ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 167 by ph10, Wed May 9 15:53:54 2007 UTC revision 1274 by ph10, Fri Mar 8 11:35:41 2013 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4  Version 7.2 01-May-07  Version 8.33 xx-xxxx-201x
5    -------------------------
7    1.  Added 'U' to some constants that are compared to unsigned integers, to
8        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
9        variables that are added to signed variables, to ensure the result is
10        signed and can be negated.
12    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
13        CMake config files.
15    3.  Revise the creation of config.h.generic so that all boolean macros are
16        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
17        overriding via -D on the command line possible.
19    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
20        to unsigned int is reported to make a quite noticeable speed difference in
21        a specific Windows environment. Testing on Linux did also appear to show
22        some benefit (and it is clearly not harmful). Also fixed the definition of
23        Xop which should be unsigned.
25    5.  Related to (4), changing the definition of the intermediate variable cc
26        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
27        improvements.
29    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
30        unnecessary spaces.
32    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
33        and later.
35    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
37    9.  Optimizing fast_forward_start_bits in JIT.
39    10. Adding experimental support for callouts in JIT, and fixing some
40        issues revealed during this work. Namely:
42        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
44        (b) Minimum length was not checked before the matching is started.
46    11. The value of capture_last that is passed to callouts was incorrect in some
47        cases when there was a capture on one path that was subsequently abandoned
48        after a backtrack. Also, the capture_last value is now reset after a
49        recursion, since all captures are also reset in this case.
51    12. The interpreter no longer returns the "too many substrings" error in the
52        case when an overflowing capture is in a branch that is subsequently
53        abandoned after a backtrack.
55    13. In the pathological case when an offset vector of size 2 is used, pcretest
56        now prints out the matched string after a yield of 0 or 1.
58    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
59        Only supported by the JIT compiler at the moment.
61    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
63    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
64        offset of the starting point of the matching process, provided the offsets
65        vector is large enough.
67    17. The \A escape now records a lookbehind value of 1, though its execution
68        does not actually inspect the previous character. This is to ensure that,
69        in partial multi-segment matching, at least one character from the old
70        segment is retained when a new segment is processed. Otherwise, if there
71        are no lookbehinds in the pattern, \A might match incorrectly at the start
72        of a new segment.
74    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
76    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
77        gives some modest performance improvement in 8-bit mode.
79    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
80        be expressed in certain programming languages using Universal Character
81        Names.
83    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
84        which points out that "non characters" are not "characters that may not
85        appear in Unicode strings" but rather "characters that are reserved for
86        internal use and have only local meaning".
88    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
89        there was a conditional group that depended on an assertion, if the
90        assertion was false, the callout that immediately followed the alternation
91        in the condition was skipped when pcre_exec() was used for matching.
93    23. Allow an explicit callout to be inserted before an assertion that is the
94        condition for a conditional group, for compatibility with automatic
95        callouts, which always insert a callout at this point.
97    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
98        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
100    25. (*PRUNE) is now supported by the JIT compiler.
102    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
104    27. Fix the case where there are two or more SKIPs with arguments that may be
105        ignored.
108    Version 8.32 30-November-2012
109    -----------------------------
111    1.  Improved JIT compiler optimizations for first character search and single
112        character iterators.
114    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
115        Patch by Daniel Richard G.
117    3.  Single character iterator optimizations in the JIT compiler.
119    4.  Improved JIT compiler optimizations for character ranges.
121    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
122        Reported by Giuseppe D'Angelo.
124    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
125        of a line, was being set incorrectly in cases where .* appeared inside
126        atomic brackets at the start of a pattern, or where there was a subsequent
127        *PRUNE or *SKIP.
129    7.  Improved instruction cache flush for POWER/PowerPC.
130        Patch by Daniel Richard G.
132    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
133        grep:
135        (a) There is now no limit to the number of patterns to be matched.
137        (b) An error is given if a pattern is too long.
139        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
140            are now supported.
142        (d) --exclude-from and --include-from (multiple use) have been added.
144        (e) Exclusions and inclusions now apply to all files and directories, not
145            just to those obtained from scanning a directory recursively.
147        (f) Multiple uses of -f and --file-list are now supported.
149        (g) In a Windows environment, the default for -d has been changed from
150            "read" (the GNU grep default) to "skip", because otherwise the presence
151            of a directory in the file list provokes an error.
153        (h) The documentation has been revised and clarified in places.
155    9.  Improve the matching speed of capturing brackets.
157    10. Changed the meaning of \X so that it now matches a Unicode extended
158        grapheme cluster.
160    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
161        out POSIX threads when JIT support is configured.
163    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
165    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
166        structure could go wrong in environments where size_t is not the same size
167        as int.
169    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
170        set.
172    15. The EBCDIC support had decayed; later updates to the code had included
173        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
174        general tidy up of EBCDIC-related issues, and the documentation was also
175        not quite right. There is now a test that can be run on ASCII systems to
176        check some of the EBCDIC-related things (but is it not a full test).
178    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
179        in a small tidy to the code.
181    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
183    18. If the --only-matching (-o) option in pcregrep is specified multiple
184        times, each one causes appropriate output. For example, -o1 -o2 outputs the
185        substrings matched by the 1st and 2nd capturing parentheses. A separating
186        string can be specified by --om-separator (default empty).
188    19. Improving the first n character searches.
190    20. Turn case lists for horizontal and vertical white space into macros so that
191        they are defined only once.
193    21. This set of changes together give more compatible Unicode case-folding
194        behaviour for characters that have more than one other case when UCP
195        support is available.
197        (a) The Unicode property table now has offsets into a new table of sets of
198            three or more characters that are case-equivalent. The MultiStage2.py
199            script that generates these tables (the pcre_ucd.c file) now scans
200            CaseFolding.txt instead of UnicodeData.txt for character case
201            information.
203        (b) The code for adding characters or ranges of characters to a character
204            class has been abstracted into a generalized function that also handles
205            case-independence. In UTF-mode with UCP support, this uses the new data
206            to handle characters with more than one other case.
208        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
209            whose other case is greater than 256 are now correctly matched
210            caselessly. Previously, the high codepoint matched the low one, but not
211            vice versa.
