - U+0fbb1
1922          U+10450 - U+1049d
1924    12. The -o option (show only the matching part of a line) for pcregrep was not
1925        compatible with GNU grep in that, if there was more than one match in a
1926        line, it showed only the first of them. It now behaves in the same way as
1927        GNU grep.
1929    13. If the -o and -v options were combined for pcregrep, it printed a blank
1930        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1931        does the same. The return code can be used to tell if there were any
1932        non-matching lines.
1934    14. Added --file-offsets and --line-offsets to pcregrep.
1936    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1937        infinitely looping recursion. The bug was that positive lookaheads were not
1938        being skipped when checking for a possible empty match (negative lookaheads
1939        and both kinds of lookbehind were skipped).
1941    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1942        inclusion of <windows.h> to before rather than after the definition of
1943        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1945    17. Specifying a possessive quantifier with a specific limit for a Unicode
1946        character property caused pcre_compile() to compile bad code, which led at
1947        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1948        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1949        caused the error; without that there was no problem.
1951    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1953    19. Added --enable-pcretest-libreadline.
1955    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1956        RE::GlobalReplace(). As a result, the number of replacements returned was
1957        double what it should be. I removed one of the increments, but Craig sent a
1958        later patch that removed the other one (the right fix) and added unit tests
1959        that check the return values (which was not done before).
1961    21. Several CMake things:
1963        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1964            the names libpcre and libpcreposix, not just pcre and pcreposix.
1966        (2) The above change means that pcretest and pcregrep are now correctly
1967            linked with the newly-built libraries, not previously installed ones.
1971    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1972        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1973        UTF-8 newline character). The key issue is that the pattern starts .*;
1974        this means that the match must be either at the beginning, or after a
1975        newline. The bug was in the code for advancing after a failed match and
1976        checking that the new position followed a newline. It was not taking
1977        account of UTF-8 characters correctly.
1979    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1980        character classes. PCRE was not treating the sequence [:...:] as a
1981        character class unless the ... were all letters. Perl, however, seems to
1982        allow any characters between [: and :], though of course it rejects as
1983        unknown any "names" that contain non-letters, because all the known class
1984        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1985        for example, whereas PCRE did not - it did not recognize a POSIX character
1986        class. This seemed a bit dangerous, so the code has been changed to be
1987        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1988        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1989        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1990        Perl does, and where it didn't before.
1992    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1993        Windows environments %n is disabled by default.
1996    Version 7.4 21-Sep-07
1997    ---------------------
1999    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
2000        means that a class such as [\s] counted as "explicit reference to CR or
2001        LF". That isn't really right - the whole point of the change was to try to
2002        help when there was an actual mention of one of the two characters. So now
2003        the change happens only if \r or \n (or a literal CR or LF) character is
2004        encountered.
2006    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
2007        of both had grown to the point where there were only 3 bits left.
2008        Fortunately, there was spare space in the data structure, and so I have
2009        moved the internal flags into a new 16-bit field to free up more option
2010        bits.
2012    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
2013        but did not set the internal JCHANGED flag - either of these is enough to
2014        control the way the "get" function works - but the PCRE_INFO_JCHANGED
2015        facility is supposed to tell if (?J) was ever used, so now (?J) at the
2016        start sets both bits.
2018    4.  Added options (at build time, compile time, exec time) to change \R from
2019        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
2021    5.  doc/pcresyntax.html was missing from the distribution.
2023    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
2024        compatibility, even though it is no longer used.
2026    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
2027        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
2028        windows.h file is present (where different names are used). [This was
2029        reversed later after testing - see 16 below.]
2031    8.  Changed all #include <config.h> to #include "config.h". There were also
2032        some further <pcre.h> cases that I changed to "pcre.h".
2034    9.  When pcregrep was used with the --colour option, it missed the line ending
2035        sequence off the lines that it output.
