ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 326 by ph10, Sat Mar 8 17:24:02 2008 UTC revision 1363 by ph10, Tue Oct 1 16:54:40 2013 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4  Version 7.7 05-Mar-08  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.
92    Version 8.33 28-May-2013
93    ------------------------
95    1.  Added 'U' to some constants that are compared to unsigned integers, to
96        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
97        variables that are added to signed variables, to ensure the result is
98        signed and can be negated.
100    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
101        CMake config files.
103    3.  Revise the creation of config.h.generic so that all boolean macros are
104        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
105        overriding via -D on the command line possible.
107    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
108        to unsigned int is reported to make a quite noticeable speed difference in
109        a specific Windows environment. Testing on Linux did also appear to show
110        some benefit (and it is clearly not harmful). Also fixed the definition of
111        Xop which should be unsigned.
113    5.  Related to (4), changing the definition of the intermediate variable cc
114        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
115        improvements.
117    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
118        unnecessary spaces.
120    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
121        and later.
123    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
125    9.  Optimizing fast_forward_start_bits in JIT.
127    10. Adding support for callouts in JIT, and fixing some issues revealed
128        during this work. Namely:
130        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
132        (b) Minimum length was not checked before the matching is started.
134    11. The value of capture_last that is passed to callouts was incorrect in some
135        cases when there was a capture on one path that was subsequently abandoned
136        after a backtrack. Also, the capture_last value is now reset after a
137        recursion, since all captures are also reset in this case.
139    12. The interpreter no longer returns the "too many substrings" error in the
140        case when an overflowing capture is in a branch that is subsequently
141        abandoned after a backtrack.
143    13. In the pathological case when an offset vector of size 2 is used, pcretest
144        now prints out the matched string after a yield of 0 or 1.
146    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
147        Only supported by the JIT compiler at the moment.
149    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
151    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
152        offset of the starting point of the matching process, provided the offsets
153        vector is large enough.
155    17. The \A escape now records a lookbehind value of 1, though its execution
156        does not actually inspect the previous character. This is to ensure that,
157        in partial multi-segment matching, at least one character from the old
158        segment is retained when a new segment is processed. Otherwise, if there
159        are no lookbehinds in the pattern, \A might match incorrectly at the start
160        of a new segment.
162    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
164    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
165        gives some modest performance improvement in 8-bit mode.
167    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
168        be expressed in certain programming languages using Universal Character
169        Names.
171    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
172        which points out that "non characters" are not "characters that may not
173        appear in Unicode strings" but rather "characters that are reserved for
174        internal use and have only local meaning".
176    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
177        there was a conditional group that depended on an assertion, if the
178        assertion was false, the callout that immediately followed the alternation
179        in the condition was skipped when pcre_exec() was used for matching.
181    23. Allow an explicit callout to be inserted before an assertion that is the
182        condition for a conditional group, for compatibility with automatic
183        callouts, which always insert a callout at this point.
185    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
186        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
188    25. (*PRUNE) is now supported by the JIT compiler.
190    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
192    27. Fix the case where there are two or more SKIPs with arguments that may be
193        ignored.
195    28. (*SKIP) is now supported by the JIT compiler.
197    29. (*THEN) is now supported by the JIT compiler.
199    30. Update RunTest with additional test selector options.
201    31. The way PCRE handles backtracking verbs has been changed in two ways.
203        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
204        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
205        backtracking. In some cases this makes it more Perl-compatible, but Perl's
206        rather obscure rules do not always do the same thing.
208        (2) Previously, backtracking verbs were confined within assertions. This is
209        no longer the case for positive assertions, except for (*ACCEPT). Again,
210        this sometimes improves Perl compatibility, and sometimes does not.
212    32. A number of tests that were in test 2 because Perl did things differently
213        have been moved to test 1, because either Perl or PCRE has changed, and
214        these tests are now compatible.
216    32. Backtracking control verbs are now handled in the same way in JIT and
217        interpreter.
219    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
220        contained a forward subroutine reference caused a compile error.
222    34. Auto-detect and optimize limited repetitions in JIT.
224    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
225        blocking (*UTF) etc.
227    36. In the interpreter, maximizing pattern repetitions for characters and
228        character types now use tail recursion, which reduces stack usage.
230    37. The value of the max lookbehind was not correctly preserved if a compiled
231        and saved regex was reloaded on a host of different endianness.
233    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
234        of the compiled pattern block, expand the flags field from 16 to 32 bits
235        because it was almost full.
237    39. Try madvise first before posix_madvise.
239    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
240        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
241        original change locked out all matches of empty strings. This has been
242        changed so that one match of an empty string per line is recognized.
243        Subsequent searches on the same line (for colouring or for --only-matching,
244        for example) do not recognize empty strings.
246    41. Applied a user patch to fix a number of spelling mistakes in comments.
248    42. Data lines longer than 65536 caused pcretest to crash.
250    43. Clarified the data type for length and startoffset arguments for pcre_exec
251        and pcre_dfa_exec in the function-specific man pages, where they were
252        explicitly stated to be in bytes, never having been updated. I also added
253        some clarification to the pcreapi man page.
255    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
256        a segmentation fault.
259    Version 8.32 30-November-2012
260    -----------------------------
262    1.  Improved JIT compiler optimizations for first character search and single
263        character iterators.
265    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
266        Patch by Daniel Richard G.
268    3.  Single character iterator optimizations in the JIT compiler.
270    4.  Improved JIT compiler optimizations for character ranges.
272    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
273        Reported by Giuseppe D'Angelo.
275    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
276        of a line, was being set incorrectly in cases where .* appeared inside
277        atomic brackets at the start of a pattern, or where there was a subsequent
278        *PRUNE or *SKIP.
280    7.  Improved instruction cache flush for POWER/PowerPC.
281        Patch by Daniel Richard G.
283    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
284        grep:
286        (a) There is now no limit to the number of patterns to be matched.
288        (b) An error is given if a pattern is too long.
290        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
291            are now supported.
293        (d) --exclude-from and --include-from (multiple use) have been added.
295        (e) Exclusions and inclusions now apply to all files and directories, not
296            just to those obtained from scanning a directory recursively.
298        (f) Multiple uses of -f and --file-list are now supported.
300        (g) In a Windows environment, the default for -d has been changed from
301            "read" (the GNU grep default) to "skip", because otherwise the presence
302            of a directory in the file list provokes an error.
304        (h) The documentation has been revised and clarified in places.
306    9.  Improve the matching speed of capturing brackets.
308    10. Changed the meaning of \X so that it now matches a Unicode extended
309        grapheme cluster.
311    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
312        out POSIX threads when JIT support is configured.
314    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
316    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
317        structure could go wrong in environments where size_t is not the same size
318        as int.
320    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
321        set.
323    15. The EBCDIC support had decayed; later updates to the code had included
324        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
325        general tidy up of EBCDIC-related issues, and the documentation was also
326        not quite right. There is now a test that can be run on ASCII systems to
327        check some of the EBCDIC-related things (but is it not a full test).
329    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
330        in a small tidy to the code.
332    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
334    18. If the --only-matching (-o) option in pcregrep is specified multiple
335        times, each one causes appropriate output. For example, -o1 -o2 outputs the
336        substrings matched by the 1st and 2nd capturing parentheses. A separating
337        string can be specified by --om-separator (default empty).
339    19. Improving the first n character searches.
341    20. Turn case lists for horizontal and vertical white space into macros so that
342        they are defined only once.
344    21. This set of changes together give more compatible Unicode case-folding
345        behaviour for characters that have more than one other case when UCP
346        support is available.
348        (a) The Unicode property table now has offsets into a new table of sets of
349            three or more characters that are case-equivalent. The MultiStage2.py
350            script that generates these tables (the pcre_ucd.c file) now scans
351            CaseFolding.txt instead of UnicodeData.txt for character case
352            information.
354        (b) The code for adding characters or ranges of characters to a character
355            class has been abstracted into a generalized function that also handles
356            case-independence. In UTF-mode with UCP support, this uses the new data
357            to handle characters with more than one other case.