213        (d) The processing of \h, \H, \v, and \ in character classes now makes use
214            of the new class addition function, using character lists defined as
215            macros alongside the case definitions of 20 above.
217        (e) Caseless back references now work with characters that have more than
218            one other case.
220        (f) General caseless matching of characters with more than one other case
221            is supported.
223    22. Unicode character properties were updated from Unicode 6.2.0
225    23. Improved CMake support under Windows. Patch by Daniel Richard G.
227    24. Add support for 32-bit character strings, and UTF-32
229    25. Major JIT compiler update (code refactoring and bugfixing).
230        Experimental Sparc 32 support is added.
232    26. Applied a modified version of Daniel Richard G's patch to create
233        pcre.h.generic and config.h.generic by "make" instead of in the
234        PrepareRelease script.
236    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
237        pcre_compile.c when checking for a zero character.
239    28. Introducing a native interface for JIT. Through this interface, the compiled
240        machine code can be directly executed. The purpose of this interface is to
241        provide fast pattern matching, so several sanity checks are not performed.
242        However, feature tests are still performed. The new interface provides
243        1.4x speedup compared to the old one.
245    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
246        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
247        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
249    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
250        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
251        the "old" RFC 2279). Instead, it ended up passing a negative length to
252        pcre_exec().
254    31. Add support for GCC's visibility feature to hide internal functions.
256    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
257        "unknown -C option" after outputting 0 or 1.
259    33. There is now support for generating a code coverage report for the test
260        suite in environments where gcc is the compiler and lcov is installed. This
261        is mainly for the benefit of the developers.
263    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
264        unaddressable using valgrind annotations, allowing valgrind to detect
265        invalid memory accesses. This is mainly for the benefit of the developers.
267    25. (*UTF) can now be used to start a pattern in any of the three libraries.
269    26. Give configure error if --enable-cpp but no C++ compiler found.
272    Version 8.31 06-July-2012
273    -------------------------
275    1.  Fixing a wrong JIT test case and some compiler warnings.
277    2.  Removed a bashism from the RunTest script.
279    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
280        to unsigned type, result still unsigned" that was given by an MS compiler
281        on encountering the code "-sizeof(xxx)".
283    4.  Partial matching support is added to the JIT compiler.
285    5.  Fixed several bugs concerned with partial matching of items that consist
286        of more than one character:
288        (a) /^(..)\1/ did not partially match "aba" because checking references was
289            done on an "all or nothing" basis. This also applied to repeated
290            references.
292        (b) \R did not give a hard partial match if \r was found at the end of the
293            subject.
295        (c) \X did not give a hard partial match after matching one or more
296            characters at the end of the subject.
298        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
299            a partial match for the string "\r".
301        (e) When newline was set to CRLF, the metacharacter "." did not recognize
302            a partial match for a CR character at the end of the subject string.
304    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
305        running pcretest, the text "(JIT)" added to the output whenever JIT is
306        actually used to run the match.
308    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
309        or /S+[+] with a digit between 1 and 7.
311    8.  OP_NOT now supports any UTF character not just single-byte ones.
313    9.  (*MARK) control verb is now supported by the JIT compiler.
315    10. The command "./RunTest list" lists the available tests without actually
316        running any of them. (Because I keep forgetting what they all are.)
320    12. Applied a (slightly modified) user-supplied patch that improves performance
321        when the heap is used for recursion (compiled with --disable-stack-for-
322        recursion). Instead of malloc and free for each heap frame each time a
323        logical recursion happens, frames are retained on a chain and re-used where
324        possible. This sometimes gives as much as 30% improvement.
326    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
327        call.
329    14. As documented, (*COMMIT) is now confined to within a positive assertion.
331    15. It is now possible to link pcretest with libedit as an alternative to
332        libreadline.
334    16. (*COMMIT) control verb is now supported by the JIT compiler.
336    17. The Unicode data tables have been updated to Unicode 6.1.0.
338    18. Added --file-list option to pcregrep.
340    19. Added binary file support to pcregrep, including the -a, --binary-files,
341        -I, and --text options.
343    20. The madvise function is renamed for posix_madvise for QNX compatibility
344        reasons. Fixed by Giuseppe D'Angelo.
346    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
348    22. Changed the option for creating symbolic links for 16-bit man pages from
349        -s to -sf so that re-installing does not cause issues.
351    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
353    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
354        matches in certain environments (the workspace was not being correctly
355        retained). Also added to pcre_dfa_exec() a simple plausibility check on
356        some of the workspace data at the beginning of a restart.
358    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
359        was not doing so when it should - probably a typo introduced by SVN 528
360        (change 8.10/14).
362    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
363        \w+ when the character tables indicated that \x{c4} was a word character.
364        There were several related cases, all because the tests for doing a table
365        lookup were testing for characters less than 127 instead of 255.
367    27. If a pattern contains capturing parentheses that are not used in a match,
368        their slots in the ovector are set to -1. For those that are higher than
369        any matched groups, this happens at the end of processing. In the case when
370        there were back references that the ovector was too small to contain
371        (causing temporary malloc'd memory to be used during matching), and the
372        highest capturing number was not used, memory off the end of the ovector
373        was incorrectly being set to -1. (It was using the size of the temporary
374        memory instead of the true size.)
376    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
377        ovector size, it uses memory at the end of the block that it has got.
379    29. Check for an overlong MARK name and give an error at compile time. The
380        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
382    30. JIT compiler update.
384    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
385        Rill for the patch.
387    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
389    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
391    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
392        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
393        was enabled.
395    35. Improve JIT code generation for greedy plus quantifier.
397    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
398        1 to "aa" instead of to an empty string. The bug affected repeated groups
399        that could potentially match an empty string.
401    37. Optimizing single character iterators in JIT.
403    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
404        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
405        codepoints that are too big for the mode are faulted, and in a UTF mode,
406        disallowed codepoints are also faulted.
408    39. If PCRE was compiled with UTF support, in three places in the DFA
409        matcher there was code that should only have been obeyed in UTF mode, but
410        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
411        processing when bytes with values greater than 127 were present. In 16-bit
412        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
413        both cases the values are those that cannot be the first data item in a UTF
414        character. The three items that might have provoked this were recursions,
415        possessively repeated groups, and atomic groups.
