ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 113 by ph10, Fri Mar 9 10:13:26 2007 UTC revision 1261 by ph10, Wed Feb 27 16:27:01 2013 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4  Version 7.1 05-Mar-07  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.
29    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
30        unnecessary spaces.
32    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
33        and later.
35    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
37    9.  Optimizing fast_forward_start_bits in JIT.
39    10. Adding experimental support for callouts in JIT, and fixing some
40        issues revealed during this work. Namely:
42        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
44        (b) Minimum length was not checked before the matching is started.
46    11. The value of capture_last that is passed to callouts was incorrect in some
47        cases when there was a capture on one path that was subsequently abandoned
48        after a backtrack. Also, the capture_last value is now reset after a
49        recursion, since all captures are also reset in this case.
51    12. The interpreter no longer returns the "too many substrings" error in the
52        case when an overflowing capture is in a branch that is subsequently
53        abandoned after a backtrack.
55    13. In the pathological case when an offset vector of size 2 is used, pcretest
56        now prints out the matched string after a yield of 0 or 1.
58    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
59        Only supported by the JIT compiler at the moment.
61    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
63    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
64        offset of the starting point of the matching process, provided the offsets
65        vector is large enough.
67    17. The \A escape now records a lookbehind value of 1, though its execution
68        does not actually inspect the previous character. This is to ensure that,
69        in partial multi-segment matching, at least one character from the old
70        segment is retained when a new segment is processed. Otherwise, if there
71        are no lookbehinds in the pattern, \A might match incorrectly at the start
72        of a new segment.
74    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
76    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
77        gives some modest performance improvement in 8-bit mode.
79    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
80        be expressed in certain programming languages using Universal Character
81        Names.
83    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
84        which points out that "non characters" are not "characters that may not
85        appear in Unicode strings" but rather "characters that are reserved for
86        internal use and have only local meaning".
89    Version 8.32 30-November-2012
90    -----------------------------
92    1.  Improved JIT compiler optimizations for first character search and single
93        character iterators.
95    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
96        Patch by Daniel Richard G.
98    3.  Single character iterator optimizations in the JIT compiler.
100    4.  Improved JIT compiler optimizations for character ranges.
102    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
103        Reported by Giuseppe D'Angelo.
105    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
106        of a line, was being set incorrectly in cases where .* appeared inside
107        atomic brackets at the start of a pattern, or where there was a subsequent
108        *PRUNE or *SKIP.
110    7.  Improved instruction cache flush for POWER/PowerPC.
111        Patch by Daniel Richard G.
113    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
114        grep:
116        (a) There is now no limit to the number of patterns to be matched.
118        (b) An error is given if a pattern is too long.
120        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
121            are now supported.
123        (d) --exclude-from and --include-from (multiple use) have been added.
125        (e) Exclusions and inclusions now apply to all files and directories, not
126            just to those obtained from scanning a directory recursively.
128        (f) Multiple uses of -f and --file-list are now supported.
130        (g) In a Windows environment, the default for -d has been changed from
131            "read" (the GNU grep default) to "skip", because otherwise the presence
132            of a directory in the file list provokes an error.
134        (h) The documentation has been revised and clarified in places.
136    9.  Improve the matching speed of capturing brackets.
138    10. Changed the meaning of \X so that it now matches a Unicode extended
139        grapheme cluster.
141    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
142        out POSIX threads when JIT support is configured.
144    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
146    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
147        structure could go wrong in environments where size_t is not the same size
148        as int.
150    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
151        set.
153    15. The EBCDIC support had decayed; later updates to the code had included
154        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
155        general tidy up of EBCDIC-related issues, and the documentation was also
156        not quite right. There is now a test that can be run on ASCII systems to
157        check some of the EBCDIC-related things (but is it not a full test).
159    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
160        in a small tidy to the code.
162    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
164    18. If the --only-matching (-o) option in pcregrep is specified multiple
165        times, each one causes appropriate output. For example, -o1 -o2 outputs the
166        substrings matched by the 1st and 2nd capturing parentheses. A separating
167        string can be specified by --om-separator (default empty).
169    19. Improving the first n character searches.
171    20. Turn case lists for horizontal and vertical white space into macros so that
172        they are defined only once.
174    21. This set of changes together give more compatible Unicode case-folding
175        behaviour for characters that have more than one other case when UCP
176        support is available.
178        (a) The Unicode property table now has offsets into a new table of sets of
179            three or more characters that are case-equivalent. The MultiStage2.py
180            script that generates these tables (the pcre_ucd.c file) now scans
181            CaseFolding.txt instead of UnicodeData.txt for character case
182            information.
184        (b) The code for adding characters or ranges of characters to a character
185            class has been abstracted into a generalized function that also handles
186            case-independence. In UTF-mode with UCP support, this uses the new data
187            to handle characters with more than one other case.
189        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
190            whose other case is greater than 256 are now correctly matched
191            caselessly. Previously, the high codepoint matched the low one, but not
192            vice versa.
194        (d) The processing of \h, \H, \v, and \ in character classes now makes use
195            of the new class addition function, using character lists defined as
196            macros alongside the case definitions of 20 above.
198        (e) Caseless back references now work with characters that have more than
199            one other case.
201        (f) General caseless matching of characters with more than one other case
202            is supported.
204    22. Unicode character properties were updated from Unicode 6.2.0
206    23. Improved CMake support under Windows. Patch by Daniel Richard G.
208    24. Add support for 32-bit character strings, and UTF-32
210    25. Major JIT compiler update (code refactoring and bugfixing).
211        Experimental Sparc 32 support is added.
213    26. Applied a modified version of Daniel Richard G's patch to create
214        pcre.h.generic and config.h.generic by "make" instead of in the
215        PrepareRelease script.
217    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
218        pcre_compile.c when checking for a zero character.
220    28. Introducing a native interface for JIT. Through this interface, the compiled
221        machine code can be directly executed. The purpose of this interface is to
222        provide fast pattern matching, so several sanity checks are not performed.
223        However, feature tests are still performed. The new interface provides
224        1.4x speedup compared to the old one.
226    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
227        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
228        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
230    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
231        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
232        the "old" RFC 2279). Instead, it ended up passing a negative length to
233        pcre_exec().
235    31. Add support for GCC's visibility feature to hide internal functions.
237    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
238        "unknown -C option" after outputting 0 or 1.
240    33. There is now support for generating a code coverage report for the test
241        suite in environments where gcc is the compiler and lcov is installed. This
242        is mainly for the benefit of the developers.
244    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
245        unaddressable using valgrind annotations, allowing valgrind to detect
246        invalid memory accesses. This is mainly for the benefit of the developers.
248    25. (*UTF) can now be used to start a pattern in any of the three libraries.
250    26. Give configure error if --enable-cpp but no C++ compiler found.
253    Version 8.31 06-July-2012
254    -------------------------
256    1.  Fixing a wrong JIT test case and some compiler warnings.
258    2.  Removed a bashism from the RunTest script.
260    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
261        to unsigned type, result still unsigned" that was given by an MS compiler
262        on encountering the code "-sizeof(xxx)".
264    4.  Partial matching support is added to the JIT compiler.
266    5.  Fixed several bugs concerned with partial matching of items that consist
267        of more than one character:
269        (a) /^(..)\1/ did not partially match "aba" because checking references was
270            done on an "all or nothing" basis. This also applied to repeated
271            references.
273        (b) \R did not give a hard partial match if \r was found at the end of the
274            subject.
276        (c) \X did not give a hard partial match after matching one or more
277            characters at the end of the subject.
279        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
280            a partial match for the string "\r".
282        (e) When newline was set to CRLF, the metacharacter "." did not recognize
283            a partial match for a CR character at the end of the subject string.
285    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
286        running pcretest, the text "(JIT)" added to the output whenever JIT is
287        actually used to run the match.
289    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
290        or /S+[+] with a digit between 1 and 7.
292    8.  OP_NOT now supports any UTF character not just single-byte ones.
294    9.  (*MARK) control verb is now supported by the JIT compiler.
296    10. The command "./RunTest list" lists the available tests without actually
297        running any of them. (Because I keep forgetting what they all are.)
301    12. Applied a (slightly modified) user-supplied patch that improves performance
302        when the heap is used for recursion (compiled with --disable-stack-for-
303        recursion). Instead of malloc and free for each heap frame each time a
304        logical recursion happens, frames are retained on a chain and re-used where
305        possible. This sometimes gives as much as 30% improvement.
307    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
308        call.
310    14. As documented, (*COMMIT) is now confined to within a positive assertion.
312    15. It is now possible to link pcretest with libedit as an alternative to
313        libreadline.
315    16. (*COMMIT) control verb is now supported by the JIT compiler.
317    17. The Unicode data tables have been updated to Unicode 6.1.0.
319    18. Added --file-list option to pcregrep.
321    19. Added binary file support to pcregrep, including the -a, --binary-files,
322        -I, and --text options.
324    20. The madvise function is renamed for posix_madvise for QNX compatibility
325        reasons. Fixed by Giuseppe D'Angelo.
327    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
329    22. Changed the option for creating symbolic links for 16-bit man pages from
330        -s to -sf so that re-installing does not cause issues.
332    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
334    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
335        matches in certain environments (the workspace was not being correctly
336        retained). Also added to pcre_dfa_exec() a simple plausibility check on
337        some of the workspace data at the beginning of a restart.
339    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
340        was not doing so when it should - probably a typo introduced by SVN 528
341        (change 8.10/14).
343    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
344        \w+ when the character tables indicated that \x{c4} was a word character.
345        There were several related cases, all because the tests for doing a table
346        lookup were testing for characters less than 127 instead of 255.
348    27. If a pattern contains capturing parentheses that are not used in a match,
349        their slots in the ovector are set to -1. For those that are higher than
350        any matched groups, this happens at the end of processing. In the case when
351        there were back references that the ovector was too small to contain
352        (causing temporary malloc'd memory to be used during matching), and the
353        highest capturing number was not used, memory off the end of the ovector
354        was incorrectly being set to -1. (It was using the size of the temporary
355        memory instead of the true size.)
357    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
358        ovector size, it uses memory at the end of the block that it has got.
