ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 91 by nigel, Sat Feb 24 21:41:34 2007 UTC revision 1364 by ph10, Sat Oct 5 15:45:11 2013 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.34 xx-xxxx-201x
5    --------------------------
7    1.  Add pcre[16|32]_jit_free_unused_memory to forcibly free unused JIT
8        executable memory. Patch inspired by Carsten Klein.
10    2.  ./configure --enable-coverage defined SUPPORT_GCOV in config.h, although
11        this macro is never tested and has no effect, because the work to support
12        coverage involves only compiling and linking options and special targets in
13        the Makefile. The comment in config.h implied that defining the macro would
14        enable coverage support, which is totally false. There was also support for
15        setting this macro in the CMake files (my fault, I just copied it from
16        configure). SUPPORT_GCOV has now been removed.
18    3.  Make a small performance improvement in strlen16() and strlen32() in
19        pcretest.
21    4.  Change 36 for 8.33 left some unreachable statements in pcre_exec.c,
22        detected by the Solaris compiler (gcc doesn't seem to be able to diagnose
23        these cases). There was also one in pcretest.c.
25    5.  Cleaned up a "may be uninitialized" compiler warning in pcre_exec.c.
27    6.  In UTF mode, the code for checking whether a group could match an empty
28        string (which is used for indefinitely repeated groups to allow for
29        breaking an infinite loop) was broken when the group contained a repeated
30        negated single-character class with a character that occupied more than one
31        data item and had a minimum repetition of zero (for example, [^\x{100}]* in
32        UTF-8 mode). The effect was undefined: the group might or might not be
33        deemed as matching an empty string, or the program might have crashed.
35    7.  The code for checking whether a group could match an empty string was not
36        recognizing that \h, \H, \v, \V, and \R must match a character.
38    8.  Implemented PCRE_INFO_MATCH_EMPTY, which yields 1 if the pattern can match
39        an empty string. If it can, pcretest shows this in its information output.
41    9.  Fixed two related bugs that applied to Unicode extended grapheme clusters
42        that were repeated with a maximizing qualifier (e.g. \X* or \X{2,5}) when
43        matched by pcre_exec() without using JIT:
45        (a) If the rest of the pattern did not match after a maximal run of
46            grapheme clusters, the code for backing up to try with fewer of them
47            did not always back up over a full grapheme when characters that do not
48            have the modifier quality were involved, e.g. Hangul syllables.
50        (b) If the match point in a subject started with modifier character, and
51            there was no match, the code could incorrectly back up beyond the match
52            point, and potentially beyond the first character in the subject,
53            leading to a segfault or an incorrect match result.
55    10. A conditional group with an assertion condition could lead to PCRE
56        recording an incorrect first data item for a match if no other first data
57        item was recorded. For example, the pattern (?(?=ab)ab) recorded "a" as a
58        first data item, and therefore matched "ca" after "c" instead of at the
59        start.
61    11. Change 40 for 8.33 (allowing pcregrep to find empty strings) showed up a
62        bug that caused the command "echo a | ./pcregrep -M '|a'" to loop.
64    12. The source of pcregrep now includes z/OS-specific code so that it can be
65        compiled for z/OS as part of the special z/OS distribution.
67    13. Added the -T and -TM options to pcretest.
69    14. The code in pcre_compile.c for creating the table of named capturing groups
70        has been refactored. Instead of creating the table dynamically during the
71        actual compiling pass, the information is remembered during the pre-compile
72        pass (on the stack unless there are more than 20 named groups, in which
73        case malloc() is used) and the whole table is created before the actual
74        compile happens. This has simplified the code (it is now nearly 150 lines
75        shorter) and prepared the way for better handling of references to groups
76        with duplicate names.
78    15. A back reference to a named subpattern when there is more than one of the
79        same name now checks them in the order in which they appear in the pattern.
80        The first one that is set is used for the reference. Previously only the
81        first one was inspected. This change makes PCRE more compatible with Perl.
83    16. Unicode character properties were updated from Unicode 6.3.0.
85    17. The compile-time code for auto-possessification has been refactored, based
86        on a patch by Zoltan Herczeg. It now happens after instead of during
87        compilation. The code is cleaner, and more cases are handled. The option
88        PCRE_NO_AUTO_POSSESSIFY is added for testing purposes, and the -O and /O
89        options in pcretest are provided to set it.
91    18. The character VT has been added to the set of characters that match \s and
92        are generally treated as white space, following this same change in Perl
93        5.18. There is now no difference between "Perl space" and "POSIX space".
96    Version 8.33 28-May-2013
97    ------------------------
99    1.  Added 'U' to some constants that are compared to unsigned integers, to
100        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
101        variables that are added to signed variables, to ensure the result is
102        signed and can be negated.
104    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
105        CMake config files.
107    3.  Revise the creation of config.h.generic so that all boolean macros are
108        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
109        overriding via -D on the command line possible.
111    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
112        to unsigned int is reported to make a quite noticeable speed difference in
113        a specific Windows environment. Testing on Linux did also appear to show
114        some benefit (and it is clearly not harmful). Also fixed the definition of
115        Xop which should be unsigned.
117    5.  Related to (4), changing the definition of the intermediate variable cc
118        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
119        improvements.
121    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
122        unnecessary spaces.
124    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
125        and later.
127    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
129    9.  Optimizing fast_forward_start_bits in JIT.
131    10. Adding support for callouts in JIT, and fixing some issues revealed
132        during this work. Namely:
134        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
136        (b) Minimum length was not checked before the matching is started.
138    11. The value of capture_last that is passed to callouts was incorrect in some
139        cases when there was a capture on one path that was subsequently abandoned
140        after a backtrack. Also, the capture_last value is now reset after a
141        recursion, since all captures are also reset in this case.
143    12. The interpreter no longer returns the "too many substrings" error in the
144        case when an overflowing capture is in a branch that is subsequently
145        abandoned after a backtrack.
147    13. In the pathological case when an offset vector of size 2 is used, pcretest
148        now prints out the matched string after a yield of 0 or 1.
150    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
151        Only supported by the JIT compiler at the moment.
153    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
155    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
156        offset of the starting point of the matching process, provided the offsets
157        vector is large enough.
159    17. The \A escape now records a lookbehind value of 1, though its execution
160        does not actually inspect the previous character. This is to ensure that,
161        in partial multi-segment matching, at least one character from the old
162        segment is retained when a new segment is processed. Otherwise, if there
163        are no lookbehinds in the pattern, \A might match incorrectly at the start
164        of a new segment.
166    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
168    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
169        gives some modest performance improvement in 8-bit mode.
171    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
172        be expressed in certain programming languages using Universal Character
173        Names.
175    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
176        which points out that "non characters" are not "characters that may not
177        appear in Unicode strings" but rather "characters that are reserved for
178        internal use and have only local meaning".
180    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
181        there was a conditional group that depended on an assertion, if the
182        assertion was false, the callout that immediately followed the alternation
183        in the condition was skipped when pcre_exec() was used for matching.
185    23. Allow an explicit callout to be inserted before an assertion that is the
186        condition for a conditional group, for compatibility with automatic
187        callouts, which always insert a callout at this point.
189    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
190        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
192    25. (*PRUNE) is now supported by the JIT compiler.
194    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
196    27. Fix the case where there are two or more SKIPs with arguments that may be
197        ignored.
199    28. (*SKIP) is now supported by the JIT compiler.
201    29. (*THEN) is now supported by the JIT compiler.
203    30. Update RunTest with additional test selector options.
205    31. The way PCRE handles backtracking verbs has been changed in two ways.
207        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
208        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
209        backtracking. In some cases this makes it more Perl-compatible, but Perl's
210        rather obscure rules do not always do the same thing.
212        (2) Previously, backtracking verbs were confined within assertions. This is
213        no longer the case for positive assertions, except for (*ACCEPT). Again,
214        this sometimes improves Perl compatibility, and sometimes does not.
216    32. A number of tests that were in test 2 because Perl did things differently
217        have been moved to test 1, because either Perl or PCRE has changed, and
218        these tests are now compatible.
220    32. Backtracking control verbs are now handled in the same way in JIT and
221        interpreter.
223    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
224        contained a forward subroutine reference caused a compile error.
226    34. Auto-detect and optimize limited repetitions in JIT.
228    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
229        blocking (*UTF) etc.
231    36. In the interpreter, maximizing pattern repetitions for characters and
232        character types now use tail recursion, which reduces stack usage.
234    37. The value of the max lookbehind was not correctly preserved if a compiled
235        and saved regex was reloaded on a host of different endianness.
237    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
238        of the compiled pattern block, expand the flags field from 16 to 32 bits
239        because it was almost full.
241    39. Try madvise first before posix_madvise.
243    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
244        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
245        original change locked out all matches of empty strings. This has been
246        changed so that one match of an empty string per line is recognized.
247        Subsequent searches on the same line (for colouring or for --only-matching,
248        for example) do not recognize empty strings.
250    41. Applied a user patch to fix a number of spelling mistakes in comments.
252    42. Data lines longer than 65536 caused pcretest to crash.
254    43. Clarified the data type for length and startoffset arguments for pcre_exec
255        and pcre_dfa_exec in the function-specific man pages, where they were
256        explicitly stated to be in bytes, never having been updated. I also added
257        some clarification to the pcreapi man page.
259    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
260        a segmentation fault.
263    Version 8.32 30-November-2012
264    -----------------------------
266    1.  Improved JIT compiler optimizations for first character search and single
267        character iterators.
269    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
270        Patch by Daniel Richard G.
272    3.  Single character iterator optimizations in the JIT compiler.
274    4.  Improved JIT compiler optimizations for character ranges.
276    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
277        Reported by Giuseppe D'Angelo.
279    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
280        of a line, was being set incorrectly in cases where .* appeared inside
281        atomic brackets at the start of a pattern, or where there was a subsequent
282        *PRUNE or *SKIP.
284    7.  Improved instruction cache flush for POWER/PowerPC.
285        Patch by Daniel Richard G.
287    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
288        grep:
290        (a) There is now no limit to the number of patterns to be matched.
292        (b) An error is given if a pattern is too long.
294        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
295            are now supported.
297        (d) --exclude-from and --include-from (multiple use) have been added.
299        (e) Exclusions and inclusions now apply to all files and directories, not
300            just to those obtained from scanning a directory recursively.
302        (f) Multiple uses of -f and --file-list are now supported.
304        (g) In a Windows environment, the default for -d has been changed from
305            "read" (the GNU grep default) to "skip", because otherwise the presence
306            of a directory in the file list provokes an error.
308        (h) The documentation has been revised and clarified in places.
310    9.  Improve the matching speed of capturing brackets.
312    10. Changed the meaning of \X so that it now matches a Unicode extended
313        grapheme cluster.
315    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
316        out POSIX threads when JIT support is configured.
318    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
320    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
321        structure could go wrong in environments where size_t is not the same size
322        as int.
324    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
325        set.
327    15. The EBCDIC support had decayed; later updates to the code had included
328        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
329        general tidy up of EBCDIC-related issues, and the documentation was also
330        not quite right. There is now a test that can be run on ASCII systems to
331        check some of the EBCDIC-related things (but is it not a full test).
333    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
334        in a small tidy to the code.
336    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
338    18. If the --only-matching (-o) option in pcregrep is specified multiple
339        times, each one causes appropriate output. For example, -o1 -o2 outputs the
340        substrings matched by the 1st and 2nd capturing parentheses. A separating
341        string can be specified by --om-separator (default empty).
343    19. Improving the first n character searches.
345    20. Turn case lists for horizontal and vertical white space into macros so that
346        they are defined only once.
348    21. This set of changes together give more compatible Unicode case-folding
349        behaviour for characters that have more than one other case when UCP
350        support is available.
352        (a) The Unicode property table now has offsets into a new table of sets of
353            three or more characters that are case-equivalent. The MultiStage2.py
354            script that generates these tables (the pcre_ucd.c file) now scans
355            CaseFolding.txt instead of UnicodeData.txt for character case
356            information.
358        (b) The code for adding characters or ranges of characters to a character
359            class has been abstracted into a generalized function that also handles
360            case-independence. In UTF-mode with UCP support, this uses the new data
361            to handle characters with more than one other case.
363        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
364            whose other case is greater than 256 are now correctly matched
365            caselessly. Previously, the high codepoint matched the low one, but not
366            vice versa.