417    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
418        and pcregrep, because some OS require shared objects to be explicitly
419        passed to ld, causing the link step to fail if they are not.
421    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
422        mode, patterns that started with \h* or \R* might be incorrectly matched.
425    Version 8.30 04-February-2012
426    -----------------------------
428    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
429        name is defined in ctype.h.
431    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
432        only in quite long subpatterns.
434    3.  Removed the function pcre_info(), which has been obsolete and deprecated
435        since it was replaced by pcre_fullinfo() in February 2000.
437    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
438        match a (*MARK), and the match failed at the start of the subject, a
439        reference to memory before the start of the subject could occur. This bug
440        was introduced by fix 17 of release 8.21.
442    5.  A reference to an unset group with zero minimum repetition was giving
443        totally wrong answers (in non-JavaScript-compatibility mode). For example,
444        /(another)?(\1?)test/ matched against "hello world test". This bug was
445        introduced in release 8.13.
447    6.  Add support for 16-bit character strings (a large amount of work involving
448        many changes and refactorings).
450    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
451        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
452        from a file.
454    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
455        rounding is not applied in this particular case).
457    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
458        if they appear, or are escaped, in patterns.
460    10. Get rid of a number of -Wunused-but-set-variable warnings.
462    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
463        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
464        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
465        also returns the mark "x". This bug applied to capturing parentheses,
466        non-capturing parentheses, and atomic parentheses. It also applied to some
467        assertions.
469    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
470        information out of configure.ac instead of relying on pcre.h.generic, which
471        is not stored in the repository.
473    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
474        -lreadline.
476    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
478    15. Applied Graycode's patch to put the top-level frame on the stack rather
479        than the heap when not using the stack for recursion. This gives a
480        performance improvement in many cases when recursion is not deep.
482    16. Experimental code added to "pcretest -C" to output the stack frame size.
485    Version 8.21 12-Dec-2011
486    ------------------------
488    1.  Updating the JIT compiler.
490    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
491        are added as well.
493    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
494        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
495        calling _pcre_jit_exec. Some extra comments are added.
497    4.  (*MARK) settings inside atomic groups that do not contain any capturing
498        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
499        was introduced by change 18 for 8.20.
501    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
502        ECMA-262 standard.
504    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
505        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
506        This bug was probably introduced by change 9 of 8.13.
508    7.  While fixing 6 above, I noticed that a number of other items were being
509        incorrectly rejected as "not fixed length". This arose partly because newer
510        opcodes had not been added to the fixed-length checking code. I have (a)
511        corrected the bug and added tests for these items, and (b) arranged for an
512        error to occur if an unknown opcode is encountered while checking for fixed
513        length instead of just assuming "not fixed length". The items that were
514        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
515        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
516        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
518    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
519        being incorrectly compiled and would have given unpredicatble results.
521    9.  A possessively repeated subpattern with minimum repeat count greater than
522        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
523        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
524        the first (A) could occur when it should not.
526    10. Add a cast and remove a redundant test from the code.
528    11. JIT should use pcre_malloc/pcre_free for allocation.
530    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
531        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
532        is anything other than /usr, -L is still shown).
534    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
536    14. Perl does not support \N without a following name in a [] class; PCRE now
537        also gives an error.
539    15. If a forward reference was repeated with an upper limit of around 2000,
540        it caused the error "internal error: overran compiling workspace". The
541        maximum number of forward references (including repeats) was limited by the
542        internal workspace, and dependent on the LINK_SIZE. The code has been
543        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
544        the default depends on LINK_SIZE. There is a new upper limit (for safety)
545        of around 200,000 forward references. While doing this, I also speeded up
546        the filling in of repeated forward references.
548    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
549        incorrectly expecting the subject to contain another "a" after the start.
551    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
552        in the match, the SKIP should be ignored. This was not happening; instead
553        the SKIP was being treated as NOMATCH. For patterns such as
554        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
555        tested.
557    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
558        now much more compatible with Perl, in particular in cases where the result
559        is a non-match for a non-anchored pattern. For example, if
560        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
561        "m", where previously it did not return a name. A side effect of this
562        change is that for partial matches, the last encountered mark name is
563        returned, as for non matches. A number of tests that were previously not
564        Perl-compatible have been moved into the Perl-compatible test files. The
565        refactoring has had the pleasing side effect of removing one argument from
566        the match() function, thus reducing its stack requirements.
568    19. If the /S+ option was used in pcretest to study a pattern using JIT,
569        subsequent uses of /S (without +) incorrectly behaved like /S+.
571    21. Retrieve executable code size support for the JIT compiler and fixing
572        some warnings.
574    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
575        not work when the shorter character appeared right at the end of the
576        subject string.
578    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
579        systems.
581    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
582        output it when the /M option is used in pcretest.
584    25. The CheckMan script was not being included in the distribution. Also, added
585        an explicit "perl" to run Perl scripts from the PrepareRelease script
586        because this is reportedly needed in Windows.
588    26. If study data was being save in a file and studying had not found a set of
589        "starts with" bytes for the pattern, the data written to the file (though
590        never used) was taken from uninitialized memory and so caused valgrind to
591        complain.
593    27. Updated RunTest.bat as provided by Sheri Pierce.
595    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
597    29. Computation of memory usage for the table of capturing group names was
598        giving an unnecessarily large value.
601    Version 8.20 21-Oct-2011
602    ------------------------
604    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
605        a POSIX class. After further experiments with Perl, which convinced me that
606        Perl has bugs and confusions, a closing square bracket is no longer allowed
607        in a POSIX name. This bug also affected patterns with classes that started
608        with full stops.
610    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
611        captured substring, but while checking the failing first alternative,
612        substring 1 is temporarily captured. If the output vector supplied to
613        pcre_exec() was not big enough for this capture, the yield of the function
614        was still zero ("insufficient space for captured substrings"). This cannot
615        be totally fixed without adding another stack variable, which seems a lot
616        of expense for a edge case. However, I have improved the situation in cases
617        such as /(a)(b)x|abc/ matched against "abc", where the return code
618        indicates that fewer than the maximum number of slots in the ovector have
619        been set.