2037    10. It was pointed out to me that arrays of string pointers cause lots of
2038        relocations when a shared library is dynamically loaded. A technique of
2039        using a single long string with a table of offsets can drastically reduce
2040        these. I have refactored PCRE in four places to do this. The result is
2041        dramatic:
2043          Originally:                          290
2044          After changing UCP table:            187
2045          After changing error message table:   43
2046          After changing table of "verbs"       36
2047          After changing table of Posix names   22
2049        Thanks to the folks working on Gregex for glib for this insight.
2051    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
2052        unicode-properties was also set.
2054    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
2056    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
2057        checked only for CRLF.
2059    14. Added casts to pcretest.c to avoid compiler warnings.
2061    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
2063    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
2064        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2065        entirely. This removes changes made in 7 above.
2067    17. The CMake files have been updated, and there is now more information about
2068        building with CMake in the NON-UNIX-USE document.
2071    Version 7.3 28-Aug-07
2072    ---------------------
2074     1. In the rejigging of the build system that eventually resulted in 7.1, the
2075        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
2076        brackets there is not right, since it causes compilers to look for an
2077        installed pcre.h, not the version that is in the source that is being
2078        compiled (which of course may be different). I have changed it back to:
2080          #include "pcre.h"
2082        I have a vague recollection that the change was concerned with compiling in
2083        different directories, but in the new build system, that is taken care of
2084        by the VPATH setting the Makefile.
2086     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
2087        when the subject happened to end in the byte 0x85 (e.g. if the last
2088        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
2089        characters but of course it shouldn't be taken as a newline when it is part
2090        of another character. The bug was that, for an unlimited repeat of . in
2091        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
2092        characters when looking for a newline.
2094     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
2096     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
2097        in debug output.
2099     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
2100        long printing in the pcrecpp unittest when running under MinGW.
2102     6. ESC_K was left out of the EBCDIC table.
2104     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
2105        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
2106        limit also applies to "virtual nesting" when a pattern is recursive, and in
2107        this case 1000 isn't so big. I have been able to remove this limit at the
2108        expense of backing off one optimization in certain circumstances. Normally,
2109        when pcre_exec() would call its internal match() function recursively and
2110        immediately return the result unconditionally, it uses a "tail recursion"
2111        feature to save stack. However, when a subpattern that can match an empty
2112        string has an unlimited repetition quantifier, it no longer makes this
2113        optimization. That gives it a stack frame in which to save the data for
2114        checking that an empty string has been matched. Previously this was taken
2115        from the 1000-entry workspace that had been reserved. So now there is no
2116        explicit limit, but more stack is used.
2118     8. Applied Daniel's patches to solve problems with the import/export magic
2119        syntax that is required for Windows, and which was going wrong for the
2120        pcreposix and pcrecpp parts of the library. These were overlooked when this
2121        problem was solved for the main library.
2123     9. There were some crude static tests to avoid integer overflow when computing
2124        the size of patterns that contain repeated groups with explicit upper
2125        limits. As the maximum quantifier is 65535, the maximum group length was
2126        set at 30,000 so that the product of these two numbers did not overflow a
2127        32-bit integer. However, it turns out that people want to use groups that
2128        are longer than 30,000 bytes (though not repeat them that many times).
2129        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
2130        made it possible to implement the integer overflow checks in a much more
2131        dynamic way, which I have now done. The artificial limitation on group
2132        length has been removed - we now have only the limit on the total length of
2133        the compiled pattern, which depends on the LINK_SIZE setting.
2135    10. Fixed a bug in the documentation for get/copy named substring when
2136        duplicate names are permitted. If none of the named substrings are set, the
2137        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
2138        empty string.
2140    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
2141        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
2142        because the ] is interpreted as the first data character and the
2143        terminating ] is not found. PCRE has been made compatible with Perl in this
2144        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
2145        cause memory overwriting.
2147    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
2148        string has been matched (to stop an infinite loop). It was not recognizing
2149        a conditional subpattern that could match an empty string if that
2150        subpattern was within another subpattern. For example, it looped when
2151        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
2152        condition was not nested. This bug has been fixed.
2154    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
2155        past the start of the subject in the presence of bytes with the top bit
2156        set, for example "\x8aBCD".
2158    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
2159        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
2161    14. Optimized (?!) to (*FAIL).