368        (d) The processing of \h, \H, \v, and \ in character classes now makes use
369            of the new class addition function, using character lists defined as
370            macros alongside the case definitions of 20 above.
372        (e) Caseless back references now work with characters that have more than
373            one other case.
375        (f) General caseless matching of characters with more than one other case
376            is supported.
378    22. Unicode character properties were updated from Unicode 6.2.0
380    23. Improved CMake support under Windows. Patch by Daniel Richard G.
382    24. Add support for 32-bit character strings, and UTF-32
384    25. Major JIT compiler update (code refactoring and bugfixing).
385        Experimental Sparc 32 support is added.
387    26. Applied a modified version of Daniel Richard G's patch to create
388        pcre.h.generic and config.h.generic by "make" instead of in the
389        PrepareRelease script.
391    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
392        pcre_compile.c when checking for a zero character.
394    28. Introducing a native interface for JIT. Through this interface, the compiled
395        machine code can be directly executed. The purpose of this interface is to
396        provide fast pattern matching, so several sanity checks are not performed.
397        However, feature tests are still performed. The new interface provides
398        1.4x speedup compared to the old one.
400    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
401        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
402        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
404    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
405        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
406        the "old" RFC 2279). Instead, it ended up passing a negative length to
407        pcre_exec().
409    31. Add support for GCC's visibility feature to hide internal functions.
411    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
412        "unknown -C option" after outputting 0 or 1.
414    33. There is now support for generating a code coverage report for the test
415        suite in environments where gcc is the compiler and lcov is installed. This
416        is mainly for the benefit of the developers.
418    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
419        unaddressable using valgrind annotations, allowing valgrind to detect
420        invalid memory accesses. This is mainly for the benefit of the developers.
422    25. (*UTF) can now be used to start a pattern in any of the three libraries.
424    26. Give configure error if --enable-cpp but no C++ compiler found.
427    Version 8.31 06-July-2012
428    -------------------------
430    1.  Fixing a wrong JIT test case and some compiler warnings.
432    2.  Removed a bashism from the RunTest script.
434    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
435        to unsigned type, result still unsigned" that was given by an MS compiler
436        on encountering the code "-sizeof(xxx)".
438    4.  Partial matching support is added to the JIT compiler.
440    5.  Fixed several bugs concerned with partial matching of items that consist
441        of more than one character:
443        (a) /^(..)\1/ did not partially match "aba" because checking references was
444            done on an "all or nothing" basis. This also applied to repeated
445            references.
447        (b) \R did not give a hard partial match if \r was found at the end of the
448            subject.
450        (c) \X did not give a hard partial match after matching one or more
451            characters at the end of the subject.
453        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
454            a partial match for the string "\r".
456        (e) When newline was set to CRLF, the metacharacter "." did not recognize
457            a partial match for a CR character at the end of the subject string.
459    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
460        running pcretest, the text "(JIT)" added to the output whenever JIT is
461        actually used to run the match.
463    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
464        or /S+[+] with a digit between 1 and 7.
466    8.  OP_NOT now supports any UTF character not just single-byte ones.
468    9.  (*MARK) control verb is now supported by the JIT compiler.
470    10. The command "./RunTest list" lists the available tests without actually
471        running any of them. (Because I keep forgetting what they all are.)
475    12. Applied a (slightly modified) user-supplied patch that improves performance
476        when the heap is used for recursion (compiled with --disable-stack-for-
477        recursion). Instead of malloc and free for each heap frame each time a
478        logical recursion happens, frames are retained on a chain and re-used where
479        possible. This sometimes gives as much as 30% improvement.
481    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
482        call.
484    14. As documented, (*COMMIT) is now confined to within a positive assertion.
486    15. It is now possible to link pcretest with libedit as an alternative to
487        libreadline.
489    16. (*COMMIT) control verb is now supported by the JIT compiler.
491    17. The Unicode data tables have been updated to Unicode 6.1.0.
493    18. Added --file-list option to pcregrep.
495    19. Added binary file support to pcregrep, including the -a, --binary-files,
496        -I, and --text options.
498    20. The madvise function is renamed for posix_madvise for QNX compatibility
499        reasons. Fixed by Giuseppe D'Angelo.
501    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
503    22. Changed the option for creating symbolic links for 16-bit man pages from
504        -s to -sf so that re-installing does not cause issues.
506    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
508    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
509        matches in certain environments (the workspace was not being correctly
510        retained). Also added to pcre_dfa_exec() a simple plausibility check on
511        some of the workspace data at the beginning of a restart.
513    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
514        was not doing so when it should - probably a typo introduced by SVN 528
515        (change 8.10/14).
517    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
518        \w+ when the character tables indicated that \x{c4} was a word character.
519        There were several related cases, all because the tests for doing a table
520        lookup were testing for characters less than 127 instead of 255.
522    27. If a pattern contains capturing parentheses that are not used in a match,
523        their slots in the ovector are set to -1. For those that are higher than
524        any matched groups, this happens at the end of processing. In the case when
525        there were back references that the ovector was too small to contain
526        (causing temporary malloc'd memory to be used during matching), and the
527        highest capturing number was not used, memory off the end of the ovector
528        was incorrectly being set to -1. (It was using the size of the temporary
529        memory instead of the true size.)
531    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
532        ovector size, it uses memory at the end of the block that it has got.
534    29. Check for an overlong MARK name and give an error at compile time. The
535        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
537    30. JIT compiler update.
539    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
540        Rill for the patch.
542    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
544    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
546    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
547        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
548        was enabled.
550    35. Improve JIT code generation for greedy plus quantifier.
552    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
553        1 to "aa" instead of to an empty string. The bug affected repeated groups
554        that could potentially match an empty string.
556    37. Optimizing single character iterators in JIT.
558    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
559        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
560        codepoints that are too big for the mode are faulted, and in a UTF mode,
561        disallowed codepoints are also faulted.
563    39. If PCRE was compiled with UTF support, in three places in the DFA
564        matcher there was code that should only have been obeyed in UTF mode, but
565        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
566        processing when bytes with values greater than 127 were present. In 16-bit
567        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
568        both cases the values are those that cannot be the first data item in a UTF
569        character. The three items that might have provoked this were recursions,
570        possessively repeated groups, and atomic groups.
572    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
573        and pcregrep, because some OS require shared objects to be explicitly
574        passed to ld, causing the link step to fail if they are not.
576    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
577        mode, patterns that started with \h* or \R* might be incorrectly matched.
580    Version 8.30 04-February-2012
581    -----------------------------
583    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
584        name is defined in ctype.h.
586    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
587        only in quite long subpatterns.
589    3.  Removed the function pcre_info(), which has been obsolete and deprecated
590        since it was replaced by pcre_fullinfo() in February 2000.
592    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
593        match a (*MARK), and the match failed at the start of the subject, a
594        reference to memory before the start of the subject could occur. This bug
595        was introduced by fix 17 of release 8.21.
597    5.  A reference to an unset group with zero minimum repetition was giving
598        totally wrong answers (in non-JavaScript-compatibility mode). For example,
599        /(another)?(\1?)test/ matched against "hello world test". This bug was
600        introduced in release 8.13.
602    6.  Add support for 16-bit character strings (a large amount of work involving
603        many changes and refactorings).
605    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
606        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
607        from a file.
609    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
610        rounding is not applied in this particular case).
612    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
613        if they appear, or are escaped, in patterns.
615    10. Get rid of a number of -Wunused-but-set-variable warnings.
617    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
618        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
619        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
620        also returns the mark "x". This bug applied to capturing parentheses,
621        non-capturing parentheses, and atomic parentheses. It also applied to some
622        assertions.
624    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
625        information out of configure.ac instead of relying on pcre.h.generic, which
626        is not stored in the repository.
628    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
629        -lreadline.
631    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
633    15. Applied Graycode's patch to put the top-level frame on the stack rather
634        than the heap when not using the stack for recursion. This gives a
635        performance improvement in many cases when recursion is not deep.
637    16. Experimental code added to "pcretest -C" to output the stack frame size.
640    Version 8.21 12-Dec-2011
641    ------------------------
643    1.  Updating the JIT compiler.
645    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
646        are added as well.
648    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
649        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
650        calling _pcre_jit_exec. Some extra comments are added.
652    4.  (*MARK) settings inside atomic groups that do not contain any capturing
653        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
654        was introduced by change 18 for 8.20.
656    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
657        ECMA-262 standard.
659    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
660        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
661        This bug was probably introduced by change 9 of 8.13.
663    7.  While fixing 6 above, I noticed that a number of other items were being
664        incorrectly rejected as "not fixed length". This arose partly because newer
665        opcodes had not been added to the fixed-length checking code. I have (a)
666        corrected the bug and added tests for these items, and (b) arranged for an
667        error to occur if an unknown opcode is encountered while checking for fixed
668        length instead of just assuming "not fixed length". The items that were
669        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
670        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
671        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
673    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
674        being incorrectly compiled and would have given unpredicatble results.
676    9.  A possessively repeated subpattern with minimum repeat count greater than
677        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
678        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
679        the first (A) could occur when it should not.
681    10. Add a cast and remove a redundant test from the code.
683    11. JIT should use pcre_malloc/pcre_free for allocation.
685    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
686        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
687        is anything other than /usr, -L is still shown).
689    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
691    14. Perl does not support \N without a following name in a [] class; PCRE now
692        also gives an error.
694    15. If a forward reference was repeated with an upper limit of around 2000,
695        it caused the error "internal error: overran compiling workspace". The
696        maximum number of forward references (including repeats) was limited by the
697        internal workspace, and dependent on the LINK_SIZE. The code has been
698        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
699        the default depends on LINK_SIZE. There is a new upper limit (for safety)
700        of around 200,000 forward references. While doing this, I also speeded up
701        the filling in of repeated forward references.
703    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
704        incorrectly expecting the subject to contain another "a" after the start.
706    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
707        in the match, the SKIP should be ignored. This was not happening; instead
708        the SKIP was being treated as NOMATCH. For patterns such as
709        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
710        tested.
712    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
713        now much more compatible with Perl, in particular in cases where the result
714        is a non-match for a non-anchored pattern. For example, if
715        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
716        "m", where previously it did not return a name. A side effect of this
717        change is that for partial matches, the last encountered mark name is
718        returned, as for non matches. A number of tests that were previously not
719        Perl-compatible have been moved into the Perl-compatible test files. The
720        refactoring has had the pleasing side effect of removing one argument from
721        the match() function, thus reducing its stack requirements.
723    19. If the /S+ option was used in pcretest to study a pattern using JIT,
724        subsequent uses of /S (without +) incorrectly behaved like /S+.
726    21. Retrieve executable code size support for the JIT compiler and fixing
727        some warnings.
729    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
730        not work when the shorter character appeared right at the end of the
731        subject string.
733    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
734        systems.
736    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
737        output it when the /M option is used in pcretest.
739    25. The CheckMan script was not being included in the distribution. Also, added
740        an explicit "perl" to run Perl scripts from the PrepareRelease script
741        because this is reportedly needed in Windows.
743    26. If study data was being save in a file and studying had not found a set of
744        "starts with" bytes for the pattern, the data written to the file (though
745        never used) was taken from uninitialized memory and so caused valgrind to
746        complain.
748    27. Updated RunTest.bat as provided by Sheri Pierce.
750    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
752    29. Computation of memory usage for the table of capturing group names was
753        giving an unnecessarily large value.
756    Version 8.20 21-Oct-2011
757    ------------------------
759    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
760        a POSIX class. After further experiments with Perl, which convinced me that
761        Perl has bugs and confusions, a closing square bracket is no longer allowed
762        in a POSIX name. This bug also affected patterns with classes that started
763        with full stops.
765    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
766        captured substring, but while checking the failing first alternative,
767        substring 1 is temporarily captured. If the output vector supplied to
768        pcre_exec() was not big enough for this capture, the yield of the function
769        was still zero ("insufficient space for captured substrings"). This cannot
770        be totally fixed without adding another stack variable, which seems a lot
771        of expense for a edge case. However, I have improved the situation in cases
772        such as /(a)(b)x|abc/ matched against "abc", where the return code
773        indicates that fewer than the maximum number of slots in the ovector have
774        been set.
776    3.  Related to (2) above: when there are more back references in a pattern than
777        slots in the output vector, pcre_exec() uses temporary memory during
778        matching, and copies in the captures as far as possible afterwards. It was
779        using the entire output vector, but this conflicts with the specification
780        that only 2/3 is used for passing back captured substrings. Now it uses
781        only the first 2/3, for compatibility. This is, of course, another edge
782        case.