360    29. Check for an overlong MARK name and give an error at compile time. The
361        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
363    30. JIT compiler update.
365    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
366        Rill for the patch.
368    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
370    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
372    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
373        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
374        was enabled.
376    35. Improve JIT code generation for greedy plus quantifier.
378    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
379        1 to "aa" instead of to an empty string. The bug affected repeated groups
380        that could potentially match an empty string.
382    37. Optimizing single character iterators in JIT.
384    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
385        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
386        codepoints that are too big for the mode are faulted, and in a UTF mode,
387        disallowed codepoints are also faulted.
389    39. If PCRE was compiled with UTF support, in three places in the DFA
390        matcher there was code that should only have been obeyed in UTF mode, but
391        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
392        processing when bytes with values greater than 127 were present. In 16-bit
393        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
394        both cases the values are those that cannot be the first data item in a UTF
395        character. The three items that might have provoked this were recursions,
396        possessively repeated groups, and atomic groups.
398    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
399        and pcregrep, because some OS require shared objects to be explicitly
400        passed to ld, causing the link step to fail if they are not.
402    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
403        mode, patterns that started with \h* or \R* might be incorrectly matched.
406    Version 8.30 04-February-2012
407    -----------------------------
409    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
410        name is defined in ctype.h.
412    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
413        only in quite long subpatterns.
415    3.  Removed the function pcre_info(), which has been obsolete and deprecated
416        since it was replaced by pcre_fullinfo() in February 2000.
418    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
419        match a (*MARK), and the match failed at the start of the subject, a
420        reference to memory before the start of the subject could occur. This bug
421        was introduced by fix 17 of release 8.21.
423    5.  A reference to an unset group with zero minimum repetition was giving
424        totally wrong answers (in non-JavaScript-compatibility mode). For example,
425        /(another)?(\1?)test/ matched against "hello world test". This bug was
426        introduced in release 8.13.
428    6.  Add support for 16-bit character strings (a large amount of work involving
429        many changes and refactorings).
431    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
432        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
433        from a file.
435    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
436        rounding is not applied in this particular case).
438    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
439        if they appear, or are escaped, in patterns.
441    10. Get rid of a number of -Wunused-but-set-variable warnings.
443    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
444        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
445        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
446        also returns the mark "x". This bug applied to capturing parentheses,
447        non-capturing parentheses, and atomic parentheses. It also applied to some
448        assertions.
450    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
451        information out of configure.ac instead of relying on pcre.h.generic, which
452        is not stored in the repository.
454    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
455        -lreadline.
457    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
459    15. Applied Graycode's patch to put the top-level frame on the stack rather
460        than the heap when not using the stack for recursion. This gives a
461        performance improvement in many cases when recursion is not deep.
463    16. Experimental code added to "pcretest -C" to output the stack frame size.
466    Version 8.21 12-Dec-2011
467    ------------------------
469    1.  Updating the JIT compiler.
471    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
472        are added as well.
474    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
475        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
476        calling _pcre_jit_exec. Some extra comments are added.
478    4.  (*MARK) settings inside atomic groups that do not contain any capturing
479        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
480        was introduced by change 18 for 8.20.
482    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
483        ECMA-262 standard.
485    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
486        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
487        This bug was probably introduced by change 9 of 8.13.
489    7.  While fixing 6 above, I noticed that a number of other items were being
490        incorrectly rejected as "not fixed length". This arose partly because newer
491        opcodes had not been added to the fixed-length checking code. I have (a)
492        corrected the bug and added tests for these items, and (b) arranged for an
493        error to occur if an unknown opcode is encountered while checking for fixed
494        length instead of just assuming "not fixed length". The items that were
495        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
496        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
497        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
499    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
500        being incorrectly compiled and would have given unpredicatble results.
502    9.  A possessively repeated subpattern with minimum repeat count greater than
503        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
504        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
505        the first (A) could occur when it should not.
507    10. Add a cast and remove a redundant test from the code.
509    11. JIT should use pcre_malloc/pcre_free for allocation.
511    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
512        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
513        is anything other than /usr, -L is still shown).
515    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
517    14. Perl does not support \N without a following name in a [] class; PCRE now
518        also gives an error.
520    15. If a forward reference was repeated with an upper limit of around 2000,
521        it caused the error "internal error: overran compiling workspace". The
522        maximum number of forward references (including repeats) was limited by the
523        internal workspace, and dependent on the LINK_SIZE. The code has been
524        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
525        the default depends on LINK_SIZE. There is a new upper limit (for safety)
526        of around 200,000 forward references. While doing this, I also speeded up
527        the filling in of repeated forward references.
529    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
530        incorrectly expecting the subject to contain another "a" after the start.
532    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
533        in the match, the SKIP should be ignored. This was not happening; instead
534        the SKIP was being treated as NOMATCH. For patterns such as
535        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
536        tested.
538    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
539        now much more compatible with Perl, in particular in cases where the result
540        is a non-match for a non-anchored pattern. For example, if
541        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
542        "m", where previously it did not return a name. A side effect of this
543        change is that for partial matches, the last encountered mark name is
544        returned, as for non matches. A number of tests that were previously not
545        Perl-compatible have been moved into the Perl-compatible test files. The
546        refactoring has had the pleasing side effect of removing one argument from
547        the match() function, thus reducing its stack requirements.
549    19. If the /S+ option was used in pcretest to study a pattern using JIT,
550        subsequent uses of /S (without +) incorrectly behaved like /S+.
552    21. Retrieve executable code size support for the JIT compiler and fixing
553        some warnings.
555    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
556        not work when the shorter character appeared right at the end of the
557        subject string.
559    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
560        systems.
562    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
563        output it when the /M option is used in pcretest.
565    25. The CheckMan script was not being included in the distribution. Also, added
566        an explicit "perl" to run Perl scripts from the PrepareRelease script
567        because this is reportedly needed in Windows.
569    26. If study data was being save in a file and studying had not found a set of
570        "starts with" bytes for the pattern, the data written to the file (though
571        never used) was taken from uninitialized memory and so caused valgrind to
572        complain.
574    27. Updated RunTest.bat as provided by Sheri Pierce.
576    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
578    29. Computation of memory usage for the table of capturing group names was
579        giving an unnecessarily large value.
582    Version 8.20 21-Oct-2011
583    ------------------------
585    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
586        a POSIX class. After further experiments with Perl, which convinced me that
587        Perl has bugs and confusions, a closing square bracket is no longer allowed
588        in a POSIX name. This bug also affected patterns with classes that started
589        with full stops.
591    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
592        captured substring, but while checking the failing first alternative,
593        substring 1 is temporarily captured. If the output vector supplied to
594        pcre_exec() was not big enough for this capture, the yield of the function
595        was still zero ("insufficient space for captured substrings"). This cannot
596        be totally fixed without adding another stack variable, which seems a lot
597        of expense for a edge case. However, I have improved the situation in cases
598        such as /(a)(b)x|abc/ matched against "abc", where the return code
599        indicates that fewer than the maximum number of slots in the ovector have
600        been set.
602    3.  Related to (2) above: when there are more back references in a pattern than
603        slots in the output vector, pcre_exec() uses temporary memory during
604        matching, and copies in the captures as far as possible afterwards. It was
605        using the entire output vector, but this conflicts with the specification
606        that only 2/3 is used for passing back captured substrings. Now it uses
607        only the first 2/3, for compatibility. This is, of course, another edge
608        case.
610    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
611        main code base, and can be used by building with --enable-jit. When this is
612        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
613        runtime --no-jit option is given.
615    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
616        ovector, the return from the function was zero, implying that there were
617        other matches that did not fit. The correct "exactly full" value is now
618        returned.
620    6.  If a subpattern that was called recursively or as a subroutine contained
621        (*PRUNE) or any other control that caused it to give a non-standard return,
622        invalid errors such as "Error -26 (nested recursion at the same subject
623        position)" or even infinite loops could occur.
625    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
626        computing the minimum length on reaching *ACCEPT, and so ended up with the
627        wrong value of 1 rather than 0. Further investigation indicates that
628        computing a minimum subject length in the presence of *ACCEPT is difficult
629        (think back references, subroutine calls), and so I have changed the code
630        so that no minimum is registered for a pattern that contains *ACCEPT.
632    8.  If (*THEN) was present in the first (true) branch of a conditional group,
633        it was not handled as intended. [But see 16 below.]
635    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
636        Sheri Pierce.
638    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
639        the first byte in a match must be "a".
641    11. Change 17 for 8.13 increased the recursion depth for patterns like
642        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
643        pattern contains any instances of (*THEN). If it does not, the old
644        optimizations are restored. It would be nice to do this on a per-group
645        basis, but at the moment that is not feasible.
647    12. In some environments, the output of pcretest -C is CRLF terminated. This
648        broke RunTest's code that checks for the link size. A single white space
649        character after the value is now allowed for.
651    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
652        For "fr", it uses the Windows-specific input and output files.
654    14. If (*THEN) appeared in a group that was called recursively or as a
655        subroutine, it did not work as intended. [But see next item.]
657    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
658        pattern fragments (but not containing any | characters). If A and B are
659        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
660        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
661        D. In other words, Perl considers parentheses that do not contain any |
662        characters to be part of a surrounding alternative, whereas PCRE was
663        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
664        differently. PCRE now behaves in the same way as Perl, except in the case
665        of subroutine/recursion calls such as (?1) which have in any case always
666        been different (but PCRE had them first :-).
668    16. Related to 15 above: Perl does not treat the | in a conditional group as
669        creating alternatives. Such a group is treated in the same way as an
670        ordinary group without any | characters when processing (*THEN). PCRE has
671        been changed to match Perl's behaviour.
673    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
674        RunGrepTest script failed.
676    18. Change 22 for version 13 caused atomic groups to use more stack. This is
677        inevitable for groups that contain captures, but it can lead to a lot of
678        stack use in large patterns. The old behaviour has been restored for atomic
679        groups that do not contain any capturing parentheses.
681    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
682        suppress the check for a minimum subject length at run time. (If it was
683        given to pcre_exec() or pcre_dfa_exec() it did work.)
685    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
686        fail to work when decoding hex characters in data strings in EBCDIC
687        environments.