2163    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
2164        This restricts code points to be within the range 0 to 0x10FFFF, excluding
2165        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
2166        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
2167        does: it's just the validity check that is more restrictive.
2169    16. Inserted checks for integer overflows during escape sequence (backslash)
2170        processing, and also fixed erroneous offset values for syntax errors during
2171        backslash processing.
2173    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
2174        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
2176    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
2177        caused an overrun.
2179    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
2180        something other than just ASCII characters) inside a group that had an
2181        unlimited repeat caused a loop at compile time (while checking to see
2182        whether the group could match an empty string).
2184    20. Debugging a pattern containing \p or \P could cause a crash. For example,
2185        [\P{Any}] did so. (Error in the code for printing property names.)
2187    21. An orphan \E inside a character class could cause a crash.
2189    22. A repeated capturing bracket such as (A)? could cause a wild memory
2190        reference during compilation.
2192    23. There are several functions in pcre_compile() that scan along a compiled
2193        expression for various reasons (e.g. to see if it's fixed length for look
2194        behind). There were bugs in these functions when a repeated \p or \P was
2195        present in the pattern. These operators have additional parameters compared
2196        with \d, etc, and these were not being taken into account when moving along
2197        the compiled data. Specifically:
2199        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2200            length.
2202        (b) An item such as \pL+ within a repeated group could cause crashes or
2203            loops.
2205        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2206            "reference to non-existent subpattern" error.
2208        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2210    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2211        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2213    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2215    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2216        character were causing crashes (broken optimization).
2218    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2219        \p or \P) caused a compile-time loop.
2221    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2222        break. For example, the unstudied pattern [\r\n]A does not match the string
2223        "\r\nA" because change 7.0/46 below moves the current point on by two
2224        characters after failing to match at the start. However, the pattern \nA
2225        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2226        the same is true. There doesn't seem any very clean way out of this, but
2227        what I have chosen to do makes the common cases work: PCRE now takes note
2228        of whether there can be an explicit match for \r or \n anywhere in the
2229        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2230        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2231        pattern has explicit CR or LF references.
2233    29. Added (*CR) etc for changing newline setting at start of pattern.
2236    Version 7.2 19-Jun-07
2237    ---------------------
2239     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
2240        which is apparently normally available under Windows.
2242     2. Re-jig the pcregrep tests with different newline settings in an attempt
2243        to make them independent of the local environment's newline setting.
2245     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
2247     4. Some of the "internals" tests were previously cut out when the link size
2248        was not 2, because the output contained actual offsets. The recent new
2249        "Z" feature of pcretest means that these can be cut out, making the tests
2250        usable with all link sizes.
2252     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
2253        stack recursion. This gives a massive performance boost under BSD, but just
2254        a small improvement under Linux. However, it saves one field in the frame
2255        in all cases.
2257     6. Added more features from the forthcoming Perl 5.10:
2259        (a) (?-n) (where n is a string of digits) is a relative subroutine or
2260            recursion call. It refers to the nth most recently opened parentheses.
2262        (b) (?+n) is also a relative subroutine call; it refers to the nth next
2263            to be opened parentheses.
2265        (c) Conditions that refer to capturing parentheses can be specified
2266            relatively, for example, (?(-2)... or (?(+3)...
2268        (d) \K resets the start of the current match so that everything before
2269            is not part of it.
2271        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2273        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2274            reference syntax.
2276        (g) (?| introduces a group in which the numbering of parentheses in each
2277            alternative starts with the same number.
2279        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2281     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2284     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2285        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2286        for detecting groups that can match an empty string.
2288     9. A pattern with a very large number of alternatives (more than several
2289        hundred) was running out of internal workspace during the pre-compile
2290        phase, where pcre_compile() figures out how much memory will be needed. A
2291        bit of new cunning has reduced the workspace needed for groups with
2292        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2293        workspace instead of running out of the 4096 that are available.
2295    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2297    11. Applied patch from Google to remove an optimization that didn't quite work.
2298        The report of the bug said:
2300          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2301          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2302          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2304    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2305        it matched the wrong number of bytes.