784    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
785        main code base, and can be used by building with --enable-jit. When this is
786        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
787        runtime --no-jit option is given.
789    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
790        ovector, the return from the function was zero, implying that there were
791        other matches that did not fit. The correct "exactly full" value is now
792        returned.
794    6.  If a subpattern that was called recursively or as a subroutine contained
795        (*PRUNE) or any other control that caused it to give a non-standard return,
796        invalid errors such as "Error -26 (nested recursion at the same subject
797        position)" or even infinite loops could occur.
799    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
800        computing the minimum length on reaching *ACCEPT, and so ended up with the
801        wrong value of 1 rather than 0. Further investigation indicates that
802        computing a minimum subject length in the presence of *ACCEPT is difficult
803        (think back references, subroutine calls), and so I have changed the code
804        so that no minimum is registered for a pattern that contains *ACCEPT.
806    8.  If (*THEN) was present in the first (true) branch of a conditional group,
807        it was not handled as intended. [But see 16 below.]
809    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
810        Sheri Pierce.
812    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
813        the first byte in a match must be "a".
815    11. Change 17 for 8.13 increased the recursion depth for patterns like
816        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
817        pattern contains any instances of (*THEN). If it does not, the old
818        optimizations are restored. It would be nice to do this on a per-group
819        basis, but at the moment that is not feasible.
821    12. In some environments, the output of pcretest -C is CRLF terminated. This
822        broke RunTest's code that checks for the link size. A single white space
823        character after the value is now allowed for.
825    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
826        For "fr", it uses the Windows-specific input and output files.
828    14. If (*THEN) appeared in a group that was called recursively or as a
829        subroutine, it did not work as intended. [But see next item.]
831    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
832        pattern fragments (but not containing any | characters). If A and B are
833        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
834        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
835        D. In other words, Perl considers parentheses that do not contain any |
836        characters to be part of a surrounding alternative, whereas PCRE was
837        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
838        differently. PCRE now behaves in the same way as Perl, except in the case
839        of subroutine/recursion calls such as (?1) which have in any case always
840        been different (but PCRE had them first :-).
842    16. Related to 15 above: Perl does not treat the | in a conditional group as
843        creating alternatives. Such a group is treated in the same way as an
844        ordinary group without any | characters when processing (*THEN). PCRE has
845        been changed to match Perl's behaviour.
847    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
848        RunGrepTest script failed.
850    18. Change 22 for version 13 caused atomic groups to use more stack. This is
851        inevitable for groups that contain captures, but it can lead to a lot of
852        stack use in large patterns. The old behaviour has been restored for atomic
853        groups that do not contain any capturing parentheses.
855    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
856        suppress the check for a minimum subject length at run time. (If it was
857        given to pcre_exec() or pcre_dfa_exec() it did work.)
859    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
860        fail to work when decoding hex characters in data strings in EBCDIC
861        environments.
863    21. It appears that in at least one Mac OS environment, the isxdigit() function
864        is implemented as a macro that evaluates to its argument more than once,
865        contravening the C 90 Standard (I haven't checked a later standard). There
866        was an instance in pcretest which caused it to go wrong when processing
867        \x{...} escapes in subject strings. The has been rewritten to avoid using
868        things like p++ in the argument of isxdigit().
871    Version 8.13 16-Aug-2011
872    ------------------------
874    1.  The Unicode data tables have been updated to Unicode 6.0.0.
876    2.  Two minor typos in pcre_internal.h have been fixed.
878    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
879        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
880        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
882    4.  There were a number of related bugs in the code for matching backrefences
883        caselessly in UTF-8 mode when codes for the characters concerned were
884        different numbers of bytes. For example, U+023A and U+2C65 are an upper
885        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
886        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
887        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
888        2-byte code at the end of the subject (it thought there wasn't enough data
889        left).
891    5.  Comprehensive information about what went wrong is now returned by
892        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
893        as the output vector has at least 2 elements. The offset of the start of
894        the failing character and a reason code are placed in the vector.
896    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
897        now returned is for the first byte of the failing character, instead of the
898        last byte inspected. This is an incompatible change, but I hope it is small
899        enough not to be a problem. It makes the returned offset consistent with
900        pcre_exec() and pcre_dfa_exec().
902    7.  pcretest now gives a text phrase as well as the error number when
903        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
904        failure, the offset and reason code are output.
906    8.  When \R was used with a maximizing quantifier it failed to skip backwards
907        over a \r\n pair if the subsequent match failed. Instead, it just skipped
908        back over a single character (\n). This seems wrong (because it treated the
909        two characters as a single entity when going forwards), conflicts with the
910        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
911        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
912        has been changed.
914    9.  Some internal refactoring has changed the processing so that the handling
915        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
916        time (the PCRE_DOTALL option was changed this way some time ago: version
917        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
918        which was always a bit of a fudge. It also means that there is one less
919        argument for the match() function, which reduces its stack requirements
920        slightly. This change also fixes an incompatibility with Perl: the pattern
921        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
923    10. More internal refactoring has drastically reduced the number of recursive
924        calls to match() for possessively repeated groups such as (abc)++ when
925        using pcre_exec().
927    11. While implementing 10, a number of bugs in the handling of groups were
928        discovered and fixed:
930        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
931        (a|)*(?1) gave a compile-time internal error.
932        ((a|)+)+  did not notice that the outer group could match an empty string.
933        (^a|^)+   was not marked as anchored.
934        (.*a|.*)+ was not marked as matching at start or after a newline.
936    12. Yet more internal refactoring has removed another argument from the match()
937        function. Special calls to this function are now indicated by setting a
938        value in a variable in the "match data" data block.
940    13. Be more explicit in pcre_study() instead of relying on "default" for
941        opcodes that mean there is no starting character; this means that when new
942        ones are added and accidentally left out of pcre_study(), testing should
943        pick them up.
945    14. The -s option of pcretest has been documented for ages as being an old
946        synonym of -m (show memory usage). I have changed it to mean "force study
947        for every regex", that is, assume /S for every regex. This is similar to -i
948        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
949        using it. It makes it easier to run collections of tests with and without
950        study enabled, and thereby test pcre_study() more easily. All the standard
951        tests are now run with and without -s (but some patterns can be marked as
952        "never study" - see 20 below).
954    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
955        restoration of the capturing data to the outer values was not happening
956        correctly.
958    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
959        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
960        pattern had matched an empty string, and so incorrectly returned a no
961        match.
963    17. There was optimizing code for the last branch of non-capturing parentheses,
964        and also for the obeyed branch of a conditional subexpression, which used
965        tail recursion to cut down on stack usage. Unfortunately, now that there is
966        the possibility of (*THEN) occurring in these branches, tail recursion is
967        no longer possible because the return has to be checked for (*THEN). These
968        two optimizations have therefore been removed. [But see 8.20/11 above.]
970    18. If a pattern containing \R was studied, it was assumed that \R always
971        matched two bytes, thus causing the minimum subject length to be
972        incorrectly computed because \R can also match just one byte.
974    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
975        was incorrectly computed.
977    20. If /S is present twice on a test pattern in pcretest input, it now
978        *disables* studying, thereby overriding the use of -s on the command line
979        (see 14 above). This is necessary for one or two tests to keep the output
980        identical in both cases.
982    21. When (*ACCEPT) was used in an assertion that matched an empty string and
983        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
985    22. When an atomic group that contained a capturing parenthesis was
986        successfully matched, but the branch in which it appeared failed, the
987        capturing was not being forgotten if a higher numbered group was later
988        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
989        group 1 to "a", when in fact it should be unset. This applied to multi-
990        branched capturing and non-capturing groups, repeated or not, and also to
991        positive assertions (capturing in negative assertions does not happen
992        in PCRE) and also to nested atomic groups.
994    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
995        subject after a captured substring, to make it easier to tell which of a
996        number of identical substrings has been captured.
998    24. The way atomic groups are processed by pcre_exec() has been changed so that
999        if they are repeated, backtracking one repetition now resets captured
1000        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
1001        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
1002        "aaa". Previously, it would have been "a". As part of this code
1003        refactoring, the way recursive calls are handled has also been changed.
1005    25. If an assertion condition captured any substrings, they were not passed
1006        back unless some other capturing happened later. For example, if
1007        (?(?=(a))a) was matched against "a", no capturing was returned.
1009    26. When studying a pattern that contained subroutine calls or assertions,
1010        the code for finding the minimum length of a possible match was handling
1011        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
1012        group 1 called group 2 while simultaneously a separate group 2 called group
1013        1). A stack overflow occurred in this case. I have fixed this by limiting
1014        the recursion depth to 10.
1016    27. Updated RunTest.bat in the distribution to the version supplied by Tom
1017        Fortmann. This supports explicit test numbers on the command line, and has
1018        argument validation and error reporting.
1020    28. An instance of \X with an unlimited repeat could fail if at any point the
1021        first character it looked at was a mark character.
1023    29. Some minor code refactoring concerning Unicode properties and scripts
1024        should reduce the stack requirement of match() slightly.
1026    30. Added the '=' option to pcretest to check the setting of unused capturing
1027        slots at the end of the pattern, which are documented as being -1, but are
1028        not included in the return count.
1030    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
1031        compiled something random. Now it gives a compile-time error (as does
1032        Perl).
1034    32. A *MARK encountered during the processing of a positive assertion is now
1035        recorded and passed back (compatible with Perl).
1037    33. If --only-matching or --colour was set on a pcregrep call whose pattern
1038        had alternative anchored branches, the search for a second match in a line
1039        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
1040        matched the line "0102" twice. The same bug affected patterns that started
1041        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
1042        twice.
1044    34. Previously, PCRE did not allow quantification of assertions. However, Perl
1045        does, and because of capturing effects, quantifying parenthesized
1046        assertions may at times be useful. Quantifiers are now allowed for
1047        parenthesized assertions.
1049    35. A minor code tidy in pcre_compile() when checking options for \R usage.
1051    36. \g was being checked for fancy things in a character class, when it should
1052        just be a literal "g".
1054    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
1055        appearance of a nested POSIX class supersedes an apparent external class.
1056        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
1057        unescaped square brackets may also appear as part of class names. For
1058        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
1059        more like Perl. (But see 8.20/1 above.)
1061    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
1062        was because it thought it was \N{name}, which is not supported).
1064    39. Add minix to OS list not supporting the -S option in pcretest.
1066    40. PCRE tries to detect cases of infinite recursion at compile time, but it
1067        cannot analyze patterns in sufficient detail to catch mutual recursions
1068        such as ((?1))((?2)). There is now a runtime test that gives an error if a
1069        subgroup is called recursively as a subpattern for a second time at the
1070        same position in the subject string. In previous releases this might have
1071        been caught by the recursion limit, or it might have run out of stack.
1073    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
1074        happen only once. PCRE was, however incorrectly giving a compile time error
1075        "recursive call could loop indefinitely" because it cannot analyze the
1076        pattern in sufficient detail. The compile time test no longer happens when
1077        PCRE is compiling a conditional subpattern, but actual runaway loops are
1078        now caught at runtime (see 40 above).
1080    42. It seems that Perl allows any characters other than a closing parenthesis
1081        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
1082        has been changed to be the same.
1084    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
1085        as not to get warnings when autogen.sh is called. Also changed
1086        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
1088    44. To help people who use pcregrep to scan files containing exceedingly long
1089        lines, the following changes have been made:
1091        (a) The default value of the buffer size parameter has been increased from
1092            8K to 20K. (The actual buffer used is three times this size.)
1094        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
1095            PCRE is built.
1097        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
1098            to be set at run time.
1100        (d) Numerical values in pcregrep options can be followed by K or M, for
1101            example --buffer-size=50K.
1103        (e) If a line being scanned overflows pcregrep's buffer, an error is now
1104            given and the return code is set to 2.
1106    45. Add a pointer to the latest mark to the callout data block.
1108    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
1109        partial match of an empty string instead of no match. This was specific to
1110        the use of ".".
1112    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
1113        complete match instead of a partial match. This bug was dependent on both
1114        the PCRE_UTF8 and PCRE_DOTALL options being set.
1116    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1117        starting byte set, because \b was not being ignored.
1120    Version 8.12 15-Jan-2011
1121    ------------------------
1123    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1124        checks for such things as part of the documentation building process.