621    3.  Related to (2) above: when there are more back references in a pattern than
622        slots in the output vector, pcre_exec() uses temporary memory during
623        matching, and copies in the captures as far as possible afterwards. It was
624        using the entire output vector, but this conflicts with the specification
625        that only 2/3 is used for passing back captured substrings. Now it uses
626        only the first 2/3, for compatibility. This is, of course, another edge
627        case.
629    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
630        main code base, and can be used by building with --enable-jit. When this is
631        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
632        runtime --no-jit option is given.
634    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
635        ovector, the return from the function was zero, implying that there were
636        other matches that did not fit. The correct "exactly full" value is now
637        returned.
639    6.  If a subpattern that was called recursively or as a subroutine contained
640        (*PRUNE) or any other control that caused it to give a non-standard return,
641        invalid errors such as "Error -26 (nested recursion at the same subject
642        position)" or even infinite loops could occur.
644    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
645        computing the minimum length on reaching *ACCEPT, and so ended up with the
646        wrong value of 1 rather than 0. Further investigation indicates that
647        computing a minimum subject length in the presence of *ACCEPT is difficult
648        (think back references, subroutine calls), and so I have changed the code
649        so that no minimum is registered for a pattern that contains *ACCEPT.
651    8.  If (*THEN) was present in the first (true) branch of a conditional group,
652        it was not handled as intended. [But see 16 below.]
654    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
655        Sheri Pierce.
657    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
658        the first byte in a match must be "a".
660    11. Change 17 for 8.13 increased the recursion depth for patterns like
661        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
662        pattern contains any instances of (*THEN). If it does not, the old
663        optimizations are restored. It would be nice to do this on a per-group
664        basis, but at the moment that is not feasible.
666    12. In some environments, the output of pcretest -C is CRLF terminated. This
667        broke RunTest's code that checks for the link size. A single white space
668        character after the value is now allowed for.
670    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
671        For "fr", it uses the Windows-specific input and output files.
673    14. If (*THEN) appeared in a group that was called recursively or as a
674        subroutine, it did not work as intended. [But see next item.]
676    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
677        pattern fragments (but not containing any | characters). If A and B are
678        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
679        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
680        D. In other words, Perl considers parentheses that do not contain any |
681        characters to be part of a surrounding alternative, whereas PCRE was
682        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
683        differently. PCRE now behaves in the same way as Perl, except in the case
684        of subroutine/recursion calls such as (?1) which have in any case always
685        been different (but PCRE had them first :-).
687    16. Related to 15 above: Perl does not treat the | in a conditional group as
688        creating alternatives. Such a group is treated in the same way as an
689        ordinary group without any | characters when processing (*THEN). PCRE has
690        been changed to match Perl's behaviour.
692    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
693        RunGrepTest script failed.
695    18. Change 22 for version 13 caused atomic groups to use more stack. This is
696        inevitable for groups that contain captures, but it can lead to a lot of
697        stack use in large patterns. The old behaviour has been restored for atomic
698        groups that do not contain any capturing parentheses.
700    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
701        suppress the check for a minimum subject length at run time. (If it was
702        given to pcre_exec() or pcre_dfa_exec() it did work.)
704    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
705        fail to work when decoding hex characters in data strings in EBCDIC
706        environments.
708    21. It appears that in at least one Mac OS environment, the isxdigit() function
709        is implemented as a macro that evaluates to its argument more than once,
710        contravening the C 90 Standard (I haven't checked a later standard). There
711        was an instance in pcretest which caused it to go wrong when processing
712        \x{...} escapes in subject strings. The has been rewritten to avoid using
713        things like p++ in the argument of isxdigit().
716    Version 8.13 16-Aug-2011
717    ------------------------
719    1.  The Unicode data tables have been updated to Unicode 6.0.0.
721    2.  Two minor typos in pcre_internal.h have been fixed.
723    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
724        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
725        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
727    4.  There were a number of related bugs in the code for matching backrefences
728        caselessly in UTF-8 mode when codes for the characters concerned were
729        different numbers of bytes. For example, U+023A and U+2C65 are an upper
730        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
731        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
732        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
733        2-byte code at the end of the subject (it thought there wasn't enough data
734        left).
736    5.  Comprehensive information about what went wrong is now returned by
737        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
738        as the output vector has at least 2 elements. The offset of the start of
739        the failing character and a reason code are placed in the vector.
741    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
742        now returned is for the first byte of the failing character, instead of the
743        last byte inspected. This is an incompatible change, but I hope it is small
744        enough not to be a problem. It makes the returned offset consistent with
745        pcre_exec() and pcre_dfa_exec().
747    7.  pcretest now gives a text phrase as well as the error number when
748        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
749        failure, the offset and reason code are output.
751    8.  When \R was used with a maximizing quantifier it failed to skip backwards
752        over a \r\n pair if the subsequent match failed. Instead, it just skipped
753        back over a single character (\n). This seems wrong (because it treated the
754        two characters as a single entity when going forwards), conflicts with the
755        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
756        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
757        has been changed.
759    9.  Some internal refactoring has changed the processing so that the handling
760        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
761        time (the PCRE_DOTALL option was changed this way some time ago: version
762        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
763        which was always a bit of a fudge. It also means that there is one less
764        argument for the match() function, which reduces its stack requirements
765        slightly. This change also fixes an incompatibility with Perl: the pattern
766        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
768    10. More internal refactoring has drastically reduced the number of recursive
769        calls to match() for possessively repeated groups such as (abc)++ when
770        using pcre_exec().
772    11. While implementing 10, a number of bugs in the handling of groups were
773        discovered and fixed:
775        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
776        (a|)*(?1) gave a compile-time internal error.
777        ((a|)+)+  did not notice that the outer group could match an empty string.
778        (^a|^)+   was not marked as anchored.
779        (.*a|.*)+ was not marked as matching at start or after a newline.
781    12. Yet more internal refactoring has removed another argument from the match()
782        function. Special calls to this function are now indicated by setting a
783        value in a variable in the "match data" data block.