689    21. It appears that in at least one Mac OS environment, the isxdigit() function
690        is implemented as a macro that evaluates to its argument more than once,
691        contravening the C 90 Standard (I haven't checked a later standard). There
692        was an instance in pcretest which caused it to go wrong when processing
693        \x{...} escapes in subject strings. The has been rewritten to avoid using
694        things like p++ in the argument of isxdigit().
697    Version 8.13 16-Aug-2011
698    ------------------------
700    1.  The Unicode data tables have been updated to Unicode 6.0.0.
702    2.  Two minor typos in pcre_internal.h have been fixed.
704    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
705        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
706        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
708    4.  There were a number of related bugs in the code for matching backrefences
709        caselessly in UTF-8 mode when codes for the characters concerned were
710        different numbers of bytes. For example, U+023A and U+2C65 are an upper
711        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
712        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
713        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
714        2-byte code at the end of the subject (it thought there wasn't enough data
715        left).
717    5.  Comprehensive information about what went wrong is now returned by
718        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
719        as the output vector has at least 2 elements. The offset of the start of
720        the failing character and a reason code are placed in the vector.
722    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
723        now returned is for the first byte of the failing character, instead of the
724        last byte inspected. This is an incompatible change, but I hope it is small
725        enough not to be a problem. It makes the returned offset consistent with
726        pcre_exec() and pcre_dfa_exec().
728    7.  pcretest now gives a text phrase as well as the error number when
729        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
730        failure, the offset and reason code are output.
732    8.  When \R was used with a maximizing quantifier it failed to skip backwards
733        over a \r\n pair if the subsequent match failed. Instead, it just skipped
734        back over a single character (\n). This seems wrong (because it treated the
735        two characters as a single entity when going forwards), conflicts with the
736        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
737        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
738        has been changed.
740    9.  Some internal refactoring has changed the processing so that the handling
741        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
742        time (the PCRE_DOTALL option was changed this way some time ago: version
743        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
744        which was always a bit of a fudge. It also means that there is one less
745        argument for the match() function, which reduces its stack requirements
746        slightly. This change also fixes an incompatibility with Perl: the pattern
747        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
749    10. More internal refactoring has drastically reduced the number of recursive
750        calls to match() for possessively repeated groups such as (abc)++ when
751        using pcre_exec().
753    11. While implementing 10, a number of bugs in the handling of groups were
754        discovered and fixed:
756        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
757        (a|)*(?1) gave a compile-time internal error.
758        ((a|)+)+  did not notice that the outer group could match an empty string.
759        (^a|^)+   was not marked as anchored.
760        (.*a|.*)+ was not marked as matching at start or after a newline.
762    12. Yet more internal refactoring has removed another argument from the match()
763        function. Special calls to this function are now indicated by setting a
764        value in a variable in the "match data" data block.
766    13. Be more explicit in pcre_study() instead of relying on "default" for
767        opcodes that mean there is no starting character; this means that when new
768        ones are added and accidentally left out of pcre_study(), testing should
769        pick them up.
771    14. The -s option of pcretest has been documented for ages as being an old
772        synonym of -m (show memory usage). I have changed it to mean "force study
773        for every regex", that is, assume /S for every regex. This is similar to -i
774        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
775        using it. It makes it easier to run collections of tests with and without
776        study enabled, and thereby test pcre_study() more easily. All the standard
777        tests are now run with and without -s (but some patterns can be marked as
778        "never study" - see 20 below).
780    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
781        restoration of the capturing data to the outer values was not happening
782        correctly.
784    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
785        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
786        pattern had matched an empty string, and so incorrectly returned a no
787        match.
789    17. There was optimizing code for the last branch of non-capturing parentheses,
790        and also for the obeyed branch of a conditional subexpression, which used
791        tail recursion to cut down on stack usage. Unfortunately, now that there is
792        the possibility of (*THEN) occurring in these branches, tail recursion is
793        no longer possible because the return has to be checked for (*THEN). These
794        two optimizations have therefore been removed. [But see 8.20/11 above.]
796    18. If a pattern containing \R was studied, it was assumed that \R always
797        matched two bytes, thus causing the minimum subject length to be
798        incorrectly computed because \R can also match just one byte.
800    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
801        was incorrectly computed.
803    20. If /S is present twice on a test pattern in pcretest input, it now
804        *disables* studying, thereby overriding the use of -s on the command line
805        (see 14 above). This is necessary for one or two tests to keep the output
806        identical in both cases.
808    21. When (*ACCEPT) was used in an assertion that matched an empty string and
809        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
811    22. When an atomic group that contained a capturing parenthesis was
812        successfully matched, but the branch in which it appeared failed, the
813        capturing was not being forgotten if a higher numbered group was later
814        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
815        group 1 to "a", when in fact it should be unset. This applied to multi-
816        branched capturing and non-capturing groups, repeated or not, and also to
817        positive assertions (capturing in negative assertions does not happen
818        in PCRE) and also to nested atomic groups.
820    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
821        subject after a captured substring, to make it easier to tell which of a
822        number of identical substrings has been captured.
824    24. The way atomic groups are processed by pcre_exec() has been changed so that
825        if they are repeated, backtracking one repetition now resets captured
826        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
827        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
828        "aaa". Previously, it would have been "a". As part of this code
829        refactoring, the way recursive calls are handled has also been changed.
831    25. If an assertion condition captured any substrings, they were not passed
832        back unless some other capturing happened later. For example, if
833        (?(?=(a))a) was matched against "a", no capturing was returned.
835    26. When studying a pattern that contained subroutine calls or assertions,
836        the code for finding the minimum length of a possible match was handling
837        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
838        group 1 called group 2 while simultaneously a separate group 2 called group
839        1). A stack overflow occurred in this case. I have fixed this by limiting
840        the recursion depth to 10.
842    27. Updated RunTest.bat in the distribution to the version supplied by Tom
843        Fortmann. This supports explicit test numbers on the command line, and has
844        argument validation and error reporting.
846    28. An instance of \X with an unlimited repeat could fail if at any point the
847        first character it looked at was a mark character.
849    29. Some minor code refactoring concerning Unicode properties and scripts
850        should reduce the stack requirement of match() slightly.
852    30. Added the '=' option to pcretest to check the setting of unused capturing
853        slots at the end of the pattern, which are documented as being -1, but are
854        not included in the return count.
856    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
857        compiled something random. Now it gives a compile-time error (as does
858        Perl).
860    32. A *MARK encountered during the processing of a positive assertion is now
861        recorded and passed back (compatible with Perl).
863    33. If --only-matching or --colour was set on a pcregrep call whose pattern
864        had alternative anchored branches, the search for a second match in a line
865        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
866        matched the line "0102" twice. The same bug affected patterns that started
867        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
868        twice.
870    34. Previously, PCRE did not allow quantification of assertions. However, Perl
871        does, and because of capturing effects, quantifying parenthesized
872        assertions may at times be useful. Quantifiers are now allowed for
873        parenthesized assertions.
875    35. A minor code tidy in pcre_compile() when checking options for \R usage.
877    36. \g was being checked for fancy things in a character class, when it should
878        just be a literal "g".
880    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
881        appearance of a nested POSIX class supersedes an apparent external class.
882        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
883        unescaped square brackets may also appear as part of class names. For
884        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
885        more like Perl. (But see 8.20/1 above.)
887    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
888        was because it thought it was \N{name}, which is not supported).
890    39. Add minix to OS list not supporting the -S option in pcretest.
892    40. PCRE tries to detect cases of infinite recursion at compile time, but it
893        cannot analyze patterns in sufficient detail to catch mutual recursions
894        such as ((?1))((?2)). There is now a runtime test that gives an error if a
895        subgroup is called recursively as a subpattern for a second time at the
896        same position in the subject string. In previous releases this might have
897        been caught by the recursion limit, or it might have run out of stack.
899    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
900        happen only once. PCRE was, however incorrectly giving a compile time error
901        "recursive call could loop indefinitely" because it cannot analyze the
902        pattern in sufficient detail. The compile time test no longer happens when
903        PCRE is compiling a conditional subpattern, but actual runaway loops are
904        now caught at runtime (see 40 above).
906    42. It seems that Perl allows any characters other than a closing parenthesis
907        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
908        has been changed to be the same.
910    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
911        as not to get warnings when autogen.sh is called. Also changed
912        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
914    44. To help people who use pcregrep to scan files containing exceedingly long
915        lines, the following changes have been made:
917        (a) The default value of the buffer size parameter has been increased from
918            8K to 20K. (The actual buffer used is three times this size.)
920        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
921            PCRE is built.
923        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
924            to be set at run time.
926        (d) Numerical values in pcregrep options can be followed by K or M, for
927            example --buffer-size=50K.
929        (e) If a line being scanned overflows pcregrep's buffer, an error is now
930            given and the return code is set to 2.
932    45. Add a pointer to the latest mark to the callout data block.
934    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
935        partial match of an empty string instead of no match. This was specific to
936        the use of ".".
938    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
939        complete match instead of a partial match. This bug was dependent on both
940        the PCRE_UTF8 and PCRE_DOTALL options being set.
942    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
943        starting byte set, because \b was not being ignored.
946    Version 8.12 15-Jan-2011
947    ------------------------
949    1.  Fixed some typos in the markup of the man pages, and wrote a script that
950        checks for such things as part of the documentation building process.
952    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
953        --match-limit and --recursion-limit options (added for 8.11). In
954        particular, this made one of the standard tests fail. (The integer value
955        went into the wrong half of a long int.)
957    3.  If the --colour option was given to pcregrep with -v (invert match), it
958        did strange things, either producing crazy output, or crashing. It should,
959        of course, ignore a request for colour when reporting lines that do not
960        match.
962    4.  Another pcregrep bug caused similar problems if --colour was specified with
963        -M (multiline) and the pattern match finished with a line ending.
965    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
966        matched in multiline mode, the following line was shown as part of the
967        match. This seems wrong, so I have changed it.
969    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
970        the check for further matches in the same line (so they could be coloured)
971        to overrun the end of the current line. If another match was found, it was
972        incorrectly shown (and then shown again when found in the next line).