1126    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1127        --match-limit and --recursion-limit options (added for 8.11). In
1128        particular, this made one of the standard tests fail. (The integer value
1129        went into the wrong half of a long int.)
1131    3.  If the --colour option was given to pcregrep with -v (invert match), it
1132        did strange things, either producing crazy output, or crashing. It should,
1133        of course, ignore a request for colour when reporting lines that do not
1134        match.
1136    4.  Another pcregrep bug caused similar problems if --colour was specified with
1137        -M (multiline) and the pattern match finished with a line ending.
1139    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1140        matched in multiline mode, the following line was shown as part of the
1141        match. This seems wrong, so I have changed it.
1143    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1144        the check for further matches in the same line (so they could be coloured)
1145        to overrun the end of the current line. If another match was found, it was
1146        incorrectly shown (and then shown again when found in the next line).
1148    7.  If pcregrep was compiled under Windows, there was a reference to the
1149        function pcregrep_exit() before it was defined. I am assuming this was
1150        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1151        reported by a user. I've moved the definition above the reference.
1154    Version 8.11 10-Dec-2010
1155    ------------------------
1157    1.  (*THEN) was not working properly if there were untried alternatives prior
1158        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1159        backtracked to try for "b" instead of moving to the next alternative branch
1160        at the same level (in this case, to look for "c"). The Perl documentation
1161        is clear that when (*THEN) is backtracked onto, it goes to the "next
1162        alternative in the innermost enclosing group".
1164    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1165        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1166        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1167        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1168        (*THEN).
1170    3.  If \s appeared in a character class, it removed the VT character from
1171        the class, even if it had been included by some previous item, for example
1172        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1173        of \s, but is part of the POSIX "space" class.)
1175    4.  A partial match never returns an empty string (because you can always
1176        match an empty string at the end of the subject); however the checking for
1177        an empty string was starting at the "start of match" point. This has been
1178        changed to the "earliest inspected character" point, because the returned
1179        data for a partial match starts at this character. This means that, for
1180        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1181        (previously it gave "no match").
1183    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1184        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1185        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1186        has an implication that the given string is incomplete (because a partial
1187        match is preferred over a full match). For this reason, these items now
1188        give a partial match in this situation. [Aside: previously, the one case
1189        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1190        match rather than a full match, which was wrong by the old rules, but is
1191        now correct.]
1193    6.  There was a bug in the handling of #-introduced comments, recognized when
1194        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1195        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1196        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1197        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1198        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1199        places in pcre_compile().
1201    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1202        comments when looking ahead for named forward references to subpatterns,
1203        the only newline sequence it recognized was NL. It now handles newlines
1204        according to the set newline convention.
1206    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1207        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1208        cater for a lack of strerror(). These oversights have been fixed.
1210    9.  Added --match-limit and --recursion-limit to pcregrep.
1212    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1214    11. When the -o option was used, pcregrep was setting a return code of 1, even
1215        when matches were found, and --line-buffered was not being honoured.
1217    12. Added an optional parentheses number to the -o and --only-matching options
1218        of pcregrep.
1220    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1221        can match an empty string. The code to do it in pcretest and pcredemo
1222        needed fixing:
1224        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1225            only one byte after an empty string match just before CRLF (this case
1226            just got forgotten; "any" and "anycrlf" were OK).
1228        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1229            mode when an empty string match preceded an ASCII character followed by
1230            a non-ASCII character. (The code for advancing by one character rather
1231            than one byte was nonsense.)
1233        (c) The pcredemo.c sample program did not have any code at all to handle
1234            the cases when CRLF is a valid newline sequence.
1236    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1237        as a starting offset was within the subject string. There is now a new
1238        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1239        negative or greater than the length of the string. In order to test this,
1240        pcretest is extended to allow the setting of negative starting offsets.
1242    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1243        starting offset points to the beginning of a UTF-8 character was
1244        unnecessarily clumsy. I tidied it up.
1246    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1247        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1249    17. Nobody had reported that the --include_dir option, which was added in
1250        release 7.7 should have been called --include-dir (hyphen, not underscore)
1251        for compatibility with GNU grep. I have changed it to --include-dir, but
1252        left --include_dir as an undocumented synonym, and the same for
1253        --exclude-dir, though that is not available in GNU grep, at least as of
1254        release 2.5.4.
1256    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1257        characters from a string of bytes) have been redefined so as not to use
1258        loops, in order to improve performance in some environments. At the same
1259        time, I abstracted some of the common code into auxiliary macros to save
1260        repetition (this should not affect the compiled code).
1262    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1263        compile-time error is now given if \c is not followed by an ASCII
1264        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1265        different, and any byte value is allowed.)
1267    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1268        START_OPTIMIZE option, which is now allowed at compile time - but just
1269        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1270        to pcregrep and other applications that have no direct access to PCRE
1271        options. The new /Y option in pcretest sets this option when calling
1272        pcre_compile().
1274    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1275        back references. Groups containing recursive back references were forced to
1276        be atomic by that change, but in the case of named groups, the amount of
1277        memory required was incorrectly computed, leading to "Failed: internal
1278        error: code overflow". This has been fixed.
1280    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1281        pcretest.c, to avoid build problems in some Borland environments.
1284    Version 8.10 25-Jun-2010
1285    ------------------------
1287    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1288        THEN.
1290    2.  (*ACCEPT) was not working when inside an atomic group.
1292    3.  Inside a character class, \B is treated as a literal by default, but
1293        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1294        causes the error). The code is unchanged, but I tidied the documentation.
1296    4.  Inside a character class, PCRE always treated \R and \X as literals,
1297        whereas Perl faults them if its -w option is set. I have changed PCRE so
1298        that it faults them when PCRE_EXTRA is set.
1300    5.  Added support for \N, which always matches any character other than
1301        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1303    6.  When compiling pcregrep with newer versions of gcc which may have
1304        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1305        declared with attribute warn_unused_result" were given. Just casting the
1306        result to (void) does not stop the warnings; a more elaborate fudge is
1307        needed. I've used a macro to implement this.
1309    7.  Minor change to pcretest.c to avoid a compiler warning.
1311    8.  Added four artifical Unicode properties to help with an option to make
1312        \s etc use properties (see next item). The new properties are: Xan
1313        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1315    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1316        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1317        this option. Modified pcretest to add /W to test this facility. Added
1318        REG_UCP to make it available via the POSIX interface.
1320    10. Added --line-buffered to pcregrep.
1322    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1323        studied, and the match started with a letter with a code point greater than
1324        127 whose first byte was different to the first byte of the other case of
1325        the letter, the other case of this starting letter was not recognized
1326        (#976).
1328    12. If a pattern that was studied started with a repeated Unicode property
1329        test, for example, \p{Nd}+, there was the theoretical possibility of
1330        setting up an incorrect bitmap of starting bytes, but fortunately it could
1331        not have actually happened in practice until change 8 above was made (it
1332        added property types that matched character-matching opcodes).
1334    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1335        possible starting bytes for non-anchored patterns.
1337    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1338        \R, and also a number of cases that involve Unicode properties, both
1339        explicit and implicit when PCRE_UCP is set.
1341    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1342        input, it could crash or give wrong results if characters with values
1343        greater than 0xc0 were present in the subject string. (Detail: it assumed
1344        UTF-8 input when processing these items.)
1346    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1347        size_t is 64-bit (#991).
1349    17. Added a check for running out of memory when PCRE is compiled with
1350        --disable-stack-for-recursion (#990).
1352    18. If the last data line in a file for pcretest does not have a newline on
1353        the end, a newline was missing in the output.
1355    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1356        less than 128) in its various bitmaps. However, there is a facility for
1357        generating tables according to the current locale when PCRE is compiled. It
1358        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1359        characters, are recognized by isspace() and therefore were getting set in
1360        these tables, and indeed these tables seem to approximate to ISO 8859. This
1361        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1362        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1363        which of course cannot start UTF-8 characters. I have changed the code so
1364        that only real ASCII characters (less than 128) and the correct starting
1365        bytes for UTF-8 encodings are set for characters greater than 127 when in
1366        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1367        altogether.)
1369    20. Added the /T option to pcretest so as to be able to run tests with non-
1370        standard character tables, thus making it possible to include the tests
1371        used for 19 above in the standard set of tests.
1373    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1374        reference to a subpattern the other side of a comment that contains an
1375        opening parenthesis caused either an internal compiling error, or a
1376        reference to the wrong subpattern.
1379    Version 8.02 19-Mar-2010
1380    ------------------------
1382    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1384    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1385        configured.
1387    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1388        original author of that file, following a query about its status.
1390    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1391        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1393    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1394        quantifier applied to a forward-referencing subroutine call, could compile
1395        incorrect code or give the error "internal error: previously-checked
1396        referenced subpattern not found".
1398    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1399        variables to point to external functions. For these systems, therefore,
1400        pcre_malloc etc. are now initialized to local functions that call the
1401        relevant global functions.
1403    7.  There were two entries missing in the vectors called coptable and poptable
1404        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1405        I've fixed the data, and added a kludgy way of testing at compile time that
1406        the lengths are correct (equal to the number of opcodes).
1408    8.  Following on from 7, I added a similar kludge to check the length of the
1409        eint vector in pcreposix.c.
1411    9.  Error texts for pcre_compile() are held as one long string to avoid too
1412        much relocation at load time. To find a text, the string is searched,
1413        counting zeros. There was no check for running off the end of the string,
1414        which could happen if a new error number was added without updating the
1415        string.
1417    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1419    11. \K was not working if it appeared in an atomic group or in a group that
1420        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1421        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1422        the assertion is positive, but not if it is negative.
1424    12. Change 11 fortuitously reduced the size of the stack frame used in the
1425        "match()" function of pcre_exec.c by one pointer. Forthcoming
1426        implementation of support for (*MARK) will need an extra pointer on the
1427        stack; I have reserved it now, so that the stack frame size does not
1428        decrease.
1430    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1431        item in branch that calls a recursion is a subroutine call - as in the
1432        second branch in the above example - was incorrectly given the compile-
1433        time error "recursive call could loop indefinitely" because pcre_compile()
1434        was not correctly checking the subroutine for matching a non-empty string.
1436    14. The checks for overrunning compiling workspace could trigger after an
1437        overrun had occurred. This is a "should never occur" error, but it can be
1438        triggered by pathological patterns such as hundreds of nested parentheses.
1439        The checks now trigger 100 bytes before the end of the workspace.
1441    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1444    Version 8.01 19-Jan-2010
1445    ------------------------
1447    1.  If a pattern contained a conditional subpattern with only one branch (in
1448        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1449        computed the wrong minimum data length (which is of course zero for such
1450        subpatterns). This could cause incorrect "no match" results.
1452    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1453        the pattern is reset in the first branch, pcre_compile() failed with
1454        "internal error: code overflow at offset...". This happened only when
1455        the reset was to the original external option setting. (An optimization
1456        abstracts leading options settings into an external setting, which was the
1457        cause of this.)
1459    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1460        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1461        assertion pattern did not match (meaning that the assertion was true), it
1462        was incorrectly treated as false if the SKIP had been reached during the
1463        matching. This also applied to assertions used as conditions.
1465    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1466        assertion subpattern, including such a pattern used as a condition,
1467        unpredictable results occurred, instead of the error return
1470    5.  The C++ GlobalReplace function was not working like Perl for the special
1471        situation when an empty string is matched. It now does the fancy magic
1472        stuff that is necessary.
1474    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1475        removed. (These were left over from very, very early versions of PCRE.)
1477    7.  Some cosmetic changes to the code to make life easier when compiling it
1478        as part of something else:
1480        (a) Change DEBUG to PCRE_DEBUG.
1482        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1483            called "current" as "current_branch", to prevent a collision with the
1484            Linux macro when compiled as a kernel module.
1486        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1487            prevent a collision with the Linux macro when compiled as a kernel
1488            module.
1490    8.  In pcre_compile() there are some checks for integer overflows that used to
1491        cast potentially large values to (double). This has been changed to that
1492        when building, a check for int64_t is made, and if it is found, it is used
1493        instead, thus avoiding the use of floating point arithmetic. (There is no
1494        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1495        double.
1497    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1498        2005 (difference between two addresses compared to an unsigned value).
1500    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1501        custom one, because of the following reported problem in Windows:
1503          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1504              under Win32.
1505          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1506              therefore missing the function definition.
1507          - The compiler thus generates a "C" signature for the test function.
1508          - The linker fails to find the "C" function.
1509          - PCRE fails to configure if asked to do so against libbz2.