359        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
360            whose other case is greater than 256 are now correctly matched
361            caselessly. Previously, the high codepoint matched the low one, but not
362            vice versa.
364        (d) The processing of \h, \H, \v, and \ in character classes now makes use
365            of the new class addition function, using character lists defined as
366            macros alongside the case definitions of 20 above.
368        (e) Caseless back references now work with characters that have more than
369            one other case.
371        (f) General caseless matching of characters with more than one other case
372            is supported.
374    22. Unicode character properties were updated from Unicode 6.2.0
376    23. Improved CMake support under Windows. Patch by Daniel Richard G.
378    24. Add support for 32-bit character strings, and UTF-32
380    25. Major JIT compiler update (code refactoring and bugfixing).
381        Experimental Sparc 32 support is added.
383    26. Applied a modified version of Daniel Richard G's patch to create
384        pcre.h.generic and config.h.generic by "make" instead of in the
385        PrepareRelease script.
387    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
388        pcre_compile.c when checking for a zero character.
390    28. Introducing a native interface for JIT. Through this interface, the compiled
391        machine code can be directly executed. The purpose of this interface is to
392        provide fast pattern matching, so several sanity checks are not performed.
393        However, feature tests are still performed. The new interface provides
394        1.4x speedup compared to the old one.
396    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
397        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
398        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
400    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
401        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
402        the "old" RFC 2279). Instead, it ended up passing a negative length to
403        pcre_exec().
405    31. Add support for GCC's visibility feature to hide internal functions.
407    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
408        "unknown -C option" after outputting 0 or 1.
410    33. There is now support for generating a code coverage report for the test
411        suite in environments where gcc is the compiler and lcov is installed. This
412        is mainly for the benefit of the developers.
414    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
415        unaddressable using valgrind annotations, allowing valgrind to detect
416        invalid memory accesses. This is mainly for the benefit of the developers.
418    25. (*UTF) can now be used to start a pattern in any of the three libraries.
420    26. Give configure error if --enable-cpp but no C++ compiler found.
423    Version 8.31 06-July-2012
424    -------------------------
426    1.  Fixing a wrong JIT test case and some compiler warnings.
428    2.  Removed a bashism from the RunTest script.
430    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
431        to unsigned type, result still unsigned" that was given by an MS compiler
432        on encountering the code "-sizeof(xxx)".
434    4.  Partial matching support is added to the JIT compiler.
436    5.  Fixed several bugs concerned with partial matching of items that consist
437        of more than one character:
439        (a) /^(..)\1/ did not partially match "aba" because checking references was
440            done on an "all or nothing" basis. This also applied to repeated
441            references.
443        (b) \R did not give a hard partial match if \r was found at the end of the
444            subject.
446        (c) \X did not give a hard partial match after matching one or more
447            characters at the end of the subject.
449        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
450            a partial match for the string "\r".
452        (e) When newline was set to CRLF, the metacharacter "." did not recognize
453            a partial match for a CR character at the end of the subject string.
455    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
456        running pcretest, the text "(JIT)" added to the output whenever JIT is
457        actually used to run the match.
459    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
460        or /S+[+] with a digit between 1 and 7.
462    8.  OP_NOT now supports any UTF character not just single-byte ones.
464    9.  (*MARK) control verb is now supported by the JIT compiler.
466    10. The command "./RunTest list" lists the available tests without actually
467        running any of them. (Because I keep forgetting what they all are.)
471    12. Applied a (slightly modified) user-supplied patch that improves performance
472        when the heap is used for recursion (compiled with --disable-stack-for-
473        recursion). Instead of malloc and free for each heap frame each time a
474        logical recursion happens, frames are retained on a chain and re-used where
475        possible. This sometimes gives as much as 30% improvement.
477    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
478        call.
480    14. As documented, (*COMMIT) is now confined to within a positive assertion.
482    15. It is now possible to link pcretest with libedit as an alternative to
483        libreadline.
485    16. (*COMMIT) control verb is now supported by the JIT compiler.
487    17. The Unicode data tables have been updated to Unicode 6.1.0.
489    18. Added --file-list option to pcregrep.
491    19. Added binary file support to pcregrep, including the -a, --binary-files,
492        -I, and --text options.
494    20. The madvise function is renamed for posix_madvise for QNX compatibility
495        reasons. Fixed by Giuseppe D'Angelo.
497    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
499    22. Changed the option for creating symbolic links for 16-bit man pages from
500        -s to -sf so that re-installing does not cause issues.
502    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
504    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
505        matches in certain environments (the workspace was not being correctly
506        retained). Also added to pcre_dfa_exec() a simple plausibility check on
507        some of the workspace data at the beginning of a restart.
509    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
510        was not doing so when it should - probably a typo introduced by SVN 528
511        (change 8.10/14).
513    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
514        \w+ when the character tables indicated that \x{c4} was a word character.
515        There were several related cases, all because the tests for doing a table
516        lookup were testing for characters less than 127 instead of 255.
518    27. If a pattern contains capturing parentheses that are not used in a match,
519        their slots in the ovector are set to -1. For those that are higher than
520        any matched groups, this happens at the end of processing. In the case when
521        there were back references that the ovector was too small to contain
522        (causing temporary malloc'd memory to be used during matching), and the
523        highest capturing number was not used, memory off the end of the ovector
524        was incorrectly being set to -1. (It was using the size of the temporary
525        memory instead of the true size.)
527    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
528        ovector size, it uses memory at the end of the block that it has got.
530    29. Check for an overlong MARK name and give an error at compile time. The
531        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
533    30. JIT compiler update.
535    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
536        Rill for the patch.
538    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
540    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
542    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
543        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
544        was enabled.
546    35. Improve JIT code generation for greedy plus quantifier.
548    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
549        1 to "aa" instead of to an empty string. The bug affected repeated groups
550        that could potentially match an empty string.
552    37. Optimizing single character iterators in JIT.
554    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
555        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
556        codepoints that are too big for the mode are faulted, and in a UTF mode,
557        disallowed codepoints are also faulted.
559    39. If PCRE was compiled with UTF support, in three places in the DFA
560        matcher there was code that should only have been obeyed in UTF mode, but
561        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
562        processing when bytes with values greater than 127 were present. In 16-bit
563        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
564        both cases the values are those that cannot be the first data item in a UTF
565        character. The three items that might have provoked this were recursions,
566        possessively repeated groups, and atomic groups.
568    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
569        and pcregrep, because some OS require shared objects to be explicitly
570        passed to ld, causing the link step to fail if they are not.
572    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
573        mode, patterns that started with \h* or \R* might be incorrectly matched.
576    Version 8.30 04-February-2012
577    -----------------------------
579    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
580        name is defined in ctype.h.
582    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
583        only in quite long subpatterns.
585    3.  Removed the function pcre_info(), which has been obsolete and deprecated
586        since it was replaced by pcre_fullinfo() in February 2000.
588    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
589        match a (*MARK), and the match failed at the start of the subject, a
590        reference to memory before the start of the subject could occur. This bug
591        was introduced by fix 17 of release 8.21.
593    5.  A reference to an unset group with zero minimum repetition was giving
594        totally wrong answers (in non-JavaScript-compatibility mode). For example,
595        /(another)?(\1?)test/ matched against "hello world test". This bug was
596        introduced in release 8.13.
598    6.  Add support for 16-bit character strings (a large amount of work involving
599        many changes and refactorings).
601    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
602        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
603        from a file.
605    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
606        rounding is not applied in this particular case).
608    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
609        if they appear, or are escaped, in patterns.
611    10. Get rid of a number of -Wunused-but-set-variable warnings.
613    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
614        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
615        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
616        also returns the mark "x". This bug applied to capturing parentheses,
617        non-capturing parentheses, and atomic parentheses. It also applied to some
618        assertions.
620    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
621        information out of configure.ac instead of relying on pcre.h.generic, which
622        is not stored in the repository.
624    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
625        -lreadline.
627    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
629    15. Applied Graycode's patch to put the top-level frame on the stack rather
630        than the heap when not using the stack for recursion. This gives a
631        performance improvement in many cases when recursion is not deep.
633    16. Experimental code added to "pcretest -C" to output the stack frame size.