974    7.  If pcregrep was compiled under Windows, there was a reference to the
975        function pcregrep_exit() before it was defined. I am assuming this was
976        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
977        reported by a user. I've moved the definition above the reference.
980    Version 8.11 10-Dec-2010
981    ------------------------
983    1.  (*THEN) was not working properly if there were untried alternatives prior
984        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
985        backtracked to try for "b" instead of moving to the next alternative branch
986        at the same level (in this case, to look for "c"). The Perl documentation
987        is clear that when (*THEN) is backtracked onto, it goes to the "next
988        alternative in the innermost enclosing group".
990    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
991        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
992        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
993        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
994        (*THEN).
996    3.  If \s appeared in a character class, it removed the VT character from
997        the class, even if it had been included by some previous item, for example
998        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
999        of \s, but is part of the POSIX "space" class.)
1001    4.  A partial match never returns an empty string (because you can always
1002        match an empty string at the end of the subject); however the checking for
1003        an empty string was starting at the "start of match" point. This has been
1004        changed to the "earliest inspected character" point, because the returned
1005        data for a partial match starts at this character. This means that, for
1006        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1007        (previously it gave "no match").
1009    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1010        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1011        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1012        has an implication that the given string is incomplete (because a partial
1013        match is preferred over a full match). For this reason, these items now
1014        give a partial match in this situation. [Aside: previously, the one case
1015        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1016        match rather than a full match, which was wrong by the old rules, but is
1017        now correct.]
1019    6.  There was a bug in the handling of #-introduced comments, recognized when
1020        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1021        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1022        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1023        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1024        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1025        places in pcre_compile().
1027    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1028        comments when looking ahead for named forward references to subpatterns,
1029        the only newline sequence it recognized was NL. It now handles newlines
1030        according to the set newline convention.
1032    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1033        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1034        cater for a lack of strerror(). These oversights have been fixed.
1036    9.  Added --match-limit and --recursion-limit to pcregrep.
1038    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1040    11. When the -o option was used, pcregrep was setting a return code of 1, even
1041        when matches were found, and --line-buffered was not being honoured.
1043    12. Added an optional parentheses number to the -o and --only-matching options
1044        of pcregrep.
1046    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1047        can match an empty string. The code to do it in pcretest and pcredemo
1048        needed fixing:
1050        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1051            only one byte after an empty string match just before CRLF (this case
1052            just got forgotten; "any" and "anycrlf" were OK).
1054        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1055            mode when an empty string match preceded an ASCII character followed by
1056            a non-ASCII character. (The code for advancing by one character rather
1057            than one byte was nonsense.)
1059        (c) The pcredemo.c sample program did not have any code at all to handle
1060            the cases when CRLF is a valid newline sequence.
1062    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1063        as a starting offset was within the subject string. There is now a new
1064        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1065        negative or greater than the length of the string. In order to test this,
1066        pcretest is extended to allow the setting of negative starting offsets.
1068    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1069        starting offset points to the beginning of a UTF-8 character was
1070        unnecessarily clumsy. I tidied it up.
1072    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1073        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1075    17. Nobody had reported that the --include_dir option, which was added in
1076        release 7.7 should have been called --include-dir (hyphen, not underscore)
1077        for compatibility with GNU grep. I have changed it to --include-dir, but
1078        left --include_dir as an undocumented synonym, and the same for
1079        --exclude-dir, though that is not available in GNU grep, at least as of
1080        release 2.5.4.
1082    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1083        characters from a string of bytes) have been redefined so as not to use
1084        loops, in order to improve performance in some environments. At the same
1085        time, I abstracted some of the common code into auxiliary macros to save
1086        repetition (this should not affect the compiled code).
1088    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1089        compile-time error is now given if \c is not followed by an ASCII
1090        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1091        different, and any byte value is allowed.)
1093    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1094        START_OPTIMIZE option, which is now allowed at compile time - but just
1095        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1096        to pcregrep and other applications that have no direct access to PCRE
1097        options. The new /Y option in pcretest sets this option when calling
1098        pcre_compile().
1100    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1101        back references. Groups containing recursive back references were forced to
1102        be atomic by that change, but in the case of named groups, the amount of
1103        memory required was incorrectly computed, leading to "Failed: internal
1104        error: code overflow". This has been fixed.
1106    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1107        pcretest.c, to avoid build problems in some Borland environments.
1110    Version 8.10 25-Jun-2010
1111    ------------------------
1113    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1114        THEN.
1116    2.  (*ACCEPT) was not working when inside an atomic group.
1118    3.  Inside a character class, \B is treated as a literal by default, but
1119        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1120        causes the error). The code is unchanged, but I tidied the documentation.
1122    4.  Inside a character class, PCRE always treated \R and \X as literals,
1123        whereas Perl faults them if its -w option is set. I have changed PCRE so
1124        that it faults them when PCRE_EXTRA is set.
1126    5.  Added support for \N, which always matches any character other than
1127        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1129    6.  When compiling pcregrep with newer versions of gcc which may have
1130        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1131        declared with attribute warn_unused_result" were given. Just casting the
1132        result to (void) does not stop the warnings; a more elaborate fudge is
1133        needed. I've used a macro to implement this.
1135    7.  Minor change to pcretest.c to avoid a compiler warning.
1137    8.  Added four artifical Unicode properties to help with an option to make
1138        \s etc use properties (see next item). The new properties are: Xan
1139        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1141    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1142        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1143        this option. Modified pcretest to add /W to test this facility. Added
1144        REG_UCP to make it available via the POSIX interface.
1146    10. Added --line-buffered to pcregrep.
1148    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1149        studied, and the match started with a letter with a code point greater than
1150        127 whose first byte was different to the first byte of the other case of
1151        the letter, the other case of this starting letter was not recognized
1152        (#976).
1154    12. If a pattern that was studied started with a repeated Unicode property
1155        test, for example, \p{Nd}+, there was the theoretical possibility of
1156        setting up an incorrect bitmap of starting bytes, but fortunately it could
1157        not have actually happened in practice until change 8 above was made (it
1158        added property types that matched character-matching opcodes).
1160    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1161        possible starting bytes for non-anchored patterns.
1163    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1164        \R, and also a number of cases that involve Unicode properties, both
1165        explicit and implicit when PCRE_UCP is set.
1167    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1168        input, it could crash or give wrong results if characters with values
1169        greater than 0xc0 were present in the subject string. (Detail: it assumed
1170        UTF-8 input when processing these items.)
1172    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1173        size_t is 64-bit (#991).
1175    17. Added a check for running out of memory when PCRE is compiled with
1176        --disable-stack-for-recursion (#990).
1178    18. If the last data line in a file for pcretest does not have a newline on
1179        the end, a newline was missing in the output.
1181    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1182        less than 128) in its various bitmaps. However, there is a facility for
1183        generating tables according to the current locale when PCRE is compiled. It
1184        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1185        characters, are recognized by isspace() and therefore were getting set in
1186        these tables, and indeed these tables seem to approximate to ISO 8859. This
1187        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1188        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1189        which of course cannot start UTF-8 characters. I have changed the code so
1190        that only real ASCII characters (less than 128) and the correct starting
1191        bytes for UTF-8 encodings are set for characters greater than 127 when in
1192        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1193        altogether.)
1195    20. Added the /T option to pcretest so as to be able to run tests with non-
1196        standard character tables, thus making it possible to include the tests
1197        used for 19 above in the standard set of tests.
1199    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1200        reference to a subpattern the other side of a comment that contains an
1201        opening parenthesis caused either an internal compiling error, or a
1202        reference to the wrong subpattern.
1205    Version 8.02 19-Mar-2010
1206    ------------------------
1208    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1210    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1211        configured.
1213    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1214        original author of that file, following a query about its status.
1216    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1217        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1219    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1220        quantifier applied to a forward-referencing subroutine call, could compile
1221        incorrect code or give the error "internal error: previously-checked
1222        referenced subpattern not found".
1224    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1225        variables to point to external functions. For these systems, therefore,
1226        pcre_malloc etc. are now initialized to local functions that call the
1227        relevant global functions.
1229    7.  There were two entries missing in the vectors called coptable and poptable
1230        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1231        I've fixed the data, and added a kludgy way of testing at compile time that
1232        the lengths are correct (equal to the number of opcodes).
1234    8.  Following on from 7, I added a similar kludge to check the length of the
1235        eint vector in pcreposix.c.
1237    9.  Error texts for pcre_compile() are held as one long string to avoid too
1238        much relocation at load time. To find a text, the string is searched,
1239        counting zeros. There was no check for running off the end of the string,
1240        which could happen if a new error number was added without updating the
1241        string.
1243    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1245    11. \K was not working if it appeared in an atomic group or in a group that
1246        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1247        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1248        the assertion is positive, but not if it is negative.
1250    12. Change 11 fortuitously reduced the size of the stack frame used in the
1251        "match()" function of pcre_exec.c by one pointer. Forthcoming
1252        implementation of support for (*MARK) will need an extra pointer on the
1253        stack; I have reserved it now, so that the stack frame size does not
1254        decrease.
1256    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1257        item in branch that calls a recursion is a subroutine call - as in the
1258        second branch in the above example - was incorrectly given the compile-
1259        time error "recursive call could loop indefinitely" because pcre_compile()
1260        was not correctly checking the subroutine for matching a non-empty string.
1262    14. The checks for overrunning compiling workspace could trigger after an
1263        overrun had occurred. This is a "should never occur" error, but it can be
1264        triggered by pathological patterns such as hundreds of nested parentheses.
1265        The checks now trigger 100 bytes before the end of the workspace.
1267    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1270    Version 8.01 19-Jan-2010
1271    ------------------------
1273    1.  If a pattern contained a conditional subpattern with only one branch (in
1274        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1275        computed the wrong minimum data length (which is of course zero for such
1276        subpatterns). This could cause incorrect "no match" results.
1278    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1279        the pattern is reset in the first branch, pcre_compile() failed with
1280        "internal error: code overflow at offset...". This happened only when
1281        the reset was to the original external option setting. (An optimization
1282        abstracts leading options settings into an external setting, which was the
1283        cause of this.)