1511    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1512        messages were output:
1514          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1515          rerunning libtoolize, to keep the correct libtool macros in-tree.
1516          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1518        I have done both of these things.
1520    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1521        most of the time, it *can* run out if it is given a pattern that contains a
1522        runaway infinite recursion. I updated the discussion in the pcrestack man
1523        page.
1525    13. Now that we have gone to the x.xx style of version numbers, the minor
1526        version may start with zero. Using 08 or 09 is a bad idea because users
1527        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1528        interpreted as invalid octal numbers. I've updated the previous comment in
1529        configure.ac, and also added a check that gives an error if 08 or 09 are
1530        used.
1532    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1533        causing partial matching to fail when the end of the subject matched \W
1534        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1536    15. There were some discrepancies between the declarations in pcre_internal.h
1537        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1538        their definitions. The declarations used "const uschar *" and the
1539        definitions used USPTR. Even though USPTR is normally defined as "const
1540        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1541        reported that: "This difference in casting confuses some C++ compilers, for
1542        example, SunCC recognizes above declarations as different functions and
1543        generates broken code for hbpcre." I have changed the declarations to use
1544        USPTR.
1546    16. GNU libtool is named differently on some systems. The autogen.sh script now
1547        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1548        (FreeBSD).
1550    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1551        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1552        comment: "Figure out how to create a longlong from a string: strtoll and
1553        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1554        instance, but it only takes 2 args instead of 3!"
1556    18. A subtle bug concerned with back references has been fixed by a change of
1557        specification, with a corresponding code fix. A pattern such as
1558        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1559        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1560        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1561        same bug. Such groups have to be quantified to be useful, or contained
1562        inside another quantified group. (If there's no repetition, the reference
1563        can never match.) The problem arises because, having left the group and
1564        moved on to the rest of the pattern, a later failure that backtracks into
1565        the group uses the captured value from the final iteration of the group
1566        rather than the correct earlier one. I have fixed this in PCRE by forcing
1567        any group that contains a reference to itself to be an atomic group; that
1568        is, there cannot be any backtracking into it once it has completed. This is
1569        similar to recursive and subroutine calls.
1572    Version 8.00 19-Oct-09
1573    ----------------------
1575    1.  The table for translating pcre_compile() error codes into POSIX error codes
1576        was out-of-date, and there was no check on the pcre_compile() error code
1577        being within the table. This could lead to an OK return being given in
1578        error.
1580    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1581        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1582        in a Windows environment.
1584    3.  The pcregrep --count option prints the count for each file even when it is
1585        zero, as does GNU grep. However, pcregrep was also printing all files when
1586        --files-with-matches was added. Now, when both options are given, it prints
1587        counts only for those files that have at least one match. (GNU grep just
1588        prints the file name in this circumstance, but including the count seems
1589        more useful - otherwise, why use --count?) Also ensured that the
1590        combination -clh just lists non-zero counts, with no names.
1592    4.  The long form of the pcregrep -F option was incorrectly implemented as
1593        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1594        but it seems right to fix it, and I didn't think it was worth preserving
1595        the old behaviour.
1597    5.  The command line items --regex=pattern and --regexp=pattern were not
1598        recognized by pcregrep, which required --regex pattern or --regexp pattern
1599        (with a space rather than an '='). The man page documented the '=' forms,
1600        which are compatible with GNU grep; these now work.
1602    6.  No libpcreposix.pc file was created for pkg-config; there was just
1603        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1605    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1606        when UCP support is not needed, by modifying the Python script that
1607        generates it from Unicode data files. This should not matter if the module
1608        is correctly used as a library, but I received one complaint about 50K of
1609        unwanted data. My guess is that the person linked everything into his
1610        program rather than using a library. Anyway, it does no harm.
1612    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1613        was a minimum greater than 1 for a wide character in a possessive
1614        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1615        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1616        character. Chaos in the form of incorrect output or a compiling loop could
1617        result.
1619    9.  The restrictions on what a pattern can contain when partial matching is
1620        requested for pcre_exec() have been removed. All patterns can now be
1621        partially matched by this function. In addition, if there are at least two
1622        slots in the offset vector, the offset of the earliest inspected character
1623        for the match and the offset of the end of the subject are set in them when
1624        PCRE_ERROR_PARTIAL is returned.
1626    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1627        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1628        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1629        and may be more useful for multi-segment matching.
1631    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1632        used to be given if ever the end of the subject was reached; now it is
1633        given only if matching could not proceed because another character was
1634        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1635        string "Z", which now yields "no match" instead of "partial match". In the
1636        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1637        final character ended with (*FAIL).
1639    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1640        if the pattern had a "must contain" character that was already found in the
1641        earlier partial match, unless partial matching was again requested. For
1642        example, with the pattern /dog.(body)?/, the "must contain" character is
1643        "g". If the first part-match was for the string "dog", restarting with
1644        "sbody" failed. This bug has been fixed.
1646    13. The string returned by pcre_dfa_exec() after a partial match has been
1647        changed so that it starts at the first inspected character rather than the
1648        first character of the match. This makes a difference only if the pattern
1649        starts with a lookbehind assertion or \b or \B (\K is not supported by
1650        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1651        matching functions compatible, and I think it's the right thing to do.
1653    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1654        so that the demonstration program is easily available in environments where
1655        PCRE has not been installed from source.
1657    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1658        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1659        library.
1661    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1662        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1663        is not the first non-POSIX option to be added. Clearly some people find
1664        these options useful.
1666    17. If a caller to the POSIX matching function regexec() passes a non-zero
1667        value for nmatch with a NULL value for pmatch, the value of
1668        nmatch is forced to zero.
1670    18. RunGrepTest did not have a test for the availability of the -u option of
1671        the diff command, as RunTest does. It now checks in the same way as
1672        RunTest, and also checks for the -b option.
1674    19. If an odd number of negated classes containing just a single character
1675        interposed, within parentheses, between a forward reference to a named
1676        subpattern and the definition of the subpattern, compilation crashed with
1677        an internal error, complaining that it could not find the referenced
1678        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1679        [The bug was that it was starting one character too far in when skipping
1680        over the character class, thus treating the ] as data rather than
1681        terminating the class. This meant it could skip too much.]
1683    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1684        /g option in pcretest when the pattern contains \K, which makes it possible
1685        to have an empty string match not at the start, even when the pattern is
1686        anchored. Updated pcretest and pcredemo to use this option.
1688    21. If the maximum number of capturing subpatterns in a recursion was greater
1689        than the maximum at the outer level, the higher number was returned, but
1690        with unset values at the outer level. The correct (outer level) value is
1691        now given.
1693    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1694        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1695        make it do so. The string so far is captured, making this feature
1696        compatible with Perl.
1698    23. The tests have been re-organized, adding tests 11 and 12, to make it
1699        possible to check the Perl 5.10 features against Perl 5.10.
1701    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1702        pattern matches a fixed length string. PCRE did not allow this; now it
1703        does. Neither allows recursion.
1705    25. I finally figured out how to implement a request to provide the minimum
1706        length of subject string that was needed in order to match a given pattern.
1707        (It was back references and recursion that I had previously got hung up
1708        on.) This code has now been added to pcre_study(); it finds a lower bound
1709        to the length of subject needed. It is not necessarily the greatest lower
1710        bound, but using it to avoid searching strings that are too short does give
1711        some useful speed-ups. The value is available to calling programs via
1712        pcre_fullinfo().
1714    26. While implementing 25, I discovered to my embarrassment that pcretest had
1715        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1716        study optimizations had never been tested with that matching function.
1717        Oops. What is worse, even when it was passed study data, there was a bug in
1718        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1719        were also very few tests of studied patterns with pcre_dfa_exec().
1721    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1722        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1723        on the other side of the coin, they are no longer allowed to have different
1724        names, because these cannot be distinguished in PCRE, and this has caused
1725        confusion. (This is a difference from Perl.)
1727    28. When duplicate subpattern names are present (necessarily with different
1728        numbers, as required by 27 above), and a test is made by name in a
1729        conditional pattern, either for a subpattern having been matched, or for
1730        recursion in such a pattern, all the associated numbered subpatterns are
1731        tested, and the overall condition is true if the condition is true for any
1732        one of them. This is the way Perl works, and is also more like the way
1733        testing by number works.
1736    Version 7.9 11-Apr-09
1737    ---------------------
1739    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1740        (pcretest), all targets were linked against these libraries. This included
1741        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1742        libraries. This caused unwanted dependencies to be created. This problem
1743        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1744        pcretest is linked with readline.
1746    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1747        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1748        moved outside it again, because FALSE and TRUE are already defined in AIX,
1749        but BOOL is not.
1751    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1752        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1754    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1755        hyphens) following file names and line numbers when outputting matching
1756        lines. This is not true; no spaces are inserted. I have also clarified the
1757        wording for the --colour (or --color) option.
1759    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1760        was not coloured; this is different to GNU grep, so I have changed it to be
1761        the same.
1763    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1764        each matching line was coloured. Now it goes on to look for further matches
1765        of any of the test patterns, which is the same behaviour as GNU grep.
1767    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1768        doesn't make sense to accept an empty string match in pcregrep, so I have
1769        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1770        seems to be how GNU grep behaves. [But see later change 40 for release
1771        8.33.]
1773    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1774        start or after a newline", because the conditional assertion was not being
1775        correctly handled. The rule now is that both the assertion and what follows
1776        in the first alternative must satisfy the test.
1778    9.  If auto-callout was enabled in a pattern with a conditional group whose
1779        condition was an assertion, PCRE could crash during matching, both with
1780        pcre_exec() and pcre_dfa_exec().
1782    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1783        used for matching.
1785    11. Unicode property support in character classes was not working for
1786        characters (bytes) greater than 127 when not in UTF-8 mode.
1788    12. Added the -M command line option to pcretest.
1790    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1792    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1794    16. Added comments and documentation about mis-use of no_arg in the C++
1795        wrapper.
1797    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1798        from Martin Jerabek that uses macro names for all relevant character and
1799        string constants.
1801    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1802        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1803        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1804        these, but not everybody uses configure.
1806    19. A conditional group that had only one branch was not being correctly
1807        recognized as an item that could match an empty string. This meant that an
1808        enclosing group might also not be so recognized, causing infinite looping
1809        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1810        with the subject "ab", where knowledge that the repeated group can match
1811        nothing is needed in order to break the loop.
1813    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1814        exec(), but without supplying a callout function, matching went wrong.
1816    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1817        leak if the size of the offset vector was greater than 30. When the vector
1818        is smaller, the saved offsets during recursion go onto a local stack
1819        vector, but for larger vectors malloc() is used. It was failing to free
1820        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1821        error, in fact).
1823    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1824        heapframe that is used only when UTF-8 support is enabled. This caused no
1825        problem, but was untidy.
1827    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1828        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1829        included within another project.
1831    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1832        slightly modified by me:
1834          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1835              not building pcregrep.
1837          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1838              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1840    25. Forward references, both numeric and by name, in patterns that made use of
1841        duplicate group numbers, could behave incorrectly or give incorrect errors,
1842        because when scanning forward to find the reference group, PCRE was not
1843        taking into account the duplicate group numbers. A pattern such as
1844        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1846    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1847        the feature of a custom pointer more persuasive (as requested by a user).
1849    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1850        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1851        pre-defined.
1853    28. Added support for (*UTF8) at the start of a pattern.
1855    29. Arrange for flags added by the "release type" setting in CMake to be shown
1856        in the configuration summary.
1859    Version 7.8 05-Sep-08
1860    ---------------------
1862    1.  Replaced UCP searching code with optimized version as implemented for Ad
1863        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1864        stage table and inline lookup instead of a function, giving speed ups of 2
1865        to 5 times on some simple patterns that I tested. Permission was given to
1866        distribute the MultiStage2.py script that generates the tables (it's not in
1867        the tarball, but is in the Subversion repository).
1869    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1870        scripts.
1872    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1873        a group with a zero qualifier. The result of the study could be incorrect,
1874        or the function might crash, depending on the pattern.
1876    4.  Caseless matching was not working for non-ASCII characters in back
1877        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1878        It now works when Unicode Property Support is available.
1880    5.  In pcretest, an escape such as \x{de} in the data was always generating
1881        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1882        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1883        truncation.
1885    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1887    7.  Added two (int) casts to pcregrep when printing the difference of two
1888        pointers, in case they are 64-bit values.