2308    Version 7.1 24-Apr-07
2309    ---------------------
2311     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
2312        that is more "standard", making use of automake and other Autotools. There
2313        is some re-arrangement of the files and adjustment of comments consequent
2314        on this.
2316     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
2317        for recursive directory scanning broke on some systems because the files
2318        are not scanned in any specific order and on different systems the order
2319        was different. A call to "sort" has been inserted into RunGrepTest for the
2320        approprate test as a short-term fix. In the longer term there may be an
2321        alternative.
2323     3. I had an email from Eric Raymond about problems translating some of PCRE's
2324        man pages to HTML (despite the fact that I distribute HTML pages, some
2325        people do their own conversions for various reasons). The problems
2326        concerned the use of low-level troff macros .br and .in. I have therefore
2327        removed all such uses from the man pages (some were redundant, some could
2328        be replaced by .nf/.fi pairs). The 132html script that I use to generate
2329        HTML has been updated to handle .nf/.fi and to complain if it encounters
2330        .br or .in.
2332     4. Updated comments in configure.ac that get placed in config.h.in and also
2333        arranged for config.h to be included in the distribution, with the name
2334        config.h.generic, for the benefit of those who have to compile without
2335        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
2337     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
2338        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
2339        makevp.bat for the current PCRE, using the additional files
2340        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
2342     6. A Windows user reported a minor discrepancy with test 2, which turned out
2343        to be caused by a trailing space on an input line that had got lost in his
2344        copy. The trailing space was an accident, so I've just removed it.
2346     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
2347        that is needed.
2349     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
2350        as "const" (a) because they are and (b) because it helps the PHP
2351        maintainers who have recently made a script to detect big data structures
2352        in the php code that should be moved to the .rodata section. I remembered
2353        to update Builducptable as well, so it won't revert if ucptable.h is ever
2354        re-created.
2356     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
2357        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
2358        order to be able to cut out the UTF-8 tables in the latter when UTF-8
2359        support is not required. This saves 1.5-2K of code, which is important in
2360        some applications.
2362        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
2363        so as not to refer to the tables, even though these functions will never be
2364        called when UTF-8 support is disabled. Otherwise there are problems with a
2365        shared library.
2367    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
2369        (a) It was defining its arguments as char * instead of void *.
2371        (b) It was assuming that all moves were upwards in memory; this was true
2372            a long time ago when I wrote it, but is no longer the case.
2374        The emulated memove() is provided for those environments that have neither
2375        memmove() nor bcopy(). I didn't think anyone used it these days, but that
2376        is clearly not the case, as these two bugs were recently reported.
2378    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
2379        and Detrail to create the HTML documentation, the .txt form of the man
2380        pages, and it removes trailing spaces from listed files. It also creates
2381        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
2382        case, it wraps all the #defines with #ifndefs. This script should be run
2383        before "make dist".
2385    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
2386        with Unicode property support.
2388        (a) For a maximizing quantifier, if the two different cases of the
2389            character were of different lengths in their UTF-8 codings (there are
2390            some cases like this - I found 11), and the matching function had to
2391            back up over a mixture of the two cases, it incorrectly assumed they
2392            were both the same length.
2394        (b) When PCRE was configured to use the heap rather than the stack for
2395            recursion during matching, it was not correctly preserving the data for
2396            the other case of a UTF-8 character when checking ahead for a match
2397            while processing a minimizing repeat. If the check also involved
2398            matching a wide character, but failed, corruption could cause an
2399            erroneous result when trying to check for a repeat of the original
2400            character.
2402    13. Some tidying changes to the testing mechanism:
2404        (a) The RunTest script now detects the internal link size and whether there
2405            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
2406            values substituted by "configure". (The RunGrepTest script already did
2407            this for UTF-8.) The configure.ac script no longer substitutes the
2408            relevant variables.
2410        (b) The debugging options /B and /D in pcretest show the compiled bytecode
2411            with length and offset values. This means that the output is different
2412            for different internal link sizes. Test 2 is skipped for link sizes
2413            other than 2 because of this, bypassing the problem. Unfortunately,
2414            there was also a test in test 3 (the locale tests) that used /B and
2415            failed for link sizes other than 2. Rather than cut the whole test out,
2416            I have added a new /Z option to pcretest that replaces the length and
2417            offset values with spaces. This is now used to make test 3 independent
2418            of link size. (Test 2 will be tidied up later.)