785    13. Be more explicit in pcre_study() instead of relying on "default" for
786        opcodes that mean there is no starting character; this means that when new
787        ones are added and accidentally left out of pcre_study(), testing should
788        pick them up.
790    14. The -s option of pcretest has been documented for ages as being an old
791        synonym of -m (show memory usage). I have changed it to mean "force study
792        for every regex", that is, assume /S for every regex. This is similar to -i
793        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
794        using it. It makes it easier to run collections of tests with and without
795        study enabled, and thereby test pcre_study() more easily. All the standard
796        tests are now run with and without -s (but some patterns can be marked as
797        "never study" - see 20 below).
799    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
800        restoration of the capturing data to the outer values was not happening
801        correctly.
803    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
804        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
805        pattern had matched an empty string, and so incorrectly returned a no
806        match.
808    17. There was optimizing code for the last branch of non-capturing parentheses,
809        and also for the obeyed branch of a conditional subexpression, which used
810        tail recursion to cut down on stack usage. Unfortunately, now that there is
811        the possibility of (*THEN) occurring in these branches, tail recursion is
812        no longer possible because the return has to be checked for (*THEN). These
813        two optimizations have therefore been removed. [But see 8.20/11 above.]
815    18. If a pattern containing \R was studied, it was assumed that \R always
816        matched two bytes, thus causing the minimum subject length to be
817        incorrectly computed because \R can also match just one byte.
819    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
820        was incorrectly computed.
822    20. If /S is present twice on a test pattern in pcretest input, it now
823        *disables* studying, thereby overriding the use of -s on the command line
824        (see 14 above). This is necessary for one or two tests to keep the output
825        identical in both cases.
827    21. When (*ACCEPT) was used in an assertion that matched an empty string and
828        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
830    22. When an atomic group that contained a capturing parenthesis was
831        successfully matched, but the branch in which it appeared failed, the
832        capturing was not being forgotten if a higher numbered group was later
833        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
834        group 1 to "a", when in fact it should be unset. This applied to multi-
835        branched capturing and non-capturing groups, repeated or not, and also to
836        positive assertions (capturing in negative assertions does not happen
837        in PCRE) and also to nested atomic groups.
839    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
840        subject after a captured substring, to make it easier to tell which of a
841        number of identical substrings has been captured.
843    24. The way atomic groups are processed by pcre_exec() has been changed so that
844        if they are repeated, backtracking one repetition now resets captured
845        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
846        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
847        "aaa". Previously, it would have been "a". As part of this code
848        refactoring, the way recursive calls are handled has also been changed.
850    25. If an assertion condition captured any substrings, they were not passed
851        back unless some other capturing happened later. For example, if
852        (?(?=(a))a) was matched against "a", no capturing was returned.
854    26. When studying a pattern that contained subroutine calls or assertions,
855        the code for finding the minimum length of a possible match was handling
856        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
857        group 1 called group 2 while simultaneously a separate group 2 called group
858        1). A stack overflow occurred in this case. I have fixed this by limiting
859        the recursion depth to 10.
861    27. Updated RunTest.bat in the distribution to the version supplied by Tom
862        Fortmann. This supports explicit test numbers on the command line, and has
863        argument validation and error reporting.
865    28. An instance of \X with an unlimited repeat could fail if at any point the
866        first character it looked at was a mark character.
868    29. Some minor code refactoring concerning Unicode properties and scripts
869        should reduce the stack requirement of match() slightly.
871    30. Added the '=' option to pcretest to check the setting of unused capturing
872        slots at the end of the pattern, which are documented as being -1, but are
873        not included in the return count.
875    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
876        compiled something random. Now it gives a compile-time error (as does
877        Perl).
879    32. A *MARK encountered during the processing of a positive assertion is now
880        recorded and passed back (compatible with Perl).
882    33. If --only-matching or --colour was set on a pcregrep call whose pattern
883        had alternative anchored branches, the search for a second match in a line
884        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
885        matched the line "0102" twice. The same bug affected patterns that started
886        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
887        twice.
889    34. Previously, PCRE did not allow quantification of assertions. However, Perl
890        does, and because of capturing effects, quantifying parenthesized
891        assertions may at times be useful. Quantifiers are now allowed for
892        parenthesized assertions.
894    35. A minor code tidy in pcre_compile() when checking options for \R usage.
896    36. \g was being checked for fancy things in a character class, when it should
897        just be a literal "g".
899    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
900        appearance of a nested POSIX class supersedes an apparent external class.
901        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
902        unescaped square brackets may also appear as part of class names. For
903        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
904        more like Perl. (But see 8.20/1 above.)
906    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
907        was because it thought it was \N{name}, which is not supported).
909    39. Add minix to OS list not supporting the -S option in pcretest.
911    40. PCRE tries to detect cases of infinite recursion at compile time, but it
912        cannot analyze patterns in sufficient detail to catch mutual recursions
913        such as ((?1))((?2)). There is now a runtime test that gives an error if a
914        subgroup is called recursively as a subpattern for a second time at the
915        same position in the subject string. In previous releases this might have
916        been caught by the recursion limit, or it might have run out of stack.
918    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
919        happen only once. PCRE was, however incorrectly giving a compile time error
920        "recursive call could loop indefinitely" because it cannot analyze the
921        pattern in sufficient detail. The compile time test no longer happens when
922        PCRE is compiling a conditional subpattern, but actual runaway loops are
923        now caught at runtime (see 40 above).
925    42. It seems that Perl allows any characters other than a closing parenthesis
926        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
927        has been changed to be the same.
929    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
930        as not to get warnings when autogen.sh is called. Also changed
931        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
933    44. To help people who use pcregrep to scan files containing exceedingly long
934        lines, the following changes have been made:
936        (a) The default value of the buffer size parameter has been increased from
937            8K to 20K. (The actual buffer used is three times this size.)
939        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
940            PCRE is built.
942        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
943            to be set at run time.
945        (d) Numerical values in pcregrep options can be followed by K or M, for
946            example --buffer-size=50K.