1285    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1286        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1287        assertion pattern did not match (meaning that the assertion was true), it
1288        was incorrectly treated as false if the SKIP had been reached during the
1289        matching. This also applied to assertions used as conditions.
1291    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1292        assertion subpattern, including such a pattern used as a condition,
1293        unpredictable results occurred, instead of the error return
1296    5.  The C++ GlobalReplace function was not working like Perl for the special
1297        situation when an empty string is matched. It now does the fancy magic
1298        stuff that is necessary.
1300    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1301        removed. (These were left over from very, very early versions of PCRE.)
1303    7.  Some cosmetic changes to the code to make life easier when compiling it
1304        as part of something else:
1306        (a) Change DEBUG to PCRE_DEBUG.
1308        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1309            called "current" as "current_branch", to prevent a collision with the
1310            Linux macro when compiled as a kernel module.
1312        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1313            prevent a collision with the Linux macro when compiled as a kernel
1314            module.
1316    8.  In pcre_compile() there are some checks for integer overflows that used to
1317        cast potentially large values to (double). This has been changed to that
1318        when building, a check for int64_t is made, and if it is found, it is used
1319        instead, thus avoiding the use of floating point arithmetic. (There is no
1320        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1321        double.
1323    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1324        2005 (difference between two addresses compared to an unsigned value).
1326    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1327        custom one, because of the following reported problem in Windows:
1329          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1330              under Win32.
1331          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1332              therefore missing the function definition.
1333          - The compiler thus generates a "C" signature for the test function.
1334          - The linker fails to find the "C" function.
1335          - PCRE fails to configure if asked to do so against libbz2.
1337    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1338        messages were output:
1340          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1341          rerunning libtoolize, to keep the correct libtool macros in-tree.
1342          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1344        I have done both of these things.
1346    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1347        most of the time, it *can* run out if it is given a pattern that contains a
1348        runaway infinite recursion. I updated the discussion in the pcrestack man
1349        page.
1351    13. Now that we have gone to the x.xx style of version numbers, the minor
1352        version may start with zero. Using 08 or 09 is a bad idea because users
1353        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1354        interpreted as invalid octal numbers. I've updated the previous comment in
1355        configure.ac, and also added a check that gives an error if 08 or 09 are
1356        used.
1358    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1359        causing partial matching to fail when the end of the subject matched \W
1360        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1362    15. There were some discrepancies between the declarations in pcre_internal.h
1363        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1364        their definitions. The declarations used "const uschar *" and the
1365        definitions used USPTR. Even though USPTR is normally defined as "const
1366        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1367        reported that: "This difference in casting confuses some C++ compilers, for
1368        example, SunCC recognizes above declarations as different functions and
1369        generates broken code for hbpcre." I have changed the declarations to use
1370        USPTR.
1372    16. GNU libtool is named differently on some systems. The autogen.sh script now
1373        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1374        (FreeBSD).
1376    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1377        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1378        comment: "Figure out how to create a longlong from a string: strtoll and
1379        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1380        instance, but it only takes 2 args instead of 3!"
1382    18. A subtle bug concerned with back references has been fixed by a change of
1383        specification, with a corresponding code fix. A pattern such as
1384        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1385        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1386        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1387        same bug. Such groups have to be quantified to be useful, or contained
1388        inside another quantified group. (If there's no repetition, the reference
1389        can never match.) The problem arises because, having left the group and
1390        moved on to the rest of the pattern, a later failure that backtracks into
1391        the group uses the captured value from the final iteration of the group
1392        rather than the correct earlier one. I have fixed this in PCRE by forcing
1393        any group that contains a reference to itself to be an atomic group; that
1394        is, there cannot be any backtracking into it once it has completed. This is
1395        similar to recursive and subroutine calls.
1398    Version 8.00 19-Oct-09
1399    ----------------------
1401    1.  The table for translating pcre_compile() error codes into POSIX error codes
1402        was out-of-date, and there was no check on the pcre_compile() error code
1403        being within the table. This could lead to an OK return being given in
1404        error.
1406    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1407        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1408        in a Windows environment.
1410    3.  The pcregrep --count option prints the count for each file even when it is
1411        zero, as does GNU grep. However, pcregrep was also printing all files when
1412        --files-with-matches was added. Now, when both options are given, it prints
1413        counts only for those files that have at least one match. (GNU grep just
1414        prints the file name in this circumstance, but including the count seems
1415        more useful - otherwise, why use --count?) Also ensured that the
1416        combination -clh just lists non-zero counts, with no names.
1418    4.  The long form of the pcregrep -F option was incorrectly implemented as
1419        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1420        but it seems right to fix it, and I didn't think it was worth preserving
1421        the old behaviour.
1423    5.  The command line items --regex=pattern and --regexp=pattern were not
1424        recognized by pcregrep, which required --regex pattern or --regexp pattern
1425        (with a space rather than an '='). The man page documented the '=' forms,
1426        which are compatible with GNU grep; these now work.
1428    6.  No libpcreposix.pc file was created for pkg-config; there was just
1429        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1431    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1432        when UCP support is not needed, by modifying the Python script that
1433        generates it from Unicode data files. This should not matter if the module
1434        is correctly used as a library, but I received one complaint about 50K of
1435        unwanted data. My guess is that the person linked everything into his
1436        program rather than using a library. Anyway, it does no harm.
1438    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1439        was a minimum greater than 1 for a wide character in a possessive
1440        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1441        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1442        character. Chaos in the form of incorrect output or a compiling loop could
1443        result.
1445    9.  The restrictions on what a pattern can contain when partial matching is
1446        requested for pcre_exec() have been removed. All patterns can now be
1447        partially matched by this function. In addition, if there are at least two
1448        slots in the offset vector, the offset of the earliest inspected character
1449        for the match and the offset of the end of the subject are set in them when
1450        PCRE_ERROR_PARTIAL is returned.
1452    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1453        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1454        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1455        and may be more useful for multi-segment matching.
1457    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1458        used to be given if ever the end of the subject was reached; now it is
1459        given only if matching could not proceed because another character was
1460        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1461        string "Z", which now yields "no match" instead of "partial match". In the
1462        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1463        final character ended with (*FAIL).
1465    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1466        if the pattern had a "must contain" character that was already found in the
1467        earlier partial match, unless partial matching was again requested. For
1468        example, with the pattern /dog.(body)?/, the "must contain" character is
1469        "g". If the first part-match was for the string "dog", restarting with
1470        "sbody" failed. This bug has been fixed.
1472    13. The string returned by pcre_dfa_exec() after a partial match has been
1473        changed so that it starts at the first inspected character rather than the
1474        first character of the match. This makes a difference only if the pattern
1475        starts with a lookbehind assertion or \b or \B (\K is not supported by
1476        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1477        matching functions compatible, and I think it's the right thing to do.
1479    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1480        so that the demonstration program is easily available in environments where
1481        PCRE has not been installed from source.
1483    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1484        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1485        library.
1487    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1488        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1489        is not the first non-POSIX option to be added. Clearly some people find
1490        these options useful.
1492    17. If a caller to the POSIX matching function regexec() passes a non-zero
1493        value for nmatch with a NULL value for pmatch, the value of
1494        nmatch is forced to zero.
1496    18. RunGrepTest did not have a test for the availability of the -u option of
1497        the diff command, as RunTest does. It now checks in the same way as
1498        RunTest, and also checks for the -b option.
1500    19. If an odd number of negated classes containing just a single character
1501        interposed, within parentheses, between a forward reference to a named
1502        subpattern and the definition of the subpattern, compilation crashed with
1503        an internal error, complaining that it could not find the referenced
1504        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1505        [The bug was that it was starting one character too far in when skipping
1506        over the character class, thus treating the ] as data rather than
1507        terminating the class. This meant it could skip too much.]
1509    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1510        /g option in pcretest when the pattern contains \K, which makes it possible
1511        to have an empty string match not at the start, even when the pattern is
1512        anchored. Updated pcretest and pcredemo to use this option.
1514    21. If the maximum number of capturing subpatterns in a recursion was greater
1515        than the maximum at the outer level, the higher number was returned, but
1516        with unset values at the outer level. The correct (outer level) value is
1517        now given.
1519    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1520        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1521        make it do so. The string so far is captured, making this feature
1522        compatible with Perl.
1524    23. The tests have been re-organized, adding tests 11 and 12, to make it
1525        possible to check the Perl 5.10 features against Perl 5.10.
1527    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1528        pattern matches a fixed length string. PCRE did not allow this; now it
1529        does. Neither allows recursion.
1531    25. I finally figured out how to implement a request to provide the minimum
1532        length of subject string that was needed in order to match a given pattern.
1533        (It was back references and recursion that I had previously got hung up
1534        on.) This code has now been added to pcre_study(); it finds a lower bound
1535        to the length of subject needed. It is not necessarily the greatest lower
1536        bound, but using it to avoid searching strings that are too short does give
1537        some useful speed-ups. The value is available to calling programs via
1538        pcre_fullinfo().
1540    26. While implementing 25, I discovered to my embarrassment that pcretest had
1541        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1542        study optimizations had never been tested with that matching function.
1543        Oops. What is worse, even when it was passed study data, there was a bug in
1544        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1545        were also very few tests of studied patterns with pcre_dfa_exec().
1547    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1548        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1549        on the other side of the coin, they are no longer allowed to have different
1550        names, because these cannot be distinguished in PCRE, and this has caused
1551        confusion. (This is a difference from Perl.)
1553    28. When duplicate subpattern names are present (necessarily with different
1554        numbers, as required by 27 above), and a test is made by name in a
1555        conditional pattern, either for a subpattern having been matched, or for
1556        recursion in such a pattern, all the associated numbered subpatterns are
1557        tested, and the overall condition is true if the condition is true for any
1558        one of them. This is the way Perl works, and is also more like the way
1559        testing by number works.
1562    Version 7.9 11-Apr-09
1563    ---------------------
1565    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1566        (pcretest), all targets were linked against these libraries. This included
1567        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1568        libraries. This caused unwanted dependencies to be created. This problem
1569        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1570        pcretest is linked with readline.
1572    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1573        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1574        moved outside it again, because FALSE and TRUE are already defined in AIX,
1575        but BOOL is not.