1890    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1891        test 2 if it fails.
1893    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1894        and a #define of that name to empty if it is not externally set. This is to
1895        allow users of MSVC to set it if necessary.
1897    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1898        the convenience functions in the pcre_get.c source file.
1900    11. An option change at the start of a pattern that had top-level alternatives
1901        could cause overwriting and/or a crash. This command provoked a crash in
1902        some environments:
1904          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1906        This potential security problem was recorded as CVE-2008-2371.
1908    12. For a pattern where the match had to start at the beginning or immediately
1909        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1910        pcre_dfa_exec() could read past the end of the passed subject if there was
1911        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1912        pcretest so that it places the subject at the end of its malloc-ed buffer.
1914    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1915        exec() might read past the end of the data buffer in UTF-8 mode.
1917    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1918        the data contained the byte 0x85 as part of a UTF-8 character within its
1919        first line. This applied both to normal and DFA matching.
1921    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1922        /^[^d]*?$/8 failed to match "abc".
1924    16. Added a missing copyright notice to pcrecpp_internal.h.
1926    17. Make it more clear in the documentation that values returned from
1927        pcre_exec() in ovector are byte offsets, not character counts.
1929    18. Tidied a few places to stop certain compilers from issuing warnings.
1931    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1932        supplied by Stefan Weber. I made a further small update for 7.8 because
1933        there is a change of source arrangements: the pcre_searchfuncs.c module is
1934        replaced by pcre_ucd.c.
1937    Version 7.7 07-May-08
1938    ---------------------
1940    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1941        a string to a long long, pretend we don't even have a long long." This is
1942        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1944    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1945        pre-7.6 versions, which defined a global no_arg variable instead of putting
1946        it in the RE class. (See also #8 below.)
1948    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1949        Lopes.
1951    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1953        (1) The include/exclude patterns were being applied to the whole pathnames
1954            of files, instead of just to the final components.
1956        (2) If there was more than one level of directory, the subdirectories were
1957            skipped unless they satisfied the include/exclude conditions. This is
1958            inconsistent with GNU grep (and could even be seen as contrary to the
1959            pcregrep specification - which I improved to make it absolutely clear).
1960            The action now is always to scan all levels of directory, and just
1961            apply the include/exclude patterns to regular files.
1963    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1964        --exclude_dir in the tests to avoid scanning .svn directories.
1966    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1967        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1968        doesn't support NULs in patterns.
1970    7.  Added some missing "const"s to declarations of static tables in
1971        pcre_compile.c and pcre_dfa_exec.c.
1973    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1974        caused by fix #2  above. (Subsequently also a second patch to fix the
1975        first patch. And a third patch - this was a messy problem.)
1977    9.  Applied Craig's patch to remove the use of push_back().
1979    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1980        matching function regexec().
1982    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1983        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1984        references. PCRE supports relative numbers with this syntax (I don't think
1985        Oniguruma does).
1987    12. Previously, a group with a zero repeat such as (...){0} was completely
1988        omitted from the compiled regex. However, this means that if the group
1989        was called as a subroutine from elsewhere in the pattern, things went wrong
1990        (an internal error was given). Such groups are now left in the compiled
1991        pattern, with a new opcode that causes them to be skipped at execution
1992        time.
1994    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1995        to the way PCRE behaves:
1997        (a) A lone ] character is dis-allowed (Perl treats it as data).
1999        (b) A back reference to an unmatched subpattern matches an empty string
2000            (Perl fails the current match path).
2002        (c) A data ] in a character class must be notated as \] because if the
2003            first data character in a class is ], it defines an empty class. (In
2004            Perl it is not possible to have an empty class.) The empty class []
2005            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
2006            The negative empty class [^] matches any one character, independently
2007            of the DOTALL setting.
2009    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
2010        non-existent subpattern following a character class starting with ']' and
2011        containing () gave an internal compiling error instead of "reference to
2012        non-existent subpattern". Fortunately, when the pattern did exist, the
2013        compiled code was correct. (When scanning forwards to check for the
2014        existence of the subpattern, it was treating the data ']' as terminating
2015        the class, so got the count wrong. When actually compiling, the reference
2016        was subsequently set up correctly.)
2018    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
2019        it was being rejected as not supported by pcre_dfa_exec(), even though
2020        other assertions are supported. I have made pcre_dfa_exec() support
2021        (*FAIL).
2023    16. The implementation of 13c above involved the invention of a new opcode,
2024        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
2025        cannot be changed at match time, I realized I could make a small
2026        improvement to matching performance by compiling OP_ALLANY instead of
2027        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
2028        on the OP_ANY path.
2030    17. Compiling pcretest on Windows with readline support failed without the
2031        following two fixes: (1) Make the unistd.h include conditional on
2032        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
2034    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
2035        ncurses library to be included for pcretest when ReadLine support is
2036        requested, but also to allow for it to be overridden. This patch came from
2037        Daniel Bergström.
2039    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
2040        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
2041        any errors with the current Unicode tables. Thanks to Peter Kankowski for
2042        spotting this.
2045    Version 7.6 28-Jan-08
2046    ---------------------
2048    1.  A character class containing a very large number of characters with
2049        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
2050        overflow.
2052    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
2053        HAVE_LONG_LONG is not defined.
2055    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
2056        bring them up to date and include new features. This patch includes:
2058        - Fixed PH's badly added libz and libbz2 support.
2059        - Fixed a problem with static linking.
2060        - Added pcredemo. [But later removed - see 7 below.]
2061        - Fixed dftables problem and added an option.
2062        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
2063            HAVE_LONG_LONG.
2064        - Added readline support for pcretest.
2065        - Added an listing of the option settings after cmake has run.
2067    4.  A user submitted a patch to Makefile that makes it easy to create
2068        "pcre.dll" under mingw when using Configure/Make. I added stuff to
2069        Makefile.am that cause it to include this special target, without
2070        affecting anything else. Note that the same mingw target plus all
2071        the other distribution libraries and programs are now supported
2072        when configuring with CMake (see 6 below) instead of with
2073        Configure/Make.
2075    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
2076        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
2077        exported in the Windows port". It has not yet been confirmed that the patch
2078        solves the problem, but it does no harm.
2080    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
2081        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
2082        with CMake, and also correct the comment about stack recursion.
2084    7.  Remove the automatic building of pcredemo from the ./configure system and
2085        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
2086        of a program that users should build themselves after PCRE is installed, so
2087        building it automatically is not really right. What is more, it gave
2088        trouble in some build environments.
2090    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
2093    Version 7.5 10-Jan-08
2094    ---------------------
2096    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
2097        values in parens when parsing an RE using the C++ wrapper."
2099    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
2100        Characters greater than 255 were excluded from the class instead of being
2101        included.
2103    3.  The same bug as (2) above applied to negated POSIX classes such as
2104        [:^space:].
2106    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
2107        defined or documented. It seems to have been a typo for PCRE_STATIC, so
2108        I have changed it.
2110    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
2111        first named subpattern) and a construct such as (?&a) would reference the
2112        first named subpattern whose name started with "a" (in other words, the
2113        length check was missing). Both these problems are fixed. "Subpattern name
2114        expected" is now given for (?&) (a zero-length name), and this patch also
2115        makes it give the same error for \k'' (previously it complained that that
2116        was a reference to a non-existent subpattern).
2118    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
2119        this is right because (?- can be followed by option settings as well as by
2120        digits. I have, however, made the messages clearer.
2122    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
2123        than the number used in the conditional) now cause a compile-time error.
2124        This is actually not compatible with Perl, which accepts such patterns, but
2125        treats the conditional as always being FALSE (as PCRE used to), but it
2126        seems to me that giving a diagnostic is better.
2128    8.  Change "alphameric" to the more common word "alphanumeric" in comments
2129        and messages.
2131    9.  Fix two occurrences of "backslash" in comments that should have been
2132        "backspace".
2134    10. Remove two redundant lines of code that can never be obeyed (their function
2135        was moved elsewhere).
2137    11. The program that makes PCRE's Unicode character property table had a bug
2138        which caused it to generate incorrect table entries for sequences of
2139        characters that have the same character type, but are in different scripts.
2140        It amalgamated them into a single range, with the script of the first of
2141        them. In other words, some characters were in the wrong script. There were
2142        thirteen such cases, affecting characters in the following ranges:
2144          U+002b0 - U+002c1
2145          U+0060c - U+0060d
2146          U+0061e - U+00612
2147          U+0064b - U+0065e
2148          U+0074d - U+0076d
2149          U+01800 - U+01805
2150          U+01d00 - U+01d77
2151          U+01d9b - U+01dbf
2152          U+0200b - U+0200f
2153          U+030fc - U+030fe
2154          U+03260 - U+0327f
2155          U+0fb46 - U+0fbb1
2156          U+10450 - U+1049d
2158    12. The -o option (show only the matching part of a line) for pcregrep was not
2159        compatible with GNU grep in that, if there was more than one match in a
2160        line, it showed only the first of them. It now behaves in the same way as
2161        GNU grep.
2163    13. If the -o and -v options were combined for pcregrep, it printed a blank
2164        line for every non-matching line. GNU grep prints nothing, and pcregrep now
2165        does the same. The return code can be used to tell if there were any
2166        non-matching lines.
2168    14. Added --file-offsets and --line-offsets to pcregrep.
2170    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
2171        infinitely looping recursion. The bug was that positive lookaheads were not
2172        being skipped when checking for a possible empty match (negative lookaheads
2173        and both kinds of lookbehind were skipped).
2175    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
2176        inclusion of <windows.h> to before rather than after the definition of
2177        INVALID_FILE_ATTRIBUTES (patch from David Byron).
2179    17. Specifying a possessive quantifier with a specific limit for a Unicode
2180        character property caused pcre_compile() to compile bad code, which led at
2181        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
2182        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
2183        caused the error; without that there was no problem.
2185    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
2187    19. Added --enable-pcretest-libreadline.
2189    20. In pcrecpp.cc, the variable 'count' was incremented twice in
2190        RE::GlobalReplace(). As a result, the number of replacements returned was
2191        double what it should be. I removed one of the increments, but Craig sent a
2192        later patch that removed the other one (the right fix) and added unit tests
2193        that check the return values (which was not done before).
2195    21. Several CMake things:
2197        (1) Arranged that, when cmake is used on Unix, the libraries end up with
2198            the names libpcre and libpcreposix, not just pcre and pcreposix.
2200        (2) The above change means that pcretest and pcregrep are now correctly
2201            linked with the newly-built libraries, not previously installed ones.
2205    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
2206        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
2207        UTF-8 newline character). The key issue is that the pattern starts .*;
2208        this means that the match must be either at the beginning, or after a
2209        newline. The bug was in the code for advancing after a failed match and
2210        checking that the new position followed a newline. It was not taking
2211        account of UTF-8 characters correctly.
2213    23. PCRE was behaving differently from Perl in the way it recognized POSIX
2214        character classes. PCRE was not treating the sequence [:...:] as a
2215        character class unless the ... were all letters. Perl, however, seems to
2216        allow any characters between [: and :], though of course it rejects as
2217        unknown any "names" that contain non-letters, because all the known class
2218        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
2219        for example, whereas PCRE did not - it did not recognize a POSIX character
2220        class. This seemed a bit dangerous, so the code has been changed to be
2221        closer to Perl. The behaviour is not identical to Perl, because PCRE will
2222        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
2223        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
2224        Perl does, and where it didn't before.
2226    24. Rewrite so as to remove the single use of %n from pcregrep because in some
2227        Windows environments %n is disabled by default.
2230    Version 7.4 21-Sep-07
2231    ---------------------
2233    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
2234        means that a class such as [\s] counted as "explicit reference to CR or
2235        LF". That isn't really right - the whole point of the change was to try to
2236        help when there was an actual mention of one of the two characters. So now
2237        the change happens only if \r or \n (or a literal CR or LF) character is
2238        encountered.
2240    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
2241        of both had grown to the point where there were only 3 bits left.
2242        Fortunately, there was spare space in the data structure, and so I have
2243        moved the internal flags into a new 16-bit field to free up more option
2244        bits.
2246    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
2247        but did not set the internal JCHANGED flag - either of these is enough to
2248        control the way the "get" function works - but the PCRE_INFO_JCHANGED
2249        facility is supposed to tell if (?J) was ever used, so now (?J) at the
2250        start sets both bits.
2252    4.  Added options (at build time, compile time, exec time) to change \R from
2253        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
2255    5.  doc/pcresyntax.html was missing from the distribution.