636    Version 8.21 12-Dec-2011
637    ------------------------
639    1.  Updating the JIT compiler.
641    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
642        are added as well.
644    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
645        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
646        calling _pcre_jit_exec. Some extra comments are added.
648    4.  (*MARK) settings inside atomic groups that do not contain any capturing
649        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
650        was introduced by change 18 for 8.20.
652    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
653        ECMA-262 standard.
655    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
656        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
657        This bug was probably introduced by change 9 of 8.13.
659    7.  While fixing 6 above, I noticed that a number of other items were being
660        incorrectly rejected as "not fixed length". This arose partly because newer
661        opcodes had not been added to the fixed-length checking code. I have (a)
662        corrected the bug and added tests for these items, and (b) arranged for an
663        error to occur if an unknown opcode is encountered while checking for fixed
664        length instead of just assuming "not fixed length". The items that were
665        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
666        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
667        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
669    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
670        being incorrectly compiled and would have given unpredicatble results.
672    9.  A possessively repeated subpattern with minimum repeat count greater than
673        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
674        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
675        the first (A) could occur when it should not.
677    10. Add a cast and remove a redundant test from the code.
679    11. JIT should use pcre_malloc/pcre_free for allocation.
681    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
682        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
683        is anything other than /usr, -L is still shown).
685    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
687    14. Perl does not support \N without a following name in a [] class; PCRE now
688        also gives an error.
690    15. If a forward reference was repeated with an upper limit of around 2000,
691        it caused the error "internal error: overran compiling workspace". The
692        maximum number of forward references (including repeats) was limited by the
693        internal workspace, and dependent on the LINK_SIZE. The code has been
694        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
695        the default depends on LINK_SIZE. There is a new upper limit (for safety)
696        of around 200,000 forward references. While doing this, I also speeded up
697        the filling in of repeated forward references.
699    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
700        incorrectly expecting the subject to contain another "a" after the start.
702    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
703        in the match, the SKIP should be ignored. This was not happening; instead
704        the SKIP was being treated as NOMATCH. For patterns such as
705        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
706        tested.
708    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
709        now much more compatible with Perl, in particular in cases where the result
710        is a non-match for a non-anchored pattern. For example, if
711        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
712        "m", where previously it did not return a name. A side effect of this
713        change is that for partial matches, the last encountered mark name is
714        returned, as for non matches. A number of tests that were previously not
715        Perl-compatible have been moved into the Perl-compatible test files. The
716        refactoring has had the pleasing side effect of removing one argument from
717        the match() function, thus reducing its stack requirements.
719    19. If the /S+ option was used in pcretest to study a pattern using JIT,
720        subsequent uses of /S (without +) incorrectly behaved like /S+.
722    21. Retrieve executable code size support for the JIT compiler and fixing
723        some warnings.
725    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
726        not work when the shorter character appeared right at the end of the
727        subject string.
729    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
730        systems.
732    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
733        output it when the /M option is used in pcretest.
735    25. The CheckMan script was not being included in the distribution. Also, added
736        an explicit "perl" to run Perl scripts from the PrepareRelease script
737        because this is reportedly needed in Windows.
739    26. If study data was being save in a file and studying had not found a set of
740        "starts with" bytes for the pattern, the data written to the file (though
741        never used) was taken from uninitialized memory and so caused valgrind to
742        complain.
744    27. Updated RunTest.bat as provided by Sheri Pierce.
746    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
748    29. Computation of memory usage for the table of capturing group names was
749        giving an unnecessarily large value.
752    Version 8.20 21-Oct-2011
753    ------------------------
755    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
756        a POSIX class. After further experiments with Perl, which convinced me that
757        Perl has bugs and confusions, a closing square bracket is no longer allowed
758        in a POSIX name. This bug also affected patterns with classes that started
759        with full stops.
761    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
762        captured substring, but while checking the failing first alternative,
763        substring 1 is temporarily captured. If the output vector supplied to
764        pcre_exec() was not big enough for this capture, the yield of the function
765        was still zero ("insufficient space for captured substrings"). This cannot
766        be totally fixed without adding another stack variable, which seems a lot
767        of expense for a edge case. However, I have improved the situation in cases
768        such as /(a)(b)x|abc/ matched against "abc", where the return code
769        indicates that fewer than the maximum number of slots in the ovector have
770        been set.
772    3.  Related to (2) above: when there are more back references in a pattern than
773        slots in the output vector, pcre_exec() uses temporary memory during
774        matching, and copies in the captures as far as possible afterwards. It was
775        using the entire output vector, but this conflicts with the specification
776        that only 2/3 is used for passing back captured substrings. Now it uses
777        only the first 2/3, for compatibility. This is, of course, another edge
778        case.
780    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
781        main code base, and can be used by building with --enable-jit. When this is
782        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
783        runtime --no-jit option is given.
785    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
786        ovector, the return from the function was zero, implying that there were
787        other matches that did not fit. The correct "exactly full" value is now
788        returned.
790    6.  If a subpattern that was called recursively or as a subroutine contained
791        (*PRUNE) or any other control that caused it to give a non-standard return,
792        invalid errors such as "Error -26 (nested recursion at the same subject
793        position)" or even infinite loops could occur.
795    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
796        computing the minimum length on reaching *ACCEPT, and so ended up with the
797        wrong value of 1 rather than 0. Further investigation indicates that
798        computing a minimum subject length in the presence of *ACCEPT is difficult
799        (think back references, subroutine calls), and so I have changed the code
800        so that no minimum is registered for a pattern that contains *ACCEPT.
802    8.  If (*THEN) was present in the first (true) branch of a conditional group,
803        it was not handled as intended. [But see 16 below.]
805    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
806        Sheri Pierce.
808    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
809        the first byte in a match must be "a".
811    11. Change 17 for 8.13 increased the recursion depth for patterns like
812        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
813        pattern contains any instances of (*THEN). If it does not, the old
814        optimizations are restored. It would be nice to do this on a per-group
815        basis, but at the moment that is not feasible.
817    12. In some environments, the output of pcretest -C is CRLF terminated. This
818        broke RunTest's code that checks for the link size. A single white space
819        character after the value is now allowed for.
821    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
822        For "fr", it uses the Windows-specific input and output files.
824    14. If (*THEN) appeared in a group that was called recursively or as a
825        subroutine, it did not work as intended. [But see next item.]
827    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
828        pattern fragments (but not containing any | characters). If A and B are
829        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
830        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
831        D. In other words, Perl considers parentheses that do not contain any |
832        characters to be part of a surrounding alternative, whereas PCRE was
833        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
834        differently. PCRE now behaves in the same way as Perl, except in the case
835        of subroutine/recursion calls such as (?1) which have in any case always
836        been different (but PCRE had them first :-).
838    16. Related to 15 above: Perl does not treat the | in a conditional group as
839        creating alternatives. Such a group is treated in the same way as an
840        ordinary group without any | characters when processing (*THEN). PCRE has
841        been changed to match Perl's behaviour.
843    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
844        RunGrepTest script failed.
846    18. Change 22 for version 13 caused atomic groups to use more stack. This is
847        inevitable for groups that contain captures, but it can lead to a lot of
848        stack use in large patterns. The old behaviour has been restored for atomic
849        groups that do not contain any capturing parentheses.
851    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
852        suppress the check for a minimum subject length at run time. (If it was
853        given to pcre_exec() or pcre_dfa_exec() it did work.)
855    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
856        fail to work when decoding hex characters in data strings in EBCDIC
857        environments.
859    21. It appears that in at least one Mac OS environment, the isxdigit() function
860        is implemented as a macro that evaluates to its argument more than once,
861        contravening the C 90 Standard (I haven't checked a later standard). There
862        was an instance in pcretest which caused it to go wrong when processing
863        \x{...} escapes in subject strings. The has been rewritten to avoid using
864        things like p++ in the argument of isxdigit().
867    Version 8.13 16-Aug-2011
868    ------------------------
870    1.  The Unicode data tables have been updated to Unicode 6.0.0.
872    2.  Two minor typos in pcre_internal.h have been fixed.