1577    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1578        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1580    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1581        hyphens) following file names and line numbers when outputting matching
1582        lines. This is not true; no spaces are inserted. I have also clarified the
1583        wording for the --colour (or --color) option.
1585    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1586        was not coloured; this is different to GNU grep, so I have changed it to be
1587        the same.
1589    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1590        each matching line was coloured. Now it goes on to look for further matches
1591        of any of the test patterns, which is the same behaviour as GNU grep.
1593    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1594        doesn't make sense to accept an empty string match in pcregrep, so I have
1595        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1596        seems to be how GNU grep behaves.
1598    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1599        start or after a newline", because the conditional assertion was not being
1600        correctly handled. The rule now is that both the assertion and what follows
1601        in the first alternative must satisfy the test.
1603    9.  If auto-callout was enabled in a pattern with a conditional group whose
1604        condition was an assertion, PCRE could crash during matching, both with
1605        pcre_exec() and pcre_dfa_exec().
1607    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1608        used for matching.
1610    11. Unicode property support in character classes was not working for
1611        characters (bytes) greater than 127 when not in UTF-8 mode.
1613    12. Added the -M command line option to pcretest.
1615    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1617    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1619    16. Added comments and documentation about mis-use of no_arg in the C++
1620        wrapper.
1622    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1623        from Martin Jerabek that uses macro names for all relevant character and
1624        string constants.
1626    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1627        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1628        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1629        these, but not everybody uses configure.
1631    19. A conditional group that had only one branch was not being correctly
1632        recognized as an item that could match an empty string. This meant that an
1633        enclosing group might also not be so recognized, causing infinite looping
1634        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1635        with the subject "ab", where knowledge that the repeated group can match
1636        nothing is needed in order to break the loop.
1638    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1639        exec(), but without supplying a callout function, matching went wrong.
1641    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1642        leak if the size of the offset vector was greater than 30. When the vector
1643        is smaller, the saved offsets during recursion go onto a local stack
1644        vector, but for larger vectors malloc() is used. It was failing to free
1645        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1646        error, in fact).
1648    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1649        heapframe that is used only when UTF-8 support is enabled. This caused no
1650        problem, but was untidy.
1652    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1653        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1654        included within another project.
1656    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1657        slightly modified by me:
1659          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1660              not building pcregrep.
1662          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1663              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1665    25. Forward references, both numeric and by name, in patterns that made use of
1666        duplicate group numbers, could behave incorrectly or give incorrect errors,
1667        because when scanning forward to find the reference group, PCRE was not
1668        taking into account the duplicate group numbers. A pattern such as
1669        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1671    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1672        the feature of a custom pointer more persuasive (as requested by a user).
1674    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1675        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1676        pre-defined.
1678    28. Added support for (*UTF8) at the start of a pattern.
1680    29. Arrange for flags added by the "release type" setting in CMake to be shown
1681        in the configuration summary.
1684    Version 7.8 05-Sep-08
1685    ---------------------
1687    1.  Replaced UCP searching code with optimized version as implemented for Ad
1688        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1689        stage table and inline lookup instead of a function, giving speed ups of 2
1690        to 5 times on some simple patterns that I tested. Permission was given to
1691        distribute the MultiStage2.py script that generates the tables (it's not in
1692        the tarball, but is in the Subversion repository).
1694    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1695        scripts.
1697    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1698        a group with a zero qualifier. The result of the study could be incorrect,
1699        or the function might crash, depending on the pattern.
1701    4.  Caseless matching was not working for non-ASCII characters in back
1702        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1703        It now works when Unicode Property Support is available.
1705    5.  In pcretest, an escape such as \x{de} in the data was always generating
1706        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1707        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1708        truncation.
1710    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1712    7.  Added two (int) casts to pcregrep when printing the difference of two
1713        pointers, in case they are 64-bit values.
1715    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1716        test 2 if it fails.
1718    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1719        and a #define of that name to empty if it is not externally set. This is to
1720        allow users of MSVC to set it if necessary.
1722    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1723        the convenience functions in the pcre_get.c source file.
1725    11. An option change at the start of a pattern that had top-level alternatives
1726        could cause overwriting and/or a crash. This command provoked a crash in
1727        some environments:
1729          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1731        This potential security problem was recorded as CVE-2008-2371.
1733    12. For a pattern where the match had to start at the beginning or immediately
1734        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1735        pcre_dfa_exec() could read past the end of the passed subject if there was
1736        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1737        pcretest so that it places the subject at the end of its malloc-ed buffer.
1739    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1740        exec() might read past the end of the data buffer in UTF-8 mode.
1742    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1743        the data contained the byte 0x85 as part of a UTF-8 character within its
1744        first line. This applied both to normal and DFA matching.
1746    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1747        /^[^d]*?$/8 failed to match "abc".
1749    16. Added a missing copyright notice to pcrecpp_internal.h.
1751    17. Make it more clear in the documentation that values returned from
1752        pcre_exec() in ovector are byte offsets, not character counts.
1754    18. Tidied a few places to stop certain compilers from issuing warnings.
1756    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1757        supplied by Stefan Weber. I made a further small update for 7.8 because
1758        there is a change of source arrangements: the pcre_searchfuncs.c module is
1759        replaced by pcre_ucd.c.
1762    Version 7.7 07-May-08
1763    ---------------------
1765    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1766        a string to a long long, pretend we don't even have a long long." This is
1767        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1769    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1770        pre-7.6 versions, which defined a global no_arg variable instead of putting
1771        it in the RE class. (See also #8 below.)
1773    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1774        Lopes.
1776    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1778        (1) The include/exclude patterns were being applied to the whole pathnames
1779            of files, instead of just to the final components.
1781        (2) If there was more than one level of directory, the subdirectories were
1782            skipped unless they satisfied the include/exclude conditions. This is
1783            inconsistent with GNU grep (and could even be seen as contrary to the
1784            pcregrep specification - which I improved to make it absolutely clear).
1785            The action now is always to scan all levels of directory, and just
1786            apply the include/exclude patterns to regular files.
1788    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1789        --exclude_dir in the tests to avoid scanning .svn directories.
1791    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1792        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1793        doesn't support NULs in patterns.
1795    7.  Added some missing "const"s to declarations of static tables in
1796        pcre_compile.c and pcre_dfa_exec.c.
1798    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1799        caused by fix #2  above. (Subsequently also a second patch to fix the
1800        first patch. And a third patch - this was a messy problem.)
1802    9.  Applied Craig's patch to remove the use of push_back().
1804    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1805        matching function regexec().
1807    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1808        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1809        references. PCRE supports relative numbers with this syntax (I don't think
1810        Oniguruma does).
1812    12. Previously, a group with a zero repeat such as (...){0} was completely
1813        omitted from the compiled regex. However, this means that if the group
1814        was called as a subroutine from elsewhere in the pattern, things went wrong
1815        (an internal error was given). Such groups are now left in the compiled
1816        pattern, with a new opcode that causes them to be skipped at execution
1817        time.
1819    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1820        to the way PCRE behaves:
1822        (a) A lone ] character is dis-allowed (Perl treats it as data).
1824        (b) A back reference to an unmatched subpattern matches an empty string
1825            (Perl fails the current match path).
1827        (c) A data ] in a character class must be notated as \] because if the
1828            first data character in a class is ], it defines an empty class. (In
1829            Perl it is not possible to have an empty class.) The empty class []
1830            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1831            The negative empty class [^] matches any one character, independently
1832            of the DOTALL setting.
1834    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1835        non-existent subpattern following a character class starting with ']' and
1836        containing () gave an internal compiling error instead of "reference to
1837        non-existent subpattern". Fortunately, when the pattern did exist, the
1838        compiled code was correct. (When scanning forwards to check for the
1839        existencd of the subpattern, it was treating the data ']' as terminating
1840        the class, so got the count wrong. When actually compiling, the reference
1841        was subsequently set up correctly.)
1843    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1844        it was being rejected as not supported by pcre_dfa_exec(), even though
1845        other assertions are supported. I have made pcre_dfa_exec() support
1846        (*FAIL).
1848    16. The implementation of 13c above involved the invention of a new opcode,
1849        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1850        cannot be changed at match time, I realized I could make a small
1851        improvement to matching performance by compiling OP_ALLANY instead of
1852        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1853        on the OP_ANY path.
1855    17. Compiling pcretest on Windows with readline support failed without the
1856        following two fixes: (1) Make the unistd.h include conditional on
1857        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1859    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1860        ncurses library to be included for pcretest when ReadLine support is
1861        requested, but also to allow for it to be overridden. This patch came from
1862        Daniel Bergström.
1864    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1865        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1866        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1867        spotting this.
1870    Version 7.6 28-Jan-08
1871    ---------------------
1873    1.  A character class containing a very large number of characters with
1874        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1875        overflow.
1877    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1878        HAVE_LONG_LONG is not defined.
1880    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1881        bring them up to date and include new features. This patch includes:
1883        - Fixed PH's badly added libz and libbz2 support.
1884        - Fixed a problem with static linking.
1885        - Added pcredemo. [But later removed - see 7 below.]
1886        - Fixed dftables problem and added an option.
1887        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1888            HAVE_LONG_LONG.
1889        - Added readline support for pcretest.
1890        - Added an listing of the option settings after cmake has run.
1892    4.  A user submitted a patch to Makefile that makes it easy to create
1893        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1894        Makefile.am that cause it to include this special target, without
1895        affecting anything else. Note that the same mingw target plus all
1896        the other distribution libraries and programs are now supported
1897        when configuring with CMake (see 6 below) instead of with
1898        Configure/Make.
1900    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1901        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1902        exported in the Windows port". It has not yet been confirmed that the patch
1903        solves the problem, but it does no harm.
1905    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1906        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1907        with CMake, and also correct the comment about stack recursion.
1909    7.  Remove the automatic building of pcredemo from the ./configure system and
1910        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1911        of a program that users should build themselves after PCRE is installed, so
1912        building it automatically is not really right. What is more, it gave
1913        trouble in some build environments.
1915    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1918    Version 7.5 10-Jan-08
1919    ---------------------
1921    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1922        values in parens when parsing an RE using the C++ wrapper."