948        (e) If a line being scanned overflows pcregrep's buffer, an error is now
949            given and the return code is set to 2.
951    45. Add a pointer to the latest mark to the callout data block.
953    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
954        partial match of an empty string instead of no match. This was specific to
955        the use of ".".
957    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
958        complete match instead of a partial match. This bug was dependent on both
959        the PCRE_UTF8 and PCRE_DOTALL options being set.
961    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
962        starting byte set, because \b was not being ignored.
965    Version 8.12 15-Jan-2011
966    ------------------------
968    1.  Fixed some typos in the markup of the man pages, and wrote a script that
969        checks for such things as part of the documentation building process.
971    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
972        --match-limit and --recursion-limit options (added for 8.11). In
973        particular, this made one of the standard tests fail. (The integer value
974        went into the wrong half of a long int.)
976    3.  If the --colour option was given to pcregrep with -v (invert match), it
977        did strange things, either producing crazy output, or crashing. It should,
978        of course, ignore a request for colour when reporting lines that do not
979        match.
981    4.  Another pcregrep bug caused similar problems if --colour was specified with
982        -M (multiline) and the pattern match finished with a line ending.
984    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
985        matched in multiline mode, the following line was shown as part of the
986        match. This seems wrong, so I have changed it.
988    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
989        the check for further matches in the same line (so they could be coloured)
990        to overrun the end of the current line. If another match was found, it was
991        incorrectly shown (and then shown again when found in the next line).
993    7.  If pcregrep was compiled under Windows, there was a reference to the
994        function pcregrep_exit() before it was defined. I am assuming this was
995        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
996        reported by a user. I've moved the definition above the reference.
999    Version 8.11 10-Dec-2010
1000    ------------------------
1002    1.  (*THEN) was not working properly if there were untried alternatives prior
1003        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1004        backtracked to try for "b" instead of moving to the next alternative branch
1005        at the same level (in this case, to look for "c"). The Perl documentation
1006        is clear that when (*THEN) is backtracked onto, it goes to the "next
1007        alternative in the innermost enclosing group".
1009    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1010        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1011        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1012        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1013        (*THEN).
1015    3.  If \s appeared in a character class, it removed the VT character from
1016        the class, even if it had been included by some previous item, for example
1017        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1018        of \s, but is part of the POSIX "space" class.)
1020    4.  A partial match never returns an empty string (because you can always
1021        match an empty string at the end of the subject); however the checking for
1022        an empty string was starting at the "start of match" point. This has been
1023        changed to the "earliest inspected character" point, because the returned
1024        data for a partial match starts at this character. This means that, for
1025        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1026        (previously it gave "no match").
1028    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1029        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1030        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1031        has an implication that the given string is incomplete (because a partial
1032        match is preferred over a full match). For this reason, these items now
1033        give a partial match in this situation. [Aside: previously, the one case
1034        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1035        match rather than a full match, which was wrong by the old rules, but is
1036        now correct.]
1038    6.  There was a bug in the handling of #-introduced comments, recognized when
1039        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1040        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1041        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1042        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1043        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1044        places in pcre_compile().
1046    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1047        comments when looking ahead for named forward references to subpatterns,
1048        the only newline sequence it recognized was NL. It now handles newlines
1049        according to the set newline convention.
1051    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1052        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1053        cater for a lack of strerror(). These oversights have been fixed.
1055    9.  Added --match-limit and --recursion-limit to pcregrep.
1057    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1059    11. When the -o option was used, pcregrep was setting a return code of 1, even
1060        when matches were found, and --line-buffered was not being honoured.
1062    12. Added an optional parentheses number to the -o and --only-matching options
1063        of pcregrep.
1065    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1066        can match an empty string. The code to do it in pcretest and pcredemo
1067        needed fixing:
1069        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1070            only one byte after an empty string match just before CRLF (this case
1071            just got forgotten; "any" and "anycrlf" were OK).
1073        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1074            mode when an empty string match preceded an ASCII character followed by
1075            a non-ASCII character. (The code for advancing by one character rather
1076            than one byte was nonsense.)
1078        (c) The pcredemo.c sample program did not have any code at all to handle
1079            the cases when CRLF is a valid newline sequence.
1081    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1082        as a starting offset was within the subject string. There is now a new
1083        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1084        negative or greater than the length of the string. In order to test this,
1085        pcretest is extended to allow the setting of negative starting offsets.
1087    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1088        starting offset points to the beginning of a UTF-8 character was
1089        unnecessarily clumsy. I tidied it up.
1091    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1092        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1094    17. Nobody had reported that the --include_dir option, which was added in
1095        release 7.7 should have been called --include-dir (hyphen, not underscore)
1096        for compatibility with GNU grep. I have changed it to --include-dir, but
1097        left --include_dir as an undocumented synonym, and the same for
1098        --exclude-dir, though that is not available in GNU grep, at least as of
1099        release 2.5.4.
1101    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1102        characters from a string of bytes) have been redefined so as not to use
1103        loops, in order to improve performance in some environments. At the same
1104        time, I abstracted some of the common code into auxiliary macros to save
1105        repetition (this should not affect the compiled code).
1107    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1108        compile-time error is now given if \c is not followed by an ASCII
1109        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1110        different, and any byte value is allowed.)
1112    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1113        START_OPTIMIZE option, which is now allowed at compile time - but just
1114        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1115        to pcregrep and other applications that have no direct access to PCRE
1116        options. The new /Y option in pcretest sets this option when calling
1117        pcre_compile().
1119    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1120        back references. Groups containing recursive back references were forced to
1121        be atomic by that change, but in the case of named groups, the amount of
1122        memory required was incorrectly computed, leading to "Failed: internal
1123        error: code overflow". This has been fixed.
1125    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1126        pcretest.c, to avoid build problems in some Borland environments.
1129    Version 8.10 25-Jun-2010
1130    ------------------------
1132    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1133        THEN.
1135    2.  (*ACCEPT) was not working when inside an atomic group.
1137    3.  Inside a character class, \B is treated as a literal by default, but
1138        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1139        causes the error). The code is unchanged, but I tidied the documentation.