874    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
875        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
876        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
878    4.  There were a number of related bugs in the code for matching backrefences
879        caselessly in UTF-8 mode when codes for the characters concerned were
880        different numbers of bytes. For example, U+023A and U+2C65 are an upper
881        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
882        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
883        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
884        2-byte code at the end of the subject (it thought there wasn't enough data
885        left).
887    5.  Comprehensive information about what went wrong is now returned by
888        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
889        as the output vector has at least 2 elements. The offset of the start of
890        the failing character and a reason code are placed in the vector.
892    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
893        now returned is for the first byte of the failing character, instead of the
894        last byte inspected. This is an incompatible change, but I hope it is small
895        enough not to be a problem. It makes the returned offset consistent with
896        pcre_exec() and pcre_dfa_exec().
898    7.  pcretest now gives a text phrase as well as the error number when
899        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
900        failure, the offset and reason code are output.
902    8.  When \R was used with a maximizing quantifier it failed to skip backwards
903        over a \r\n pair if the subsequent match failed. Instead, it just skipped
904        back over a single character (\n). This seems wrong (because it treated the
905        two characters as a single entity when going forwards), conflicts with the
906        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
907        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
908        has been changed.
910    9.  Some internal refactoring has changed the processing so that the handling
911        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
912        time (the PCRE_DOTALL option was changed this way some time ago: version
913        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
914        which was always a bit of a fudge. It also means that there is one less
915        argument for the match() function, which reduces its stack requirements
916        slightly. This change also fixes an incompatibility with Perl: the pattern
917        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
919    10. More internal refactoring has drastically reduced the number of recursive
920        calls to match() for possessively repeated groups such as (abc)++ when
921        using pcre_exec().
923    11. While implementing 10, a number of bugs in the handling of groups were
924        discovered and fixed:
926        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
927        (a|)*(?1) gave a compile-time internal error.
928        ((a|)+)+  did not notice that the outer group could match an empty string.
929        (^a|^)+   was not marked as anchored.
930        (.*a|.*)+ was not marked as matching at start or after a newline.
932    12. Yet more internal refactoring has removed another argument from the match()
933        function. Special calls to this function are now indicated by setting a
934        value in a variable in the "match data" data block.
936    13. Be more explicit in pcre_study() instead of relying on "default" for
937        opcodes that mean there is no starting character; this means that when new
938        ones are added and accidentally left out of pcre_study(), testing should
939        pick them up.
941    14. The -s option of pcretest has been documented for ages as being an old
942        synonym of -m (show memory usage). I have changed it to mean "force study
943        for every regex", that is, assume /S for every regex. This is similar to -i
944        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
945        using it. It makes it easier to run collections of tests with and without
946        study enabled, and thereby test pcre_study() more easily. All the standard
947        tests are now run with and without -s (but some patterns can be marked as
948        "never study" - see 20 below).
950    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
951        restoration of the capturing data to the outer values was not happening
952        correctly.
954    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
955        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
956        pattern had matched an empty string, and so incorrectly returned a no
957        match.
959    17. There was optimizing code for the last branch of non-capturing parentheses,
960        and also for the obeyed branch of a conditional subexpression, which used
961        tail recursion to cut down on stack usage. Unfortunately, now that there is
962        the possibility of (*THEN) occurring in these branches, tail recursion is
963        no longer possible because the return has to be checked for (*THEN). These
964        two optimizations have therefore been removed. [But see 8.20/11 above.]
966    18. If a pattern containing \R was studied, it was assumed that \R always
967        matched two bytes, thus causing the minimum subject length to be
968        incorrectly computed because \R can also match just one byte.
970    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
971        was incorrectly computed.
973    20. If /S is present twice on a test pattern in pcretest input, it now
974        *disables* studying, thereby overriding the use of -s on the command line
975        (see 14 above). This is necessary for one or two tests to keep the output
976        identical in both cases.
978    21. When (*ACCEPT) was used in an assertion that matched an empty string and
979        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
981    22. When an atomic group that contained a capturing parenthesis was
982        successfully matched, but the branch in which it appeared failed, the
983        capturing was not being forgotten if a higher numbered group was later
984        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
985        group 1 to "a", when in fact it should be unset. This applied to multi-
986        branched capturing and non-capturing groups, repeated or not, and also to
987        positive assertions (capturing in negative assertions does not happen
988        in PCRE) and also to nested atomic groups.
990    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
991        subject after a captured substring, to make it easier to tell which of a
992        number of identical substrings has been captured.
994    24. The way atomic groups are processed by pcre_exec() has been changed so that
995        if they are repeated, backtracking one repetition now resets captured
996        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
997        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
998        "aaa". Previously, it would have been "a". As part of this code
999        refactoring, the way recursive calls are handled has also been changed.
1001    25. If an assertion condition captured any substrings, they were not passed
1002        back unless some other capturing happened later. For example, if
1003        (?(?=(a))a) was matched against "a", no capturing was returned.
1005    26. When studying a pattern that contained subroutine calls or assertions,
1006        the code for finding the minimum length of a possible match was handling
1007        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
1008        group 1 called group 2 while simultaneously a separate group 2 called group
1009        1). A stack overflow occurred in this case. I have fixed this by limiting
1010        the recursion depth to 10.
1012    27. Updated RunTest.bat in the distribution to the version supplied by Tom
1013        Fortmann. This supports explicit test numbers on the command line, and has
1014        argument validation and error reporting.
1016    28. An instance of \X with an unlimited repeat could fail if at any point the
1017        first character it looked at was a mark character.
1019    29. Some minor code refactoring concerning Unicode properties and scripts
1020        should reduce the stack requirement of match() slightly.
1022    30. Added the '=' option to pcretest to check the setting of unused capturing
1023        slots at the end of the pattern, which are documented as being -1, but are
1024        not included in the return count.
1026    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
1027        compiled something random. Now it gives a compile-time error (as does
1028        Perl).
1030    32. A *MARK encountered during the processing of a positive assertion is now
1031        recorded and passed back (compatible with Perl).
1033    33. If --only-matching or --colour was set on a pcregrep call whose pattern
1034        had alternative anchored branches, the search for a second match in a line
1035        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
1036        matched the line "0102" twice. The same bug affected patterns that started
1037        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
1038        twice.
1040    34. Previously, PCRE did not allow quantification of assertions. However, Perl
1041        does, and because of capturing effects, quantifying parenthesized
1042        assertions may at times be useful. Quantifiers are now allowed for
1043        parenthesized assertions.
1045    35. A minor code tidy in pcre_compile() when checking options for \R usage.
1047    36. \g was being checked for fancy things in a character class, when it should
1048        just be a literal "g".
1050    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
1051        appearance of a nested POSIX class supersedes an apparent external class.
1052        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
1053        unescaped square brackets may also appear as part of class names. For
1054        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
1055        more like Perl. (But see 8.20/1 above.)
1057    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
1058        was because it thought it was \N{name}, which is not supported).
1060    39. Add minix to OS list not supporting the -S option in pcretest.
1062    40. PCRE tries to detect cases of infinite recursion at compile time, but it
1063        cannot analyze patterns in sufficient detail to catch mutual recursions
1064        such as ((?1))((?2)). There is now a runtime test that gives an error if a
1065        subgroup is called recursively as a subpattern for a second time at the
1066        same position in the subject string. In previous releases this might have
1067        been caught by the recursion limit, or it might have run out of stack.
1069    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
1070        happen only once. PCRE was, however incorrectly giving a compile time error
1071        "recursive call could loop indefinitely" because it cannot analyze the
1072        pattern in sufficient detail. The compile time test no longer happens when
1073        PCRE is compiling a conditional subpattern, but actual runaway loops are
1074        now caught at runtime (see 40 above).
1076    42. It seems that Perl allows any characters other than a closing parenthesis
1077        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
1078        has been changed to be the same.
1080    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
1081        as not to get warnings when autogen.sh is called. Also changed
1082        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
1084    44. To help people who use pcregrep to scan files containing exceedingly long
1085        lines, the following changes have been made:
1087        (a) The default value of the buffer size parameter has been increased from
1088            8K to 20K. (The actual buffer used is three times this size.)