2257    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
2258        compatibility, even though it is no longer used.
2260    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
2261        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
2262        windows.h file is present (where different names are used). [This was
2263        reversed later after testing - see 16 below.]
2265    8.  Changed all #include <config.h> to #include "config.h". There were also
2266        some further <pcre.h> cases that I changed to "pcre.h".
2268    9.  When pcregrep was used with the --colour option, it missed the line ending
2269        sequence off the lines that it output.
2271    10. It was pointed out to me that arrays of string pointers cause lots of
2272        relocations when a shared library is dynamically loaded. A technique of
2273        using a single long string with a table of offsets can drastically reduce
2274        these. I have refactored PCRE in four places to do this. The result is
2275        dramatic:
2277          Originally:                          290
2278          After changing UCP table:            187
2279          After changing error message table:   43
2280          After changing table of "verbs"       36
2281          After changing table of Posix names   22
2283        Thanks to the folks working on Gregex for glib for this insight.
2285    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
2286        unicode-properties was also set.
2288    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
2290    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
2291        checked only for CRLF.
2293    14. Added casts to pcretest.c to avoid compiler warnings.
2295    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
2297    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
2298        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2299        entirely. This removes changes made in 7 above.
2301    17. The CMake files have been updated, and there is now more information about
2302        building with CMake in the NON-UNIX-USE document.
2305    Version 7.3 28-Aug-07
2306    ---------------------
2308     1. In the rejigging of the build system that eventually resulted in 7.1, the
2309        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
2310        brackets there is not right, since it causes compilers to look for an
2311        installed pcre.h, not the version that is in the source that is being
2312        compiled (which of course may be different). I have changed it back to:
2314          #include "pcre.h"
2316        I have a vague recollection that the change was concerned with compiling in
2317        different directories, but in the new build system, that is taken care of
2318        by the VPATH setting the Makefile.
2320     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
2321        when the subject happened to end in the byte 0x85 (e.g. if the last
2322        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
2323        characters but of course it shouldn't be taken as a newline when it is part
2324        of another character. The bug was that, for an unlimited repeat of . in
2325        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
2326        characters when looking for a newline.
2328     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
2330     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
2331        in debug output.
2333     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
2334        long printing in the pcrecpp unittest when running under MinGW.
2336     6. ESC_K was left out of the EBCDIC table.
2338     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
2339        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
2340        limit also applies to "virtual nesting" when a pattern is recursive, and in
2341        this case 1000 isn't so big. I have been able to remove this limit at the
2342        expense of backing off one optimization in certain circumstances. Normally,
2343        when pcre_exec() would call its internal match() function recursively and
2344        immediately return the result unconditionally, it uses a "tail recursion"
2345        feature to save stack. However, when a subpattern that can match an empty
2346        string has an unlimited repetition quantifier, it no longer makes this
2347        optimization. That gives it a stack frame in which to save the data for
2348        checking that an empty string has been matched. Previously this was taken
2349        from the 1000-entry workspace that had been reserved. So now there is no
2350        explicit limit, but more stack is used.
2352     8. Applied Daniel's patches to solve problems with the import/export magic
2353        syntax that is required for Windows, and which was going wrong for the
2354        pcreposix and pcrecpp parts of the library. These were overlooked when this
2355        problem was solved for the main library.
2357     9. There were some crude static tests to avoid integer overflow when computing
2358        the size of patterns that contain repeated groups with explicit upper
2359        limits. As the maximum quantifier is 65535, the maximum group length was
2360        set at 30,000 so that the product of these two numbers did not overflow a
2361        32-bit integer. However, it turns out that people want to use groups that
2362        are longer than 30,000 bytes (though not repeat them that many times).
2363        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
2364        made it possible to implement the integer overflow checks in a much more
2365        dynamic way, which I have now done. The artificial limitation on group
2366        length has been removed - we now have only the limit on the total length of
2367        the compiled pattern, which depends on the LINK_SIZE setting.
2369    10. Fixed a bug in the documentation for get/copy named substring when
2370        duplicate names are permitted. If none of the named substrings are set, the
2371        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
2372        empty string.
2374    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
2375        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
2376        because the ] is interpreted as the first data character and the
2377        terminating ] is not found. PCRE has been made compatible with Perl in this
2378        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
2379        cause memory overwriting.
2381    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
2382        string has been matched (to stop an infinite loop). It was not recognizing
2383        a conditional subpattern that could match an empty string if that
2384        subpattern was within another subpattern. For example, it looped when
2385        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
2386        condition was not nested. This bug has been fixed.
2388    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
2389        past the start of the subject in the presence of bytes with the top bit
2390        set, for example "\x8aBCD".
2392    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
2393        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
2395    14. Optimized (?!) to (*FAIL).
2397    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
2398        This restricts code points to be within the range 0 to 0x10FFFF, excluding
2399        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
2400        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
2401        does: it's just the validity check that is more restrictive.
2403    16. Inserted checks for integer overflows during escape sequence (backslash)
2404        processing, and also fixed erroneous offset values for syntax errors during
2405        backslash processing.
2407    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
2408        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
2410    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
2411        caused an overrun.
2413    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
2414        something other than just ASCII characters) inside a group that had an
2415        unlimited repeat caused a loop at compile time (while checking to see
2416        whether the group could match an empty string).
2418    20. Debugging a pattern containing \p or \P could cause a crash. For example,
2419        [\P{Any}] did so. (Error in the code for printing property names.)
2421    21. An orphan \E inside a character class could cause a crash.
2423    22. A repeated capturing bracket such as (A)? could cause a wild memory
2424        reference during compilation.
2426    23. There are several functions in pcre_compile() that scan along a compiled
2427        expression for various reasons (e.g. to see if it's fixed length for look
2428        behind). There were bugs in these functions when a repeated \p or \P was
2429        present in the pattern. These operators have additional parameters compared
2430        with \d, etc, and these were not being taken into account when moving along
2431        the compiled data. Specifically:
2433        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2434            length.
2436        (b) An item such as \pL+ within a repeated group could cause crashes or
2437            loops.
2439        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2440            "reference to non-existent subpattern" error.
2442        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2444    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2445        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2447    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2449    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2450        character were causing crashes (broken optimization).
2452    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2453        \p or \P) caused a compile-time loop.
2455    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2456        break. For example, the unstudied pattern [\r\n]A does not match the string
2457        "\r\nA" because change 7.0/46 below moves the current point on by two
2458        characters after failing to match at the start. However, the pattern \nA
2459        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2460        the same is true. There doesn't seem any very clean way out of this, but
2461        what I have chosen to do makes the common cases work: PCRE now takes note
2462        of whether there can be an explicit match for \r or \n anywhere in the
2463        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2464        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2465        pattern has explicit CR or LF references.
2467    29. Added (*CR) etc for changing newline setting at start of pattern.
2470    Version 7.2 19-Jun-07
2471    ---------------------
2473     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
2474        which is apparently normally available under Windows.
2476     2. Re-jig the pcregrep tests with different newline settings in an attempt
2477        to make them independent of the local environment's newline setting.
2479     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
2481     4. Some of the "internals" tests were previously cut out when the link size
2482        was not 2, because the output contained actual offsets. The recent new
2483        "Z" feature of pcretest means that these can be cut out, making the tests
2484        usable with all link sizes.
2486     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
2487        stack recursion. This gives a massive performance boost under BSD, but just
2488        a small improvement under Linux. However, it saves one field in the frame
2489        in all cases.
2491     6. Added more features from the forthcoming Perl 5.10:
2493        (a) (?-n) (where n is a string of digits) is a relative subroutine or
2494            recursion call. It refers to the nth most recently opened parentheses.
2496        (b) (?+n) is also a relative subroutine call; it refers to the nth next
2497            to be opened parentheses.
2499        (c) Conditions that refer to capturing parentheses can be specified
2500            relatively, for example, (?(-2)... or (?(+3)...
2502        (d) \K resets the start of the current match so that everything before
2503            is not part of it.
2505        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2507        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2508            reference syntax.
2510        (g) (?| introduces a group in which the numbering of parentheses in each
2511            alternative starts with the same number.
2513        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2515     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2518     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2519        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2520        for detecting groups that can match an empty string.
2522     9. A pattern with a very large number of alternatives (more than several
2523        hundred) was running out of internal workspace during the pre-compile
2524        phase, where pcre_compile() figures out how much memory will be needed. A
2525        bit of new cunning has reduced the workspace needed for groups with
2526        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2527        workspace instead of running out of the 4096 that are available.
2529    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2531    11. Applied patch from Google to remove an optimization that didn't quite work.
2532        The report of the bug said:
2534          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2535          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2536          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2538    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2539        it matched the wrong number of bytes.
2542    Version 7.1 24-Apr-07
2543    ---------------------
2545     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
2546        that is more "standard", making use of automake and other Autotools. There
2547        is some re-arrangement of the files and adjustment of comments consequent
2548        on this.
2550     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
2551        for recursive directory scanning broke on some systems because the files
2552        are not scanned in any specific order and on different systems the order
2553        was different. A call to "sort" has been inserted into RunGrepTest for the
2554        approprate test as a short-term fix. In the longer term there may be an
2555        alternative.
2557     3. I had an email from Eric Raymond about problems translating some of PCRE's
2558        man pages to HTML (despite the fact that I distribute HTML pages, some
2559        people do their own conversions for various reasons). The problems
2560        concerned the use of low-level troff macros .br and .in. I have therefore
2561        removed all such uses from the man pages (some were redundant, some could
2562        be replaced by .nf/.fi pairs). The 132html script that I use to generate
2563        HTML has been updated to handle .nf/.fi and to complain if it encounters
2564        .br or .in.
2566     4. Updated comments in configure.ac that get placed in config.h.in and also
2567        arranged for config.h to be included in the distribution, with the name
2568        config.h.generic, for the benefit of those who have to compile without
2569        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
2571     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
2572        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
2573        makevp.bat for the current PCRE, using the additional files
2574        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
2576     6. A Windows user reported a minor discrepancy with test 2, which turned out
2577        to be caused by a trailing space on an input line that had got lost in his
2578        copy. The trailing space was an accident, so I've just removed it.
2580     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
2581        that is needed.
2583     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
2584        as "const" (a) because they are and (b) because it helps the PHP
2585        maintainers who have recently made a script to detect big data structures
2586        in the php code that should be moved to the .rodata section. I remembered
2587        to update Builducptable as well, so it won't revert if ucptable.h is ever
2588        re-created.
2590     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
2591        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
2592        order to be able to cut out the UTF-8 tables in the latter when UTF-8
2593        support is not required. This saves 1.5-2K of code, which is important in
2594        some applications.
2596        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
2597        so as not to refer to the tables, even though these functions will never be
2598        called when UTF-8 support is disabled. Otherwise there are problems with a
2599        shared library.
2601    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
2603        (a) It was defining its arguments as char * instead of void *.
2605        (b) It was assuming that all moves were upwards in memory; this was true
2606            a long time ago when I wrote it, but is no longer the case.
2608        The emulated memove() is provided for those environments that have neither
2609        memmove() nor bcopy(). I didn't think anyone used it these days, but that
2610        is clearly not the case, as these two bugs were recently reported.
2612    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
2613        and Detrail to create the HTML documentation, the .txt form of the man
2614        pages, and it removes trailing spaces from listed files. It also creates
2615        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
2616        case, it wraps all the #defines with #ifndefs. This script should be run
2617        before "make dist".
2619    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
2620        with Unicode property support.
2622        (a) For a maximizing quantifier, if the two different cases of the
2623            character were of different lengths in their UTF-8 codings (there are
2624            some cases like this - I found 11), and the matching function had to
2625            back up over a mixture of the two cases, it incorrectly assumed they
2626            were both the same length.
2628        (b) When PCRE was configured to use the heap rather than the stack for
2629            recursion during matching, it was not correctly preserving the data for
2630            the other case of a UTF-8 character when checking ahead for a match
2631            while processing a minimizing repeat. If the check also involved
2632            matching a wide character, but failed, corruption could cause an
2633            erroneous result when trying to check for a repeat of the original
2634            character.
2636    13. Some tidying changes to the testing mechanism:
2638        (a) The RunTest script now detects the internal link size and whether there
2639            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
2640            values substituted by "configure". (The RunGrepTest script already did
2641            this for UTF-8.) The configure.ac script no longer substitutes the
2642            relevant variables.