1141    4.  Inside a character class, PCRE always treated \R and \X as literals,
1142        whereas Perl faults them if its -w option is set. I have changed PCRE so
1143        that it faults them when PCRE_EXTRA is set.
1145    5.  Added support for \N, which always matches any character other than
1146        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1148    6.  When compiling pcregrep with newer versions of gcc which may have
1149        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1150        declared with attribute warn_unused_result" were given. Just casting the
1151        result to (void) does not stop the warnings; a more elaborate fudge is
1152        needed. I've used a macro to implement this.
1154    7.  Minor change to pcretest.c to avoid a compiler warning.
1156    8.  Added four artifical Unicode properties to help with an option to make
1157        \s etc use properties (see next item). The new properties are: Xan
1158        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1160    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1161        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1162        this option. Modified pcretest to add /W to test this facility. Added
1163        REG_UCP to make it available via the POSIX interface.
1165    10. Added --line-buffered to pcregrep.
1167    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1168        studied, and the match started with a letter with a code point greater than
1169        127 whose first byte was different to the first byte of the other case of
1170        the letter, the other case of this starting letter was not recognized
1171        (#976).
1173    12. If a pattern that was studied started with a repeated Unicode property
1174        test, for example, \p{Nd}+, there was the theoretical possibility of
1175        setting up an incorrect bitmap of starting bytes, but fortunately it could
1176        not have actually happened in practice until change 8 above was made (it
1177        added property types that matched character-matching opcodes).
1179    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1180        possible starting bytes for non-anchored patterns.
1182    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1183        \R, and also a number of cases that involve Unicode properties, both
1184        explicit and implicit when PCRE_UCP is set.
1186    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1187        input, it could crash or give wrong results if characters with values
1188        greater than 0xc0 were present in the subject string. (Detail: it assumed
1189        UTF-8 input when processing these items.)
1191    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1192        size_t is 64-bit (#991).
1194    17. Added a check for running out of memory when PCRE is compiled with
1195        --disable-stack-for-recursion (#990).
1197    18. If the last data line in a file for pcretest does not have a newline on
1198        the end, a newline was missing in the output.
1200    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1201        less than 128) in its various bitmaps. However, there is a facility for
1202        generating tables according to the current locale when PCRE is compiled. It
1203        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1204        characters, are recognized by isspace() and therefore were getting set in
1205        these tables, and indeed these tables seem to approximate to ISO 8859. This
1206        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1207        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1208        which of course cannot start UTF-8 characters. I have changed the code so
1209        that only real ASCII characters (less than 128) and the correct starting
1210        bytes for UTF-8 encodings are set for characters greater than 127 when in
1211        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1212        altogether.)
1214    20. Added the /T option to pcretest so as to be able to run tests with non-
1215        standard character tables, thus making it possible to include the tests
1216        used for 19 above in the standard set of tests.
1218    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1219        reference to a subpattern the other side of a comment that contains an
1220        opening parenthesis caused either an internal compiling error, or a
1221        reference to the wrong subpattern.
1224    Version 8.02 19-Mar-2010
1225    ------------------------
1227    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1229    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1230        configured.
1232    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1233        original author of that file, following a query about its status.
1235    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1236        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1238    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1239        quantifier applied to a forward-referencing subroutine call, could compile
1240        incorrect code or give the error "internal error: previously-checked
1241        referenced subpattern not found".
1243    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1244        variables to point to external functions. For these systems, therefore,
1245        pcre_malloc etc. are now initialized to local functions that call the
1246        relevant global functions.
1248    7.  There were two entries missing in the vectors called coptable and poptable
1249        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1250        I've fixed the data, and added a kludgy way of testing at compile time that
1251        the lengths are correct (equal to the number of opcodes).
1253    8.  Following on from 7, I added a similar kludge to check the length of the
1254        eint vector in pcreposix.c.
1256    9.  Error texts for pcre_compile() are held as one long string to avoid too
1257        much relocation at load time. To find a text, the string is searched,
1258        counting zeros. There was no check for running off the end of the string,
1259        which could happen if a new error number was added without updating the
1260        string.
1262    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1264    11. \K was not working if it appeared in an atomic group or in a group that
1265        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1266        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1267        the assertion is positive, but not if it is negative.
1269    12. Change 11 fortuitously reduced the size of the stack frame used in the
1270        "match()" function of pcre_exec.c by one pointer. Forthcoming
1271        implementation of support for (*MARK) will need an extra pointer on the
1272        stack; I have reserved it now, so that the stack frame size does not
1273        decrease.
1275    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1276        item in branch that calls a recursion is a subroutine call - as in the
1277        second branch in the above example - was incorrectly given the compile-
1278        time error "recursive call could loop indefinitely" because pcre_compile()
1279        was not correctly checking the subroutine for matching a non-empty string.
1281    14. The checks for overrunning compiling workspace could trigger after an
1282        overrun had occurred. This is a "should never occur" error, but it can be
1283        triggered by pathological patterns such as hundreds of nested parentheses.
1284        The checks now trigger 100 bytes before the end of the workspace.
1286    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1289    Version 8.01 19-Jan-2010
1290    ------------------------
1292    1.  If a pattern contained a conditional subpattern with only one branch (in
1293        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1294        computed the wrong minimum data length (which is of course zero for such
1295        subpatterns). This could cause incorrect "no match" results.
1297    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1298        the pattern is reset in the first branch, pcre_compile() failed with
1299        "internal error: code overflow at offset...". This happened only when
1300        the reset was to the original external option setting. (An optimization
1301        abstracts leading options settings into an external setting, which was the
1302        cause of this.)
1304    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1305        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1306        assertion pattern did not match (meaning that the assertion was true), it
1307        was incorrectly treated as false if the SKIP had been reached during the
1308        matching. This also applied to assertions used as conditions.
1310    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1311        assertion subpattern, including such a pattern used as a condition,
1312        unpredictable results occurred, instead of the error return
1315    5.  The C++ GlobalReplace function was not working like Perl for the special
1316        situation when an empty string is matched. It now does the fancy magic
1317        stuff that is necessary.
1319    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1320        removed. (These were left over from very, very early versions of PCRE.)