1090        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
1091            PCRE is built.
1093        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
1094            to be set at run time.
1096        (d) Numerical values in pcregrep options can be followed by K or M, for
1097            example --buffer-size=50K.
1099        (e) If a line being scanned overflows pcregrep's buffer, an error is now
1100            given and the return code is set to 2.
1102    45. Add a pointer to the latest mark to the callout data block.
1104    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
1105        partial match of an empty string instead of no match. This was specific to
1106        the use of ".".
1108    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
1109        complete match instead of a partial match. This bug was dependent on both
1110        the PCRE_UTF8 and PCRE_DOTALL options being set.
1112    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1113        starting byte set, because \b was not being ignored.
1116    Version 8.12 15-Jan-2011
1117    ------------------------
1119    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1120        checks for such things as part of the documentation building process.
1122    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1123        --match-limit and --recursion-limit options (added for 8.11). In
1124        particular, this made one of the standard tests fail. (The integer value
1125        went into the wrong half of a long int.)
1127    3.  If the --colour option was given to pcregrep with -v (invert match), it
1128        did strange things, either producing crazy output, or crashing. It should,
1129        of course, ignore a request for colour when reporting lines that do not
1130        match.
1132    4.  Another pcregrep bug caused similar problems if --colour was specified with
1133        -M (multiline) and the pattern match finished with a line ending.
1135    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1136        matched in multiline mode, the following line was shown as part of the
1137        match. This seems wrong, so I have changed it.
1139    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1140        the check for further matches in the same line (so they could be coloured)
1141        to overrun the end of the current line. If another match was found, it was
1142        incorrectly shown (and then shown again when found in the next line).
1144    7.  If pcregrep was compiled under Windows, there was a reference to the
1145        function pcregrep_exit() before it was defined. I am assuming this was
1146        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1147        reported by a user. I've moved the definition above the reference.
1150    Version 8.11 10-Dec-2010
1151    ------------------------
1153    1.  (*THEN) was not working properly if there were untried alternatives prior
1154        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1155        backtracked to try for "b" instead of moving to the next alternative branch
1156        at the same level (in this case, to look for "c"). The Perl documentation
1157        is clear that when (*THEN) is backtracked onto, it goes to the "next
1158        alternative in the innermost enclosing group".
1160    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1161        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1162        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1163        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1164        (*THEN).
1166    3.  If \s appeared in a character class, it removed the VT character from
1167        the class, even if it had been included by some previous item, for example
1168        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1169        of \s, but is part of the POSIX "space" class.)
1171    4.  A partial match never returns an empty string (because you can always
1172        match an empty string at the end of the subject); however the checking for
1173        an empty string was starting at the "start of match" point. This has been
1174        changed to the "earliest inspected character" point, because the returned
1175        data for a partial match starts at this character. This means that, for
1176        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1177        (previously it gave "no match").
1179    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1180        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1181        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1182        has an implication that the given string is incomplete (because a partial
1183        match is preferred over a full match). For this reason, these items now
1184        give a partial match in this situation. [Aside: previously, the one case
1185        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1186        match rather than a full match, which was wrong by the old rules, but is
1187        now correct.]
1189    6.  There was a bug in the handling of #-introduced comments, recognized when
1190        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1191        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1192        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1193        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1194        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1195        places in pcre_compile().
1197    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1198        comments when looking ahead for named forward references to subpatterns,
1199        the only newline sequence it recognized was NL. It now handles newlines
1200        according to the set newline convention.
1202    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1203        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1204        cater for a lack of strerror(). These oversights have been fixed.
1206    9.  Added --match-limit and --recursion-limit to pcregrep.
1208    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1210    11. When the -o option was used, pcregrep was setting a return code of 1, even
1211        when matches were found, and --line-buffered was not being honoured.
1213    12. Added an optional parentheses number to the -o and --only-matching options
1214        of pcregrep.
1216    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1217        can match an empty string. The code to do it in pcretest and pcredemo
1218        needed fixing:
1220        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1221            only one byte after an empty string match just before CRLF (this case
1222            just got forgotten; "any" and "anycrlf" were OK).
1224        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1225            mode when an empty string match preceded an ASCII character followed by
1226            a non-ASCII character. (The code for advancing by one character rather
1227            than one byte was nonsense.)
1229        (c) The pcredemo.c sample program did not have any code at all to handle
1230            the cases when CRLF is a valid newline sequence.
1232    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1233        as a starting offset was within the subject string. There is now a new
1234        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1235        negative or greater than the length of the string. In order to test this,
1236        pcretest is extended to allow the setting of negative starting offsets.
1238    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1239        starting offset points to the beginning of a UTF-8 character was
1240        unnecessarily clumsy. I tidied it up.
1242    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1243        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1245    17. Nobody had reported that the --include_dir option, which was added in
1246        release 7.7 should have been called --include-dir (hyphen, not underscore)
1247        for compatibility with GNU grep. I have changed it to --include-dir, but
1248        left --include_dir as an undocumented synonym, and the same for
1249        --exclude-dir, though that is not available in GNU grep, at least as of
1250        release 2.5.4.
1252    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1253        characters from a string of bytes) have been redefined so as not to use
1254        loops, in order to improve performance in some environments. At the same
1255        time, I abstracted some of the common code into auxiliary macros to save
1256        repetition (this should not affect the compiled code).
1258    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1259        compile-time error is now given if \c is not followed by an ASCII
1260        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1261        different, and any byte value is allowed.)
1263    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1264        START_OPTIMIZE option, which is now allowed at compile time - but just
1265        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1266        to pcregrep and other applications that have no direct access to PCRE
1267        options. The new /Y option in pcretest sets this option when calling
1268        pcre_compile().
1270    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1271        back references. Groups containing recursive back references were forced to
1272        be atomic by that change, but in the case of named groups, the amount of
1273        memory required was incorrectly computed, leading to "Failed: internal
1274        error: code overflow". This has been fixed.
1276    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1277        pcretest.c, to avoid build problems in some Borland environments.
1280    Version 8.10 25-Jun-2010
1281    ------------------------
1283    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1284        THEN.
1286    2.  (*ACCEPT) was not working when inside an atomic group.
1288    3.  Inside a character class, \B is treated as a literal by default, but
1289        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1290        causes the error). The code is unchanged, but I tidied the documentation.
1292    4.  Inside a character class, PCRE always treated \R and \X as literals,
1293        whereas Perl faults them if its -w option is set. I have changed PCRE so
1294        that it faults them when PCRE_EXTRA is set.
1296    5.  Added support for \N, which always matches any character other than
1297        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1299    6.  When compiling pcregrep with newer versions of gcc which may have
1300        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1301        declared with attribute warn_unused_result" were given. Just casting the
1302        result to (void) does not stop the warnings; a more elaborate fudge is
1303        needed. I've used a macro to implement this.
1305    7.  Minor change to pcretest.c to avoid a compiler warning.
1307    8.  Added four artifical Unicode properties to help with an option to make
1308        \s etc use properties (see next item). The new properties are: Xan
1309        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1311    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1312        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1313        this option. Modified pcretest to add /W to test this facility. Added
1314        REG_UCP to make it available via the POSIX interface.
1316    10. Added --line-buffered to pcregrep.
1318    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1319        studied, and the match started with a letter with a code point greater than
1320        127 whose first byte was different to the first byte of the other case of
1321        the letter, the other case of this starting letter was not recognized
1322        (#976).
1324    12. If a pattern that was studied started with a repeated Unicode property
1325        test, for example, \p{Nd}+, there was the theoretical possibility of
1326        setting up an incorrect bitmap of starting bytes, but fortunately it could
1327        not have actually happened in practice until change 8 above was made (it
1328        added property types that matched character-matching opcodes).
1330    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1331        possible starting bytes for non-anchored patterns.
1333    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1334        \R, and also a number of cases that involve Unicode properties, both
1335        explicit and implicit when PCRE_UCP is set.
1337    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1338        input, it could crash or give wrong results if characters with values
1339        greater than 0xc0 were present in the subject string. (Detail: it assumed
1340        UTF-8 input when processing these items.)
1342    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1343        size_t is 64-bit (#991).