2644        (b) The debugging options /B and /D in pcretest show the compiled bytecode
2645            with length and offset values. This means that the output is different
2646            for different internal link sizes. Test 2 is skipped for link sizes
2647            other than 2 because of this, bypassing the problem. Unfortunately,
2648            there was also a test in test 3 (the locale tests) that used /B and
2649            failed for link sizes other than 2. Rather than cut the whole test out,
2650            I have added a new /Z option to pcretest that replaces the length and
2651            offset values with spaces. This is now used to make test 3 independent
2652            of link size. (Test 2 will be tidied up later.)
2654    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
2655        segmentation fault instead of returning the appropriate error message.
2657    15. In multiline mode when the newline sequence was set to "any", the pattern
2658        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
2659        This doesn't seem right; it now treats the CRLF combination as the line
2660        ending, and so does not match in that case. It's only a pattern such as ^$
2661        that would hit this one: something like ^ABC$ would have failed after \r
2662        and then tried again after \r\n.
2664    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
2665        in an attempt to make files that differ only in their line terminators
2666        compare equal. This works on Linux.
2668    17. Under certain error circumstances pcregrep might try to free random memory
2669        as it exited. This is now fixed, thanks to valgrind.
2671    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
2672        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
2673        was because its rules for how to advance for /g after matching an empty
2674        string at the end of a line did not allow for this case. They now check for
2675        it specially.
2677    20. pcretest is supposed to handle patterns and data of any length, by
2678        extending its buffers when necessary. It was getting this wrong when the
2679        buffer for a data line had to be extended.
2681    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
2682        CRLF as a newline sequence.
2684    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
2685        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
2686        I have nevertheless tidied it up.
2688    23. Added some casts to kill warnings from HP-UX ia64 compiler.
2690    24. Added a man page for pcre-config.
2693    Version 7.0 19-Dec-06
2694    ---------------------
2696     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
2697        moving to gcc 4.1.1.
2699     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
2700        sys/time.h, which is documented as needed for this function. It doesn't
2701        seem to matter on Linux, but it showed up on some releases of OS X.
2703     3. It seems that there are systems where bytes whose values are greater than
2704        127 match isprint() in the "C" locale. The "C" locale should be the
2705        default when a C program starts up. In most systems, only ASCII printing
2706        characters match isprint(). This difference caused the output from pcretest
2707        to vary, making some of the tests fail. I have changed pcretest so that:
2709        (a) When it is outputting text in the compiled version of a pattern, bytes
2710            other than 32-126 are always shown as hex escapes.
2712        (b) When it is outputting text that is a matched part of a subject string,
2713            it does the same, unless a different locale has been set for the match
2714            (using the /L modifier). In this case, it uses isprint() to decide.
2716     4. Fixed a major bug that caused incorrect computation of the amount of memory
2717        required for a compiled pattern when options that changed within the
2718        pattern affected the logic of the preliminary scan that determines the
2719        length. The relevant options are -x, and -i in UTF-8 mode. The result was
2720        that the computed length was too small. The symptoms of this bug were
2721        either the PCRE error "internal error: code overflow" from pcre_compile(),
2722        or a glibc crash with a message such as "pcretest: free(): invalid next
2723        size (fast)". Examples of patterns that provoked this bug (shown in
2724        pcretest format) are:
2726          /(?-x: )/x
2727          /(?x)(?-x: \s*#\s*)/
2728          /((?i)[\x{c0}])/8
2729          /(?i:[\x{c0}])/8
2731        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
2732        is now done differently.
2734     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
2735        wrapper classes; (b) implement a new function in the C++ scanner that is
2736        more efficient than the old way of doing things because it avoids levels of
2737        recursion in the regex matching; (c) add a paragraph to the documentation
2738        for the FullMatch() function.
2740     6. The escape sequence \n was being treated as whatever was defined as
2741        "newline". Not only was this contrary to the documentation, which states
2742        that \n is character 10 (hex 0A), but it also went horribly wrong when
2743        "newline" was defined as CRLF. This has been fixed.
2745     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
2746        was being set to -1 for the "end of line" case (supposedly a value that no
2747        character can have). Though this value is never used (the check for end of
2748        line is "zero bytes in current character"), it caused compiler complaints.
2749        I've changed it to 0xffffffff.
2751     8. In pcre_version.c, the version string was being built by a sequence of
2752        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
2753        string (as it is for production releases) called a macro with an empty
2754        argument. The C standard says the result of this is undefined. The gcc
2755        compiler treats it as an empty string (which was what was wanted) but it is
2756        reported that Visual C gives an error. The source has been hacked around to
2757        avoid this problem.
2759     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
2760        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
2761        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
2762        of them did).
2764    10. Originally, pcretest opened its input and output without "b"; then I was
2765        told that "b" was needed in some environments, so it was added for release
2766        5.0 to both the input and output. (It makes no difference on Unix-like
2767        systems.) Later I was told that it is wrong for the input on Windows. I've
2768        now abstracted the modes into two macros, to make it easier to fiddle with
2769        them, and removed "b" from the input mode under Windows.
2771    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
2773    12. Added -help and --help to pcretest as an official way of being reminded
2774        of the options.
2776    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
2777        and pcrecpp.cc because they annoy compilers at high warning levels.
2779    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
2781    15. Fixed an occurrence of == in configure.ac that should have been = (shell
2782        scripts are not C programs :-) and which was not noticed because it works
2783        on Linux.
2785    16. pcretest is supposed to handle any length of pattern and data line (as one
2786        line or as a continued sequence of lines) by extending its input buffer if
2787        necessary. This feature was broken for very long pattern lines, leading to
2788        a string of junk being passed to pcre_compile() if the pattern was longer
2789        than about 50K.
2791    17. I have done a major re-factoring of the way pcre_compile() computes the
2792        amount of memory needed for a compiled pattern. Previously, there was code
2793        that made a preliminary scan of the pattern in order to do this. That was
2794        OK when PCRE was new, but as the facilities have expanded, it has become
2795        harder and harder to keep it in step with the real compile phase, and there
2796        have been a number of bugs (see for example, 4 above). I have now found a
2797        cunning way of running the real compile function in a "fake" mode that
2798        enables it to compute how much memory it would need, while actually only
2799        ever using a few hundred bytes of working memory and without too many
2800        tests of the mode. This should make future maintenance and development
2801        easier. A side effect of this work is that the limit of 200 on the nesting
2802        depth of parentheses has been removed (though this was never a serious
2803        limitation, I suspect). However, there is a downside: pcre_compile() now
2804        runs more slowly than before (30% or more, depending on the pattern). I
2805        hope this isn't a big issue. There is no effect on runtime performance.
2807    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
2808        newline (only possible for the last line of a file) and it was a
2809        pattern that set a locale (followed by /Lsomething), pcretest crashed.
2811    19. Added additional timing features to pcretest. (1) The -tm option now times
2812        matching only, not compiling. (2) Both -t and -tm can be followed, as a
2813        separate command line item, by a number that specifies the number of
2814        repeats to use when timing. The default is 50000; this gives better
2815        precision, but takes uncomfortably long for very large patterns.
2817    20. Extended pcre_study() to be more clever in cases where a branch of a
2818        subpattern has no definite first character. For example, (a*|b*)[cd] would
2819        previously give no result from pcre_study(). Now it recognizes that the
2820        first character must be a, b, c, or d.
2822    21. There was an incorrect error "recursive call could loop indefinitely" if
2823        a subpattern (or the entire pattern) that was being tested for matching an
2824        empty string contained only one non-empty item after a nested subpattern.
2825        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
2826        incorrectly, because the \d was being skipped in the check.
2828    22. The pcretest program now has a new pattern option /B and a command line
2829        option -b, which is equivalent to adding /B to every pattern. This causes
2830        it to show the compiled bytecode, without the additional information that
2831        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
2832        is the same as /B/I).
2834    23. A new optimization is now able automatically to treat some sequences such
2835        as a*b as a*+b. More specifically, if something simple (such as a character
2836        or a simple class like \d) has an unlimited quantifier, and is followed by
2837        something that cannot possibly match the quantified thing, the quantifier
2838        is automatically "possessified".
2840    24. A recursive reference to a subpattern whose number was greater than 39
2841        went wrong under certain circumstances in UTF-8 mode. This bug could also
2842        have affected the operation of pcre_study().
2844    25. Realized that a little bit of performance could be had by replacing
2845        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
2847    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
2849    27. Possessive quantifiers such as a++ were previously implemented by turning
2850        them into atomic groups such as ($>a+). Now they have their own opcodes,
2851        which improves performance. This includes the automatically created ones
2852        from 23 above.
2854    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
2855        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
2856        the first matched character to be a colon. This applied both to named and
2857        numbered groups.
2859    29. The ucpinternal.h header file was missing its idempotency #ifdef.
2861    30. I was sent a "project" file called libpcre.a.dev which I understand makes
2862        building PCRE on Windows easier, so I have included it in the distribution.
2864    31. There is now a check in pcretest against a ridiculously large number being
2865        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
2866        loop, the loop is abandoned.
2868    32. Forward references to subpatterns in conditions such as (?(2)...) where
2869        subpattern 2 is defined later cause pcre_compile() to search forwards in
2870        the pattern for the relevant set of parentheses. This search went wrong
2871        when there were unescaped parentheses in a character class, parentheses
2872        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2874    33. "Subroutine" calls and backreferences were previously restricted to
2875        referencing subpatterns earlier in the regex. This restriction has now
2876        been removed.
2878    34. Added a number of extra features that are going to be in Perl 5.10. On the
2879        whole, these are just syntactic alternatives for features that PCRE had
2880        previously implemented using the Python syntax or my own invention. The
2881        other formats are all retained for compatibility.
2883        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2884            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2885            also .NET compatible.
2887        (b) A recursion or subroutine call to a named group can now be defined as
2888            (?&name) as well as (?P>name).
2890        (c) A backreference to a named group can now be defined as \k<name> or
2891            \k'name' as well as (?P=name). The new forms, as well as being in Perl
2892            5.10, are also .NET compatible.
2894        (d) A conditional reference to a named group can now use the syntax
2895            (?(<name>) or (?('name') as well as (?(name).
2897        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2898            groups (named and numbered) that are never evaluated inline, but can be
2899            called as "subroutines" from elsewhere. In effect, the DEFINE condition
2900            is always false. There may be only one alternative in such a group.
2902        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2903            as the simple (?(R). The condition is true only if the most recent
2904            recursion is that of the given number or name. It does not search out
2905            through the entire recursion stack.
2907        (g) The escape \gN or \g{N} has been added, where N is a positive or
2908            negative number, specifying an absolute or relative reference.
2910    35. Tidied to get rid of some further signed/unsigned compiler warnings and
2911        some "unreachable code" warnings.
2913    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2914        things, this adds five new scripts.
2916    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2917        There were also incompatibilities regarding the handling of \Q..\E inside
2918        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2919        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2921    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2922        matches an empty string, and forcibly breaks the loop. There were bugs in
2923        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2924        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2925        separate and independent bugs (that affected different cases) have been
2926        fixed.
2928    39. Refactored the code to abolish the use of different opcodes for small
2929        capturing bracket numbers. This is a tidy that I avoided doing when I
2930        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2931        The new approach is not only tidier, it makes it possible to reduce the
2932        memory needed to fix the previous bug (38).
2934    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2935        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2936        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2937        mode.
2939    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2940        report.
2942    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2943        copy construction and assignment in the C++ wrapper.
2945    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2946        couple of bugs that could have given wrong results in the "--newline=crlf"
2947        case.
2949    44. Added a number of casts and did some reorganization of signed/unsigned int
2950        variables following suggestions from Dair Grant. Also renamed the variable
2951        "this" as "item" because it is a C++ keyword.
2953    45. Arranged for dftables to add
2955          #include "pcre_internal.h"
2957        to pcre_chartables.c because without it, gcc 4.x may remove the array
2958        definition from the final binary if PCRE is built into a static library and
2959        dead code stripping is activated.
2961    46. For an unanchored pattern, if a match attempt fails at the start of a
2962        newline sequence, and the newline setting is CRLF or ANY, and the next two
2963        characters are CRLF, advance by two characters instead of one.
2966  Version 6.7 04-Jul-06  Version 6.7 04-Jul-06
2967  ---------------------  ---------------------

Removed from v.91  
changed lines
  Added in v.1364

  ViewVC Help
Powered by ViewVC 1.1.5