2420    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
2421        segmentation fault instead of returning the appropriate error message.
2423    15. In multiline mode when the newline sequence was set to "any", the pattern
2424        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
2425        This doesn't seem right; it now treats the CRLF combination as the line
2426        ending, and so does not match in that case. It's only a pattern such as ^$
2427        that would hit this one: something like ^ABC$ would have failed after \r
2428        and then tried again after \r\n.
2430    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
2431        in an attempt to make files that differ only in their line terminators
2432        compare equal. This works on Linux.
2434    17. Under certain error circumstances pcregrep might try to free random memory
2435        as it exited. This is now fixed, thanks to valgrind.
2437    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
2438        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
2439        was because its rules for how to advance for /g after matching an empty
2440        string at the end of a line did not allow for this case. They now check for
2441        it specially.
2443    20. pcretest is supposed to handle patterns and data of any length, by
2444        extending its buffers when necessary. It was getting this wrong when the
2445        buffer for a data line had to be extended.
2447    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
2448        CRLF as a newline sequence.
2450    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
2451        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
2452        I have nevertheless tidied it up.
2454    23. Added some casts to kill warnings from HP-UX ia64 compiler.
2456    24. Added a man page for pcre-config.
2459    Version 7.0 19-Dec-06
2460    ---------------------
2462     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
2463        moving to gcc 4.1.1.
2465     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
2466        sys/time.h, which is documented as needed for this function. It doesn't
2467        seem to matter on Linux, but it showed up on some releases of OS X.
2469     3. It seems that there are systems where bytes whose values are greater than
2470        127 match isprint() in the "C" locale. The "C" locale should be the
2471        default when a C program starts up. In most systems, only ASCII printing
2472        characters match isprint(). This difference caused the output from pcretest
2473        to vary, making some of the tests fail. I have changed pcretest so that:
2475        (a) When it is outputting text in the compiled version of a pattern, bytes
2476            other than 32-126 are always shown as hex escapes.
2478        (b) When it is outputting text that is a matched part of a subject string,
2479            it does the same, unless a different locale has been set for the match
2480            (using the /L modifier). In this case, it uses isprint() to decide.
2482     4. Fixed a major bug that caused incorrect computation of the amount of memory
2483        required for a compiled pattern when options that changed within the
2484        pattern affected the logic of the preliminary scan that determines the
2485        length. The relevant options are -x, and -i in UTF-8 mode. The result was
2486        that the computed length was too small. The symptoms of this bug were
2487        either the PCRE error "internal error: code overflow" from pcre_compile(),
2488        or a glibc crash with a message such as "pcretest: free(): invalid next
2489        size (fast)". Examples of patterns that provoked this bug (shown in
2490        pcretest format) are:
2492          /(?-x: )/x
2493          /(?x)(?-x: \s*#\s*)/
2494          /((?i)[\x{c0}])/8
2495          /(?i:[\x{c0}])/8
2497        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
2498        is now done differently.
2500     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
2501        wrapper classes; (b) implement a new function in the C++ scanner that is
2502        more efficient than the old way of doing things because it avoids levels of
2503        recursion in the regex matching; (c) add a paragraph to the documentation
2504        for the FullMatch() function.
2506     6. The escape sequence \n was being treated as whatever was defined as
2507        "newline". Not only was this contrary to the documentation, which states
2508        that \n is character 10 (hex 0A), but it also went horribly wrong when
2509        "newline" was defined as CRLF. This has been fixed.
2511     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
2512        was being set to -1 for the "end of line" case (supposedly a value that no
2513        character can have). Though this value is never used (the check for end of
2514        line is "zero bytes in current character"), it caused compiler complaints.
2515        I've changed it to 0xffffffff.
2517     8. In pcre_version.c, the version string was being built by a sequence of
2518        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
2519        string (as it is for production releases) called a macro with an empty
2520        argument. The C standard says the result of this is undefined. The gcc
2521        compiler treats it as an empty string (which was what was wanted) but it is
2522        reported that Visual C gives an error. The source has been hacked around to
2523        avoid this problem.