1345    17. Added a check for running out of memory when PCRE is compiled with
1346        --disable-stack-for-recursion (#990).
1348    18. If the last data line in a file for pcretest does not have a newline on
1349        the end, a newline was missing in the output.
1351    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1352        less than 128) in its various bitmaps. However, there is a facility for
1353        generating tables according to the current locale when PCRE is compiled. It
1354        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1355        characters, are recognized by isspace() and therefore were getting set in
1356        these tables, and indeed these tables seem to approximate to ISO 8859. This
1357        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1358        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1359        which of course cannot start UTF-8 characters. I have changed the code so
1360        that only real ASCII characters (less than 128) and the correct starting
1361        bytes for UTF-8 encodings are set for characters greater than 127 when in
1362        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1363        altogether.)
1365    20. Added the /T option to pcretest so as to be able to run tests with non-
1366        standard character tables, thus making it possible to include the tests
1367        used for 19 above in the standard set of tests.
1369    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1370        reference to a subpattern the other side of a comment that contains an
1371        opening parenthesis caused either an internal compiling error, or a
1372        reference to the wrong subpattern.
1375    Version 8.02 19-Mar-2010
1376    ------------------------
1378    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1380    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1381        configured.
1383    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1384        original author of that file, following a query about its status.
1386    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1387        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1389    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1390        quantifier applied to a forward-referencing subroutine call, could compile
1391        incorrect code or give the error "internal error: previously-checked
1392        referenced subpattern not found".
1394    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1395        variables to point to external functions. For these systems, therefore,
1396        pcre_malloc etc. are now initialized to local functions that call the
1397        relevant global functions.
1399    7.  There were two entries missing in the vectors called coptable and poptable
1400        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1401        I've fixed the data, and added a kludgy way of testing at compile time that
1402        the lengths are correct (equal to the number of opcodes).
1404    8.  Following on from 7, I added a similar kludge to check the length of the
1405        eint vector in pcreposix.c.
1407    9.  Error texts for pcre_compile() are held as one long string to avoid too
1408        much relocation at load time. To find a text, the string is searched,
1409        counting zeros. There was no check for running off the end of the string,
1410        which could happen if a new error number was added without updating the
1411        string.
1413    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1415    11. \K was not working if it appeared in an atomic group or in a group that
1416        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1417        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1418        the assertion is positive, but not if it is negative.
1420    12. Change 11 fortuitously reduced the size of the stack frame used in the
1421        "match()" function of pcre_exec.c by one pointer. Forthcoming
1422        implementation of support for (*MARK) will need an extra pointer on the
1423        stack; I have reserved it now, so that the stack frame size does not
1424        decrease.
1426    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1427        item in branch that calls a recursion is a subroutine call - as in the
1428        second branch in the above example - was incorrectly given the compile-
1429        time error "recursive call could loop indefinitely" because pcre_compile()
1430        was not correctly checking the subroutine for matching a non-empty string.
1432    14. The checks for overrunning compiling workspace could trigger after an
1433        overrun had occurred. This is a "should never occur" error, but it can be
1434        triggered by pathological patterns such as hundreds of nested parentheses.
1435        The checks now trigger 100 bytes before the end of the workspace.
1437    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1440    Version 8.01 19-Jan-2010
1441    ------------------------
1443    1.  If a pattern contained a conditional subpattern with only one branch (in
1444        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1445        computed the wrong minimum data length (which is of course zero for such
1446        subpatterns). This could cause incorrect "no match" results.
1448    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1449        the pattern is reset in the first branch, pcre_compile() failed with
1450        "internal error: code overflow at offset...". This happened only when
1451        the reset was to the original external option setting. (An optimization
1452        abstracts leading options settings into an external setting, which was the
1453        cause of this.)
1455    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1456        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1457        assertion pattern did not match (meaning that the assertion was true), it
1458        was incorrectly treated as false if the SKIP had been reached during the
1459        matching. This also applied to assertions used as conditions.
1461    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1462        assertion subpattern, including such a pattern used as a condition,
1463        unpredictable results occurred, instead of the error return
1466    5.  The C++ GlobalReplace function was not working like Perl for the special
1467        situation when an empty string is matched. It now does the fancy magic
1468        stuff that is necessary.
1470    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1471        removed. (These were left over from very, very early versions of PCRE.)
1473    7.  Some cosmetic changes to the code to make life easier when compiling it
1474        as part of something else:
1476        (a) Change DEBUG to PCRE_DEBUG.
1478        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1479            called "current" as "current_branch", to prevent a collision with the
1480            Linux macro when compiled as a kernel module.
1482        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1483            prevent a collision with the Linux macro when compiled as a kernel
1484            module.
1486    8.  In pcre_compile() there are some checks for integer overflows that used to
1487        cast potentially large values to (double). This has been changed to that
1488        when building, a check for int64_t is made, and if it is found, it is used
1489        instead, thus avoiding the use of floating point arithmetic. (There is no
1490        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1491        double.
1493    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1494        2005 (difference between two addresses compared to an unsigned value).
1496    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1497        custom one, because of the following reported problem in Windows:
1499          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1500              under Win32.
1501          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1502              therefore missing the function definition.
1503          - The compiler thus generates a "C" signature for the test function.
1504          - The linker fails to find the "C" function.
1505          - PCRE fails to configure if asked to do so against libbz2.
1507    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1508        messages were output:
1510          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1511          rerunning libtoolize, to keep the correct libtool macros in-tree.
1512          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1514        I have done both of these things.
1516    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1517        most of the time, it *can* run out if it is given a pattern that contains a
1518        runaway infinite recursion. I updated the discussion in the pcrestack man
1519        page.
1521    13. Now that we have gone to the x.xx style of version numbers, the minor
1522        version may start with zero. Using 08 or 09 is a bad idea because users
1523        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1524        interpreted as invalid octal numbers. I've updated the previous comment in
1525        configure.ac, and also added a check that gives an error if 08 or 09 are
1526        used.
1528    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1529        causing partial matching to fail when the end of the subject matched \W
1530        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1532    15. There were some discrepancies between the declarations in pcre_internal.h
1533        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1534        their definitions. The declarations used "const uschar *" and the
1535        definitions used USPTR. Even though USPTR is normally defined as "const
1536        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1537        reported that: "This difference in casting confuses some C++ compilers, for
1538        example, SunCC recognizes above declarations as different functions and
1539        generates broken code for hbpcre." I have changed the declarations to use
1540        USPTR.
1542    16. GNU libtool is named differently on some systems. The autogen.sh script now
1543        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1544        (FreeBSD).
1546    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1547        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1548        comment: "Figure out how to create a longlong from a string: strtoll and
1549        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1550        instance, but it only takes 2 args instead of 3!"
1552    18. A subtle bug concerned with back references has been fixed by a change of
1553        specification, with a corresponding code fix. A pattern such as
1554        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1555        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1556        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1557        same bug. Such groups have to be quantified to be useful, or contained
1558        inside another quantified group. (If there's no repetition, the reference
1559        can never match.) The problem arises because, having left the group and
1560        moved on to the rest of the pattern, a later failure that backtracks into
1561        the group uses the captured value from the final iteration of the group
1562        rather than the correct earlier one. I have fixed this in PCRE by forcing
1563        any group that contains a reference to itself to be an atomic group; that
1564        is, there cannot be any backtracking into it once it has completed. This is
1565        similar to recursive and subroutine calls.
1568    Version 8.00 19-Oct-09
1569    ----------------------
1571    1.  The table for translating pcre_compile() error codes into POSIX error codes
1572        was out-of-date, and there was no check on the pcre_compile() error code
1573        being within the table. This could lead to an OK return being given in
1574        error.
1576    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1577        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1578        in a Windows environment.
1580    3.  The pcregrep --count option prints the count for each file even when it is
1581        zero, as does GNU grep. However, pcregrep was also printing all files when
1582        --files-with-matches was added. Now, when both options are given, it prints
1583        counts only for those files that have at least one match. (GNU grep just
1584        prints the file name in this circumstance, but including the count seems
1585        more useful - otherwise, why use --count?) Also ensured that the
1586        combination -clh just lists non-zero counts, with no names.