1924    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1925        Characters greater than 255 were excluded from the class instead of being
1926        included.
1928    3.  The same bug as (2) above applied to negated POSIX classes such as
1929        [:^space:].
1931    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1932        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1933        I have changed it.
1935    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1936        first named subpattern) and a construct such as (?&a) would reference the
1937        first named subpattern whose name started with "a" (in other words, the
1938        length check was missing). Both these problems are fixed. "Subpattern name
1939        expected" is now given for (?&) (a zero-length name), and this patch also
1940        makes it give the same error for \k'' (previously it complained that that
1941        was a reference to a non-existent subpattern).
1943    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1944        this is right because (?- can be followed by option settings as well as by
1945        digits. I have, however, made the messages clearer.
1947    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1948        than the number used in the conditional) now cause a compile-time error.
1949        This is actually not compatible with Perl, which accepts such patterns, but
1950        treats the conditional as always being FALSE (as PCRE used to), but it
1951        seems to me that giving a diagnostic is better.
1953    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1954        and messages.
1956    9.  Fix two occurrences of "backslash" in comments that should have been
1957        "backspace".
1959    10. Remove two redundant lines of code that can never be obeyed (their function
1960        was moved elsewhere).
1962    11. The program that makes PCRE's Unicode character property table had a bug
1963        which caused it to generate incorrect table entries for sequences of
1964        characters that have the same character type, but are in different scripts.
1965        It amalgamated them into a single range, with the script of the first of
1966        them. In other words, some characters were in the wrong script. There were
1967        thirteen such cases, affecting characters in the following ranges:
1969          U+002b0 - U+002c1
1970          U+0060c - U+0060d
1971          U+0061e - U+00612
1972          U+0064b - U+0065e
1973          U+0074d - U+0076d
1974          U+01800 - U+01805
1975          U+01d00 - U+01d77
1976          U+01d9b - U+01dbf
1977          U+0200b - U+0200f
1978          U+030fc - U+030fe
1979          U+03260 - U+0327f
1980          U+0fb46 - U+0fbb1
1981          U+10450 - U+1049d
1983    12. The -o option (show only the matching part of a line) for pcregrep was not
1984        compatible with GNU grep in that, if there was more than one match in a
1985        line, it showed only the first of them. It now behaves in the same way as
1986        GNU grep.
1988    13. If the -o and -v options were combined for pcregrep, it printed a blank
1989        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1990        does the same. The return code can be used to tell if there were any
1991        non-matching lines.
1993    14. Added --file-offsets and --line-offsets to pcregrep.
1995    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1996        infinitely looping recursion. The bug was that positive lookaheads were not
1997        being skipped when checking for a possible empty match (negative lookaheads
1998        and both kinds of lookbehind were skipped).
2000    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
2001        inclusion of <windows.h> to before rather than after the definition of
2002        INVALID_FILE_ATTRIBUTES (patch from David Byron).
2004    17. Specifying a possessive quantifier with a specific limit for a Unicode
2005        character property caused pcre_compile() to compile bad code, which led at
2006        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
2007        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
2008        caused the error; without that there was no problem.
2010    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
2012    19. Added --enable-pcretest-libreadline.
2014    20. In pcrecpp.cc, the variable 'count' was incremented twice in
2015        RE::GlobalReplace(). As a result, the number of replacements returned was
2016        double what it should be. I removed one of the increments, but Craig sent a
2017        later patch that removed the other one (the right fix) and added unit tests
2018        that check the return values (which was not done before).
2020    21. Several CMake things:
2022        (1) Arranged that, when cmake is used on Unix, the libraries end up with
2023            the names libpcre and libpcreposix, not just pcre and pcreposix.
2025        (2) The above change means that pcretest and pcregrep are now correctly
2026            linked with the newly-built libraries, not previously installed ones.
2030    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
2031        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
2032        UTF-8 newline character). The key issue is that the pattern starts .*;
2033        this means that the match must be either at the beginning, or after a
2034        newline. The bug was in the code for advancing after a failed match and
2035        checking that the new position followed a newline. It was not taking
2036        account of UTF-8 characters correctly.
2038    23. PCRE was behaving differently from Perl in the way it recognized POSIX
2039        character classes. PCRE was not treating the sequence [:...:] as a
2040        character class unless the ... were all letters. Perl, however, seems to
2041        allow any characters between [: and :], though of course it rejects as
2042        unknown any "names" that contain non-letters, because all the known class
2043        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
2044        for example, whereas PCRE did not - it did not recognize a POSIX character
2045        class. This seemed a bit dangerous, so the code has been changed to be
2046        closer to Perl. The behaviour is not identical to Perl, because PCRE will
2047        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
2048        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
2049        Perl does, and where it didn't before.
2051    24. Rewrite so as to remove the single use of %n from pcregrep because in some
2052        Windows environments %n is disabled by default.
2055    Version 7.4 21-Sep-07
2056    ---------------------
2058    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
2059        means that a class such as [\s] counted as "explicit reference to CR or
2060        LF". That isn't really right - the whole point of the change was to try to
2061        help when there was an actual mention of one of the two characters. So now
2062        the change happens only if \r or \n (or a literal CR or LF) character is
2063        encountered.
2065    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
2066        of both had grown to the point where there were only 3 bits left.
2067        Fortunately, there was spare space in the data structure, and so I have
2068        moved the internal flags into a new 16-bit field to free up more option
2069        bits.
2071    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
2072        but did not set the internal JCHANGED flag - either of these is enough to
2073        control the way the "get" function works - but the PCRE_INFO_JCHANGED
2074        facility is supposed to tell if (?J) was ever used, so now (?J) at the
2075        start sets both bits.
2077    4.  Added options (at build time, compile time, exec time) to change \R from
2078        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
2080    5.  doc/pcresyntax.html was missing from the distribution.
2082    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
2083        compatibility, even though it is no longer used.
2085    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
2086        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
2087        windows.h file is present (where different names are used). [This was
2088        reversed later after testing - see 16 below.]
2090    8.  Changed all #include <config.h> to #include "config.h". There were also
2091        some further <pcre.h> cases that I changed to "pcre.h".
2093    9.  When pcregrep was used with the --colour option, it missed the line ending
2094        sequence off the lines that it output.
2096    10. It was pointed out to me that arrays of string pointers cause lots of
2097        relocations when a shared library is dynamically loaded. A technique of
2098        using a single long string with a table of offsets can drastically reduce
2099        these. I have refactored PCRE in four places to do this. The result is
2100        dramatic:
2102          Originally:                          290
2103          After changing UCP table:            187
2104          After changing error message table:   43
2105          After changing table of "verbs"       36
2106          After changing table of Posix names   22
2108        Thanks to the folks working on Gregex for glib for this insight.
2110    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
2111        unicode-properties was also set.
2113    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
2115    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
2116        checked only for CRLF.
2118    14. Added casts to pcretest.c to avoid compiler warnings.
2120    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
2122    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
2123        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2124        entirely. This removes changes made in 7 above.
2126    17. The CMake files have been updated, and there is now more information about
2127        building with CMake in the NON-UNIX-USE document.
2130    Version 7.3 28-Aug-07
2131    ---------------------
2133     1. In the rejigging of the build system that eventually resulted in 7.1, the
2134        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
2135        brackets there is not right, since it causes compilers to look for an
2136        installed pcre.h, not the version that is in the source that is being
2137        compiled (which of course may be different). I have changed it back to:
2139          #include "pcre.h"
2141        I have a vague recollection that the change was concerned with compiling in
2142        different directories, but in the new build system, that is taken care of
2143        by the VPATH setting the Makefile.
2145     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
2146        when the subject happened to end in the byte 0x85 (e.g. if the last
2147        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
2148        characters but of course it shouldn't be taken as a newline when it is part
2149        of another character. The bug was that, for an unlimited repeat of . in
2150        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
2151        characters when looking for a newline.
2153     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
2155     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
2156        in debug output.
2158     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
2159        long printing in the pcrecpp unittest when running under MinGW.
2161     6. ESC_K was left out of the EBCDIC table.
2163     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
2164        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
2165        limit also applies to "virtual nesting" when a pattern is recursive, and in
2166        this case 1000 isn't so big. I have been able to remove this limit at the
2167        expense of backing off one optimization in certain circumstances. Normally,
2168        when pcre_exec() would call its internal match() function recursively and
2169        immediately return the result unconditionally, it uses a "tail recursion"
2170        feature to save stack. However, when a subpattern that can match an empty
2171        string has an unlimited repetition quantifier, it no longer makes this
2172        optimization. That gives it a stack frame in which to save the data for
2173        checking that an empty string has been matched. Previously this was taken
2174        from the 1000-entry workspace that had been reserved. So now there is no
2175        explicit limit, but more stack is used.
2177     8. Applied Daniel's patches to solve problems with the import/export magic
2178        syntax that is required for Windows, and which was going wrong for the
2179        pcreposix and pcrecpp parts of the library. These were overlooked when this
2180        problem was solved for the main library.
2182     9. There were some crude static tests to avoid integer overflow when computing
2183        the size of patterns that contain repeated groups with explicit upper
2184        limits. As the maximum quantifier is 65535, the maximum group length was
2185        set at 30,000 so that the product of these two numbers did not overflow a
2186        32-bit integer. However, it turns out that people want to use groups that
2187        are longer than 30,000 bytes (though not repeat them that many times).
2188        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
2189        made it possible to implement the integer overflow checks in a much more
2190        dynamic way, which I have now done. The artificial limitation on group
2191        length has been removed - we now have only the limit on the total length of
2192        the compiled pattern, which depends on the LINK_SIZE setting.
2194    10. Fixed a bug in the documentation for get/copy named substring when
2195        duplicate names are permitted. If none of the named substrings are set, the
2196        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
2197        empty string.
2199    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
2200        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
2201        because the ] is interpreted as the first data character and the
2202        terminating ] is not found. PCRE has been made compatible with Perl in this
2203        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
2204        cause memory overwriting.
2206    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
2207        string has been matched (to stop an infinite loop). It was not recognizing
2208        a conditional subpattern that could match an empty string if that
2209        subpattern was within another subpattern. For example, it looped when
2210        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
2211        condition was not nested. This bug has been fixed.