2525     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
2526        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
2527        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
2528        of them did).
2530    10. Originally, pcretest opened its input and output without "b"; then I was
2531        told that "b" was needed in some environments, so it was added for release
2532        5.0 to both the input and output. (It makes no difference on Unix-like
2533        systems.) Later I was told that it is wrong for the input on Windows. I've
2534        now abstracted the modes into two macros, to make it easier to fiddle with
2535        them, and removed "b" from the input mode under Windows.
2537    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
2539    12. Added -help and --help to pcretest as an official way of being reminded
2540        of the options.
2542    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
2543        and pcrecpp.cc because they annoy compilers at high warning levels.
2545    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
2547    15. Fixed an occurrence of == in configure.ac that should have been = (shell
2548        scripts are not C programs :-) and which was not noticed because it works
2549        on Linux.
2551    16. pcretest is supposed to handle any length of pattern and data line (as one
2552        line or as a continued sequence of lines) by extending its input buffer if
2553        necessary. This feature was broken for very long pattern lines, leading to
2554        a string of junk being passed to pcre_compile() if the pattern was longer
2555        than about 50K.
2557    17. I have done a major re-factoring of the way pcre_compile() computes the
2558        amount of memory needed for a compiled pattern. Previously, there was code
2559        that made a preliminary scan of the pattern in order to do this. That was
2560        OK when PCRE was new, but as the facilities have expanded, it has become
2561        harder and harder to keep it in step with the real compile phase, and there
2562        have been a number of bugs (see for example, 4 above). I have now found a
2563        cunning way of running the real compile function in a "fake" mode that
2564        enables it to compute how much memory it would need, while actually only
2565        ever using a few hundred bytes of working memory and without too many
2566        tests of the mode. This should make future maintenance and development
2567        easier. A side effect of this work is that the limit of 200 on the nesting
2568        depth of parentheses has been removed (though this was never a serious
2569        limitation, I suspect). However, there is a downside: pcre_compile() now
2570        runs more slowly than before (30% or more, depending on the pattern). I
2571        hope this isn't a big issue. There is no effect on runtime performance.
2573    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
2574        newline (only possible for the last line of a file) and it was a
2575        pattern that set a locale (followed by /Lsomething), pcretest crashed.
2577    19. Added additional timing features to pcretest. (1) The -tm option now times
2578        matching only, not compiling. (2) Both -t and -tm can be followed, as a
2579        separate command line item, by a number that specifies the number of
2580        repeats to use when timing. The default is 50000; this gives better
2581        precision, but takes uncomfortably long for very large patterns.
2583    20. Extended pcre_study() to be more clever in cases where a branch of a
2584        subpattern has no definite first character. For example, (a*|b*)[cd] would
2585        previously give no result from pcre_study(). Now it recognizes that the
2586        first character must be a, b, c, or d.
2588    21. There was an incorrect error "recursive call could loop indefinitely" if
2589        a subpattern (or the entire pattern) that was being tested for matching an
2590        empty string contained only one non-empty item after a nested subpattern.
2591        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
2592        incorrectly, because the \d was being skipped in the check.
2594    22. The pcretest program now has a new pattern option /B and a command line
2595        option -b, which is equivalent to adding /B to every pattern. This causes
2596        it to show the compiled bytecode, without the additional information that
2597        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
2598        is the same as /B/I).
2600    23. A new optimization is now able automatically to treat some sequences such
2601        as a*b as a*+b. More specifically, if something simple (such as a character
2602        or a simple class like \d) has an unlimited quantifier, and is followed by
2603        something that cannot possibly match the quantified thing, the quantifier
2604        is automatically "possessified".
2606    24. A recursive reference to a subpattern whose number was greater than 39
2607        went wrong under certain circumstances in UTF-8 mode. This bug could also
2608        have affected the operation of pcre_study().