1588    4.  The long form of the pcregrep -F option was incorrectly implemented as
1589        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1590        but it seems right to fix it, and I didn't think it was worth preserving
1591        the old behaviour.
1593    5.  The command line items --regex=pattern and --regexp=pattern were not
1594        recognized by pcregrep, which required --regex pattern or --regexp pattern
1595        (with a space rather than an '='). The man page documented the '=' forms,
1596        which are compatible with GNU grep; these now work.
1598    6.  No libpcreposix.pc file was created for pkg-config; there was just
1599        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1601    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1602        when UCP support is not needed, by modifying the Python script that
1603        generates it from Unicode data files. This should not matter if the module
1604        is correctly used as a library, but I received one complaint about 50K of
1605        unwanted data. My guess is that the person linked everything into his
1606        program rather than using a library. Anyway, it does no harm.
1608    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1609        was a minimum greater than 1 for a wide character in a possessive
1610        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1611        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1612        character. Chaos in the form of incorrect output or a compiling loop could
1613        result.
1615    9.  The restrictions on what a pattern can contain when partial matching is
1616        requested for pcre_exec() have been removed. All patterns can now be
1617        partially matched by this function. In addition, if there are at least two
1618        slots in the offset vector, the offset of the earliest inspected character
1619        for the match and the offset of the end of the subject are set in them when
1620        PCRE_ERROR_PARTIAL is returned.
1622    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1623        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1624        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1625        and may be more useful for multi-segment matching.
1627    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1628        used to be given if ever the end of the subject was reached; now it is
1629        given only if matching could not proceed because another character was
1630        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1631        string "Z", which now yields "no match" instead of "partial match". In the
1632        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1633        final character ended with (*FAIL).
1635    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1636        if the pattern had a "must contain" character that was already found in the
1637        earlier partial match, unless partial matching was again requested. For
1638        example, with the pattern /dog.(body)?/, the "must contain" character is
1639        "g". If the first part-match was for the string "dog", restarting with
1640        "sbody" failed. This bug has been fixed.
1642    13. The string returned by pcre_dfa_exec() after a partial match has been
1643        changed so that it starts at the first inspected character rather than the
1644        first character of the match. This makes a difference only if the pattern
1645        starts with a lookbehind assertion or \b or \B (\K is not supported by
1646        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1647        matching functions compatible, and I think it's the right thing to do.
1649    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1650        so that the demonstration program is easily available in environments where
1651        PCRE has not been installed from source.
1653    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1654        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1655        library.
1657    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1658        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1659        is not the first non-POSIX option to be added. Clearly some people find
1660        these options useful.
1662    17. If a caller to the POSIX matching function regexec() passes a non-zero
1663        value for nmatch with a NULL value for pmatch, the value of
1664        nmatch is forced to zero.
1666    18. RunGrepTest did not have a test for the availability of the -u option of
1667        the diff command, as RunTest does. It now checks in the same way as
1668        RunTest, and also checks for the -b option.
1670    19. If an odd number of negated classes containing just a single character
1671        interposed, within parentheses, between a forward reference to a named
1672        subpattern and the definition of the subpattern, compilation crashed with
1673        an internal error, complaining that it could not find the referenced
1674        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1675        [The bug was that it was starting one character too far in when skipping
1676        over the character class, thus treating the ] as data rather than
1677        terminating the class. This meant it could skip too much.]
1679    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1680        /g option in pcretest when the pattern contains \K, which makes it possible
1681        to have an empty string match not at the start, even when the pattern is
1682        anchored. Updated pcretest and pcredemo to use this option.
1684    21. If the maximum number of capturing subpatterns in a recursion was greater
1685        than the maximum at the outer level, the higher number was returned, but
1686        with unset values at the outer level. The correct (outer level) value is
1687        now given.
1689    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1690        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1691        make it do so. The string so far is captured, making this feature
1692        compatible with Perl.
1694    23. The tests have been re-organized, adding tests 11 and 12, to make it
1695        possible to check the Perl 5.10 features against Perl 5.10.
1697    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1698        pattern matches a fixed length string. PCRE did not allow this; now it
1699        does. Neither allows recursion.
1701    25. I finally figured out how to implement a request to provide the minimum
1702        length of subject string that was needed in order to match a given pattern.
1703        (It was back references and recursion that I had previously got hung up
1704        on.) This code has now been added to pcre_study(); it finds a lower bound
1705        to the length of subject needed. It is not necessarily the greatest lower
1706        bound, but using it to avoid searching strings that are too short does give
1707        some useful speed-ups. The value is available to calling programs via
1708        pcre_fullinfo().
1710    26. While implementing 25, I discovered to my embarrassment that pcretest had
1711        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1712        study optimizations had never been tested with that matching function.
1713        Oops. What is worse, even when it was passed study data, there was a bug in
1714        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1715        were also very few tests of studied patterns with pcre_dfa_exec().
1717    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1718        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1719        on the other side of the coin, they are no longer allowed to have different
1720        names, because these cannot be distinguished in PCRE, and this has caused
1721        confusion. (This is a difference from Perl.)
1723    28. When duplicate subpattern names are present (necessarily with different
1724        numbers, as required by 27 above), and a test is made by name in a
1725        conditional pattern, either for a subpattern having been matched, or for
1726        recursion in such a pattern, all the associated numbered subpatterns are
1727        tested, and the overall condition is true if the condition is true for any
1728        one of them. This is the way Perl works, and is also more like the way
1729        testing by number works.
1732    Version 7.9 11-Apr-09
1733    ---------------------
1735    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1736        (pcretest), all targets were linked against these libraries. This included
1737        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1738        libraries. This caused unwanted dependencies to be created. This problem
1739        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1740        pcretest is linked with readline.
1742    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1743        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1744        moved outside it again, because FALSE and TRUE are already defined in AIX,
1745        but BOOL is not.
1747    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1748        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1750    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1751        hyphens) following file names and line numbers when outputting matching
1752        lines. This is not true; no spaces are inserted. I have also clarified the
1753        wording for the --colour (or --color) option.
1755    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1756        was not coloured; this is different to GNU grep, so I have changed it to be
1757        the same.
1759    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1760        each matching line was coloured. Now it goes on to look for further matches
1761        of any of the test patterns, which is the same behaviour as GNU grep.
1763    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1764        doesn't make sense to accept an empty string match in pcregrep, so I have
1765        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1766        seems to be how GNU grep behaves. [But see later change 40 for release
1767        8.33.]
1769    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1770        start or after a newline", because the conditional assertion was not being
1771        correctly handled. The rule now is that both the assertion and what follows
1772        in the first alternative must satisfy the test.
1774    9.  If auto-callout was enabled in a pattern with a conditional group whose
1775        condition was an assertion, PCRE could crash during matching, both with
1776        pcre_exec() and pcre_dfa_exec().
1778    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1779        used for matching.
1781    11. Unicode property support in character classes was not working for
1782        characters (bytes) greater than 127 when not in UTF-8 mode.
1784    12. Added the -M command line option to pcretest.
1786    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1788    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1790    16. Added comments and documentation about mis-use of no_arg in the C++
1791        wrapper.
1793    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1794        from Martin Jerabek that uses macro names for all relevant character and
1795        string constants.
1797    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1798        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1799        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1800        these, but not everybody uses configure.
1802    19. A conditional group that had only one branch was not being correctly
1803        recognized as an item that could match an empty string. This meant that an
1804        enclosing group might also not be so recognized, causing infinite looping
1805        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1806        with the subject "ab", where knowledge that the repeated group can match
1807        nothing is needed in order to break the loop.
1809    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1810        exec(), but without supplying a callout function, matching went wrong.
1812    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1813        leak if the size of the offset vector was greater than 30. When the vector
1814        is smaller, the saved offsets during recursion go onto a local stack
1815        vector, but for larger vectors malloc() is used. It was failing to free
1816        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1817        error, in fact).
1819    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1820        heapframe that is used only when UTF-8 support is enabled. This caused no
1821        problem, but was untidy.