2213    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
2214        past the start of the subject in the presence of bytes with the top bit
2215        set, for example "\x8aBCD".
2217    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
2218        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
2220    14. Optimized (?!) to (*FAIL).
2222    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
2223        This restricts code points to be within the range 0 to 0x10FFFF, excluding
2224        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
2225        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
2226        does: it's just the validity check that is more restrictive.
2228    16. Inserted checks for integer overflows during escape sequence (backslash)
2229        processing, and also fixed erroneous offset values for syntax errors during
2230        backslash processing.
2232    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
2233        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
2235    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
2236        caused an overrun.
2238    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
2239        something other than just ASCII characters) inside a group that had an
2240        unlimited repeat caused a loop at compile time (while checking to see
2241        whether the group could match an empty string).
2243    20. Debugging a pattern containing \p or \P could cause a crash. For example,
2244        [\P{Any}] did so. (Error in the code for printing property names.)
2246    21. An orphan \E inside a character class could cause a crash.
2248    22. A repeated capturing bracket such as (A)? could cause a wild memory
2249        reference during compilation.
2251    23. There are several functions in pcre_compile() that scan along a compiled
2252        expression for various reasons (e.g. to see if it's fixed length for look
2253        behind). There were bugs in these functions when a repeated \p or \P was
2254        present in the pattern. These operators have additional parameters compared
2255        with \d, etc, and these were not being taken into account when moving along
2256        the compiled data. Specifically:
2258        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2259            length.
2261        (b) An item such as \pL+ within a repeated group could cause crashes or
2262            loops.
2264        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2265            "reference to non-existent subpattern" error.
2267        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2269    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2270        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2272    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2274    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2275        character were causing crashes (broken optimization).
2277    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2278        \p or \P) caused a compile-time loop.
2280    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2281        break. For example, the unstudied pattern [\r\n]A does not match the string
2282        "\r\nA" because change 7.0/46 below moves the current point on by two
2283        characters after failing to match at the start. However, the pattern \nA
2284        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2285        the same is true. There doesn't seem any very clean way out of this, but
2286        what I have chosen to do makes the common cases work: PCRE now takes note
2287        of whether there can be an explicit match for \r or \n anywhere in the
2288        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2289        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2290        pattern has explicit CR or LF references.
2292    29. Added (*CR) etc for changing newline setting at start of pattern.
2295    Version 7.2 19-Jun-07
2296    ---------------------
2298     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
2299        which is apparently normally available under Windows.
2301     2. Re-jig the pcregrep tests with different newline settings in an attempt
2302        to make them independent of the local environment's newline setting.
2304     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
2306     4. Some of the "internals" tests were previously cut out when the link size
2307        was not 2, because the output contained actual offsets. The recent new
2308        "Z" feature of pcretest means that these can be cut out, making the tests
2309        usable with all link sizes.
2311     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
2312        stack recursion. This gives a massive performance boost under BSD, but just
2313        a small improvement under Linux. However, it saves one field in the frame
2314        in all cases.
2316     6. Added more features from the forthcoming Perl 5.10:
2318        (a) (?-n) (where n is a string of digits) is a relative subroutine or
2319            recursion call. It refers to the nth most recently opened parentheses.
2321        (b) (?+n) is also a relative subroutine call; it refers to the nth next
2322            to be opened parentheses.
2324        (c) Conditions that refer to capturing parentheses can be specified
2325            relatively, for example, (?(-2)... or (?(+3)...
2327        (d) \K resets the start of the current match so that everything before
2328            is not part of it.
2330        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2332        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2333            reference syntax.
2335        (g) (?| introduces a group in which the numbering of parentheses in each
2336            alternative starts with the same number.
2338        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2340     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2343     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2344        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2345        for detecting groups that can match an empty string.
2347     9. A pattern with a very large number of alternatives (more than several
2348        hundred) was running out of internal workspace during the pre-compile
2349        phase, where pcre_compile() figures out how much memory will be needed. A
2350        bit of new cunning has reduced the workspace needed for groups with
2351        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2352        workspace instead of running out of the 4096 that are available.
2354    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2356    11. Applied patch from Google to remove an optimization that didn't quite work.
2357        The report of the bug said:
2359          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2360          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2361          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2363    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2364        it matched the wrong number of bytes.
2367    Version 7.1 24-Apr-07
2368  ---------------------  ---------------------
2370   1. Applied Bob Rossi and Daniel G's patches to convert the build system to one   1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
2371      that is more "standard", making use of automake and other autotools. There      that is more "standard", making use of automake and other Autotools. There
2372      is some re-arrangement of the files and adjustment of comments consequent      is some re-arrangement of the files and adjustment of comments consequent
2373      on this.      on this.
# Line 26  Version 7.1 05-Mar-07 Line 2389  Version 7.1 05-Mar-07
2389      .br or .in.      .br or .in.
2391   4. Updated comments in configure.ac that get placed in config.h.in and also   4. Updated comments in configure.ac that get placed in config.h.in and also
2392      arranged for config.h to be included in the distribution, with the name      arranged for config.h to be included in the distribution, with the name
2393      config.h.generic, for the benefit of those who have to compile without      config.h.generic, for the benefit of those who have to compile without
2394      Autotools (compare pcre.h, which is now distributed as pcre.h.generic).      Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
2396   5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan   5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
2397      Weber: (1) pcre_internal.h was missing some function renames; (2) updated      Weber: (1) pcre_internal.h was missing some function renames; (2) updated
2398      makevp.bat for the current PCRE, using the additional files !compile.txt,      makevp.bat for the current PCRE, using the additional files
2399      !linklib.txt, and pcregexp.pas.      makevp_c.txt, makevp_l.txt, and pcregexp.pas.
2401   6. A Windows user reported a minor discrepancy with test 2, which turned out   6. A Windows user reported a minor discrepancy with test 2, which turned out
2402      to be caused by a trailing space on an input line that had got lost in his      to be caused by a trailing space on an input line that had got lost in his
# Line 60  Version 7.1 05-Mar-07 Line 2423  Version 7.1 05-Mar-07
2423      called when UTF-8 support is disabled. Otherwise there are problems with a      called when UTF-8 support is disabled. Otherwise there are problems with a
2424      shared library.      shared library.
2426  10. The emulated memmove() function in pcre_internal.h (provided for those  10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
2427      environments that have neither memmove() nor bcopy()) was defining its  
2428      arguments as char * instead of void *.      (a) It was defining its arguments as char * instead of void *.
2430        (b) It was assuming that all moves were upwards in memory; this was true
2431            a long time ago when I wrote it, but is no longer the case.
2433        The emulated memove() is provided for those environments that have neither
2434        memmove() nor bcopy(). I didn't think anyone used it these days, but that
2435        is clearly not the case, as these two bugs were recently reported.
2437  11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,  11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
2438      and Detrail to create the HTML documentation, the .txt form of the man      and Detrail to create the HTML documentation, the .txt form of the man
2439      pages, and it removes trailing spaces from listed files. It also creates      pages, and it removes trailing spaces from listed files. It also creates
2440      pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter      pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
2441      case, it wraps all the #defines with #ifndefs. This script should be run      case, it wraps all the #defines with #ifndefs. This script should be run
2442      before "make dist".      before "make dist".
2444    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
2445        with Unicode property support.
2447        (a) For a maximizing quantifier, if the two different cases of the
2448            character were of different lengths in their UTF-8 codings (there are
2449            some cases like this - I found 11), and the matching function had to
2450            back up over a mixture of the two cases, it incorrectly assumed they
2451            were both the same length.
2453        (b) When PCRE was configured to use the heap rather than the stack for
2454            recursion during matching, it was not correctly preserving the data for
2455            the other case of a UTF-8 character when checking ahead for a match
2456            while processing a minimizing repeat. If the check also involved
2457            matching a wide character, but failed, corruption could cause an
2458            erroneous result when trying to check for a repeat of the original
2459            character.
2461    13. Some tidying changes to the testing mechanism:
2463        (a) The RunTest script now detects the internal link size and whether there
2464            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
2465            values substituted by "configure". (The RunGrepTest script already did
2466            this for UTF-8.) The configure.ac script no longer substitutes the
2467            relevant variables.
2469        (b) The debugging options /B and /D in pcretest show the compiled bytecode
2470            with length and offset values. This means that the output is different
2471            for different internal link sizes. Test 2 is skipped for link sizes
2472            other than 2 because of this, bypassing the problem. Unfortunately,
2473            there was also a test in test 3 (the locale tests) that used /B and
2474            failed for link sizes other than 2. Rather than cut the whole test out,
2475            I have added a new /Z option to pcretest that replaces the length and
2476            offset values with spaces. This is now used to make test 3 independent
2477            of link size. (Test 2 will be tidied up later.)
2479    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
2480        segmentation fault instead of returning the appropriate error message.
2482    15. In multiline mode when the newline sequence was set to "any", the pattern
2483        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
2484        This doesn't seem right; it now treats the CRLF combination as the line
2485        ending, and so does not match in that case. It's only a pattern such as ^$
2486        that would hit this one: something like ^ABC$ would have failed after \r
2487        and then tried again after \r\n.
2489    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
2490        in an attempt to make files that differ only in their line terminators
2491        compare equal. This works on Linux.
2493    17. Under certain error circumstances pcregrep might try to free random memory
2494        as it exited. This is now fixed, thanks to valgrind.
2496    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
2497        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
2498        was because its rules for how to advance for /g after matching an empty
2499        string at the end of a line did not allow for this case. They now check for
2500        it specially.
2502    20. pcretest is supposed to handle patterns and data of any length, by
2503        extending its buffers when necessary. It was getting this wrong when the
2504        buffer for a data line had to be extended.
2506    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
2507        CRLF as a newline sequence.
2509    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
2510        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
2511        I have nevertheless tidied it up.
2513    23. Added some casts to kill warnings from HP-UX ia64 compiler.
2515    24. Added a man page for pcre-config.
2518  Version 7.0 19-Dec-06  Version 7.0 19-Dec-06
2519  ---------------------  ---------------------

Removed from v.113  
changed lines
  Added in v.1261

  ViewVC Help
Powered by ViewVC 1.1.5