2610    25. Realized that a little bit of performance could be had by replacing
2611        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
2613    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
2615    27. Possessive quantifiers such as a++ were previously implemented by turning
2616        them into atomic groups such as ($>a+). Now they have their own opcodes,
2617        which improves performance. This includes the automatically created ones
2618        from 23 above.
2620    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
2621        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
2622        the first matched character to be a colon. This applied both to named and
2623        numbered groups.
2625    29. The ucpinternal.h header file was missing its idempotency #ifdef.
2627    30. I was sent a "project" file called libpcre.a.dev which I understand makes
2628        building PCRE on Windows easier, so I have included it in the distribution.
2630    31. There is now a check in pcretest against a ridiculously large number being
2631        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
2632        loop, the loop is abandoned.
2634    32. Forward references to subpatterns in conditions such as (?(2)...) where
2635        subpattern 2 is defined later cause pcre_compile() to search forwards in
2636        the pattern for the relevant set of parentheses. This search went wrong
2637        when there were unescaped parentheses in a character class, parentheses
2638        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2640    33. "Subroutine" calls and backreferences were previously restricted to
2641        referencing subpatterns earlier in the regex. This restriction has now
2642        been removed.
2644    34. Added a number of extra features that are going to be in Perl 5.10. On the
2645        whole, these are just syntactic alternatives for features that PCRE had
2646        previously implemented using the Python syntax or my own invention. The
2647        other formats are all retained for compatibility.
2649        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2650            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2651            also .NET compatible.
2653        (b) A recursion or subroutine call to a named group can now be defined as
2654            (?&name) as well as (?P>name).
2656        (c) A backreference to a named group can now be defined as \k<name> or
2657            \k'name' as well as (?P=name). The new forms, as well as being in Perl
2658            5.10, are also .NET compatible.
2660        (d) A conditional reference to a named group can now use the syntax
2661            (?(<name>) or (?('name') as well as (?(name).
2663        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2664            groups (named and numbered) that are never evaluated inline, but can be
2665            called as "subroutines" from elsewhere. In effect, the DEFINE condition
2666            is always false. There may be only one alternative in such a group.
2668        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2669            as the simple (?(R). The condition is true only if the most recent
2670            recursion is that of the given number or name. It does not search out
2671            through the entire recursion stack.
2673        (g) The escape \gN or \g{N} has been added, where N is a positive or
2674            negative number, specifying an absolute or relative reference.
2676    35. Tidied to get rid of some further signed/unsigned compiler warnings and
2677        some "unreachable code" warnings.
2679    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2680        things, this adds five new scripts.
2682    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2683        There were also incompatibilities regarding the handling of \Q..\E inside
2684        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2685        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2687    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2688        matches an empty string, and forcibly breaks the loop. There were bugs in
2689        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2690        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2691        separate and independent bugs (that affected different cases) have been
2692        fixed.
2694    39. Refactored the code to abolish the use of different opcodes for small
2695        capturing bracket numbers. This is a tidy that I avoided doing when I
2696        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2697        The new approach is not only tidier, it makes it possible to reduce the
2698        memory needed to fix the previous bug (38).
2700    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2701        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2702        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2703        mode.
2705    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2706        report.
2708    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2709        copy construction and assignment in the C++ wrapper.
2711    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2712        couple of bugs that could have given wrong results in the "--newline=crlf"
2713        case.
2715    44. Added a number of casts and did some reorganization of signed/unsigned int
2716        variables following suggestions from Dair Grant. Also renamed the variable
2717        "this" as "item" because it is a C++ keyword.
2719    45. Arranged for dftables to add
2721          #include "pcre_internal.h"
2723        to pcre_chartables.c because without it, gcc 4.x may remove the array
2724        definition from the final binary if PCRE is built into a static library and
2725        dead code stripping is activated.
2727    46. For an unanchored pattern, if a match attempt fails at the start of a
2728        newline sequence, and the newline setting is CRLF or ANY, and the next two
2729        characters are CRLF, advance by two characters instead of one.
2732  Version 6.7 04-Jul-06  Version 6.7 04-Jul-06
2733  ---------------------  ---------------------

Removed from v.91  
changed lines
  Added in v.1238

  ViewVC Help
Powered by ViewVC 1.1.5