1823    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1824        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1825        included within another project.
1827    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1828        slightly modified by me:
1830          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1831              not building pcregrep.
1833          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1834              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1836    25. Forward references, both numeric and by name, in patterns that made use of
1837        duplicate group numbers, could behave incorrectly or give incorrect errors,
1838        because when scanning forward to find the reference group, PCRE was not
1839        taking into account the duplicate group numbers. A pattern such as
1840        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1842    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1843        the feature of a custom pointer more persuasive (as requested by a user).
1845    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1846        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1847        pre-defined.
1849    28. Added support for (*UTF8) at the start of a pattern.
1851    29. Arrange for flags added by the "release type" setting in CMake to be shown
1852        in the configuration summary.
1855    Version 7.8 05-Sep-08
1856    ---------------------
1858    1.  Replaced UCP searching code with optimized version as implemented for Ad
1859        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1860        stage table and inline lookup instead of a function, giving speed ups of 2
1861        to 5 times on some simple patterns that I tested. Permission was given to
1862        distribute the MultiStage2.py script that generates the tables (it's not in
1863        the tarball, but is in the Subversion repository).
1865    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1866        scripts.
1868    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1869        a group with a zero qualifier. The result of the study could be incorrect,
1870        or the function might crash, depending on the pattern.
1872    4.  Caseless matching was not working for non-ASCII characters in back
1873        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1874        It now works when Unicode Property Support is available.
1876    5.  In pcretest, an escape such as \x{de} in the data was always generating
1877        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1878        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1879        truncation.
1881    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1883    7.  Added two (int) casts to pcregrep when printing the difference of two
1884        pointers, in case they are 64-bit values.
1886    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1887        test 2 if it fails.
1889    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1890        and a #define of that name to empty if it is not externally set. This is to
1891        allow users of MSVC to set it if necessary.
1893    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1894        the convenience functions in the pcre_get.c source file.
1896    11. An option change at the start of a pattern that had top-level alternatives
1897        could cause overwriting and/or a crash. This command provoked a crash in
1898        some environments:
1900          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1902        This potential security problem was recorded as CVE-2008-2371.
1904    12. For a pattern where the match had to start at the beginning or immediately
1905        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1906        pcre_dfa_exec() could read past the end of the passed subject if there was
1907        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1908        pcretest so that it places the subject at the end of its malloc-ed buffer.
1910    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1911        exec() might read past the end of the data buffer in UTF-8 mode.
1913    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1914        the data contained the byte 0x85 as part of a UTF-8 character within its
1915        first line. This applied both to normal and DFA matching.
1917    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1918        /^[^d]*?$/8 failed to match "abc".
1920    16. Added a missing copyright notice to pcrecpp_internal.h.
1922    17. Make it more clear in the documentation that values returned from
1923        pcre_exec() in ovector are byte offsets, not character counts.
1925    18. Tidied a few places to stop certain compilers from issuing warnings.
1927    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1928        supplied by Stefan Weber. I made a further small update for 7.8 because
1929        there is a change of source arrangements: the pcre_searchfuncs.c module is
1930        replaced by pcre_ucd.c.
1933    Version 7.7 07-May-08
1934  ---------------------  ---------------------
1936  1.  Applied Craig's patch to sort out a long long problem: "If we can't convert  1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1937      a string to a long long, pretend we don't even have a long long." This is      a string to a long long, pretend we don't even have a long long." This is
1938      done by checking for the strtoq, strtoll, and _strtoi64 functions.      done by checking for the strtoq, strtoll, and _strtoi64 functions.
1940  2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with  2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1941      pre-7.6 versions, which defined a global no_arg variable instead of putting      pre-7.6 versions, which defined a global no_arg variable instead of putting
1942      it in the RE class.      it in the RE class. (See also #8 below.)
1944  3.  Remove a line of dead code, identified by coverity and reported by Nuno  3.  Remove a line of dead code, identified by coverity and reported by Nuno
1945      Lopes.      Lopes.
1947  4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:  4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1949      (1) The include/exclude patterns were being applied to the whole pathnames      (1) The include/exclude patterns were being applied to the whole pathnames
1950          of files, instead of just to the final components.          of files, instead of just to the final components.
1952      (2) If there was more than one level of directory, the subdirectories were      (2) If there was more than one level of directory, the subdirectories were
1953          skipped unless they satisfied the include/exclude conditions. This is          skipped unless they satisfied the include/exclude conditions. This is
1954          inconsistent with GNU grep (and could even be seen as contrary to the          inconsistent with GNU grep (and could even be seen as contrary to the
1955          pcregrep specification - which I improved to make it absolutely clear).          pcregrep specification - which I improved to make it absolutely clear).
1956          The action now is always to scan all levels of directory, and just          The action now is always to scan all levels of directory, and just
1957          apply the include/exclude patterns to regular files.          apply the include/exclude patterns to regular files.
1959  5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used  5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1960      --exclude_dir in the tests to avoid scanning .svn directories.      --exclude_dir in the tests to avoid scanning .svn directories.
1962  6.  Applied Craig's patch to the QuoteMeta function so that it escapes the  6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1963      NUL character as backslash + 0 rather than backslash + NUL, because PCRE      NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1964      doesn't support NULs in patterns.      doesn't support NULs in patterns.
1966    7.  Added some missing "const"s to declarations of static tables in
1967        pcre_compile.c and pcre_dfa_exec.c.
1969    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1970        caused by fix #2  above. (Subsequently also a second patch to fix the
1971        first patch. And a third patch - this was a messy problem.)
1973    9.  Applied Craig's patch to remove the use of push_back().
1975    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1976        matching function regexec().
1978    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1979        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1980        references. PCRE supports relative numbers with this syntax (I don't think
1981        Oniguruma does).
1983    12. Previously, a group with a zero repeat such as (...){0} was completely
1984        omitted from the compiled regex. However, this means that if the group
1985        was called as a subroutine from elsewhere in the pattern, things went wrong
1986        (an internal error was given). Such groups are now left in the compiled
1987        pattern, with a new opcode that causes them to be skipped at execution
1988        time.
1990    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1991        to the way PCRE behaves:
1993        (a) A lone ] character is dis-allowed (Perl treats it as data).
1995        (b) A back reference to an unmatched subpattern matches an empty string
1996            (Perl fails the current match path).
1998        (c) A data ] in a character class must be notated as \] because if the
1999            first data character in a class is ], it defines an empty class. (In
2000            Perl it is not possible to have an empty class.) The empty class []
2001            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
2002            The negative empty class [^] matches any one character, independently
2003            of the DOTALL setting.
2005    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
2006        non-existent subpattern following a character class starting with ']' and
2007        containing () gave an internal compiling error instead of "reference to
2008        non-existent subpattern". Fortunately, when the pattern did exist, the
2009        compiled code was correct. (When scanning forwards to check for the
2010        existence of the subpattern, it was treating the data ']' as terminating
2011        the class, so got the count wrong. When actually compiling, the reference
2012        was subsequently set up correctly.)
2014    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
2015        it was being rejected as not supported by pcre_dfa_exec(), even though
2016        other assertions are supported. I have made pcre_dfa_exec() support
2017        (*FAIL).
2019    16. The implementation of 13c above involved the invention of a new opcode,
2020        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
2021        cannot be changed at match time, I realized I could make a small
2022        improvement to matching performance by compiling OP_ALLANY instead of
2023        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
2024        on the OP_ANY path.
2026    17. Compiling pcretest on Windows with readline support failed without the
2027        following two fixes: (1) Make the unistd.h include conditional on
2028        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
2030    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
2031        ncurses library to be included for pcretest when ReadLine support is
2032        requested, but also to allow for it to be overridden. This patch came from
2033        Daniel Bergström.
2035    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
2036        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
2037        any errors with the current Unicode tables. Thanks to Peter Kankowski for
2038        spotting this.
2041  Version 7.6 28-Jan-08  Version 7.6 28-Jan-08
2042  ---------------------  ---------------------

Removed from v.326  
changed lines
  Added in v.1363

  ViewVC Help
Powered by ViewVC 1.1.5