1322    7.  Some cosmetic changes to the code to make life easier when compiling it
1323        as part of something else:
1325        (a) Change DEBUG to PCRE_DEBUG.
1327        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1328            called "current" as "current_branch", to prevent a collision with the
1329            Linux macro when compiled as a kernel module.
1331        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1332            prevent a collision with the Linux macro when compiled as a kernel
1333            module.
1335    8.  In pcre_compile() there are some checks for integer overflows that used to
1336        cast potentially large values to (double). This has been changed to that
1337        when building, a check for int64_t is made, and if it is found, it is used
1338        instead, thus avoiding the use of floating point arithmetic. (There is no
1339        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1340        double.
1342    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1343        2005 (difference between two addresses compared to an unsigned value).
1345    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1346        custom one, because of the following reported problem in Windows:
1348          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1349              under Win32.
1350          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1351              therefore missing the function definition.
1352          - The compiler thus generates a "C" signature for the test function.
1353          - The linker fails to find the "C" function.
1354          - PCRE fails to configure if asked to do so against libbz2.
1356    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1357        messages were output:
1359          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1360          rerunning libtoolize, to keep the correct libtool macros in-tree.
1361          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1363        I have done both of these things.
1365    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1366        most of the time, it *can* run out if it is given a pattern that contains a
1367        runaway infinite recursion. I updated the discussion in the pcrestack man
1368        page.
1370    13. Now that we have gone to the x.xx style of version numbers, the minor
1371        version may start with zero. Using 08 or 09 is a bad idea because users
1372        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1373        interpreted as invalid octal numbers. I've updated the previous comment in
1374        configure.ac, and also added a check that gives an error if 08 or 09 are
1375        used.
1377    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1378        causing partial matching to fail when the end of the subject matched \W
1379        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1381    15. There were some discrepancies between the declarations in pcre_internal.h
1382        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1383        their definitions. The declarations used "const uschar *" and the
1384        definitions used USPTR. Even though USPTR is normally defined as "const
1385        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1386        reported that: "This difference in casting confuses some C++ compilers, for
1387        example, SunCC recognizes above declarations as different functions and
1388        generates broken code for hbpcre." I have changed the declarations to use
1389        USPTR.
1391    16. GNU libtool is named differently on some systems. The autogen.sh script now
1392        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1393        (FreeBSD).
1395    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1396        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1397        comment: "Figure out how to create a longlong from a string: strtoll and
1398        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1399        instance, but it only takes 2 args instead of 3!"
1401    18. A subtle bug concerned with back references has been fixed by a change of
1402        specification, with a corresponding code fix. A pattern such as
1403        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1404        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1405        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1406        same bug. Such groups have to be quantified to be useful, or contained
1407        inside another quantified group. (If there's no repetition, the reference
1408        can never match.) The problem arises because, having left the group and
1409        moved on to the rest of the pattern, a later failure that backtracks into
1410        the group uses the captured value from the final iteration of the group
1411        rather than the correct earlier one. I have fixed this in PCRE by forcing
1412        any group that contains a reference to itself to be an atomic group; that
1413        is, there cannot be any backtracking into it once it has completed. This is
1414        similar to recursive and subroutine calls.
1417    Version 8.00 19-Oct-09
1418    ----------------------
1420    1.  The table for translating pcre_compile() error codes into POSIX error codes
1421        was out-of-date, and there was no check on the pcre_compile() error code
1422        being within the table. This could lead to an OK return being given in
1423        error.
1425    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1426        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1427        in a Windows environment.
1429    3.  The pcregrep --count option prints the count for each file even when it is
1430        zero, as does GNU grep. However, pcregrep was also printing all files when
1431        --files-with-matches was added. Now, when both options are given, it prints
1432        counts only for those files that have at least one match. (GNU grep just
1433        prints the file name in this circumstance, but including the count seems
1434        more useful - otherwise, why use --count?) Also ensured that the
1435        combination -clh just lists non-zero counts, with no names.
1437    4.  The long form of the pcregrep -F option was incorrectly implemented as
1438        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1439        but it seems right to fix it, and I didn't think it was worth preserving
1440        the old behaviour.
1442    5.  The command line items --regex=pattern and --regexp=pattern were not
1443        recognized by pcregrep, which required --regex pattern or --regexp pattern
1444        (with a space rather than an '='). The man page documented the '=' forms,
1445        which are compatible with GNU grep; these now work.
1447    6.  No libpcreposix.pc file was created for pkg-config; there was just
1448        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1450    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1451        when UCP support is not needed, by modifying the Python script that
1452        generates it from Unicode data files. This should not matter if the module
1453        is correctly used as a library, but I received one complaint about 50K of
1454        unwanted data. My guess is that the person linked everything into his
1455        program rather than using a library. Anyway, it does no harm.
1457    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1458        was a minimum greater than 1 for a wide character in a possessive
1459        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1460        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1461        character. Chaos in the form of incorrect output or a compiling loop could
1462        result.
1464    9.  The restrictions on what a pattern can contain when partial matching is
1465        requested for pcre_exec() have been removed. All patterns can now be
1466        partially matched by this function. In addition, if there are at least two
1467        slots in the offset vector, the offset of the earliest inspected character
1468        for the match and the offset of the end of the subject are set in them when
1469        PCRE_ERROR_PARTIAL is returned.
1471    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1472        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1473        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1474        and may be more useful for multi-segment matching.
1476    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1477        used to be given if ever the end of the subject was reached; now it is
1478        given only if matching could not proceed because another character was
1479        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1480        string "Z", which now yields "no match" instead of "partial match". In the
1481        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1482        final character ended with (*FAIL).
1484    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1485        if the pattern had a "must contain" character that was already found in the
1486        earlier partial match, unless partial matching was again requested. For
1487        example, with the pattern /dog.(body)?/, the "must contain" character is
1488        "g". If the first part-match was for the string "dog", restarting with
1489        "sbody" failed. This bug has been fixed.
1491    13. The string returned by pcre_dfa_exec() after a partial match has been
1492        changed so that it starts at the first inspected character rather than the
1493        first character of the match. This makes a difference only if the pattern
1494        starts with a lookbehind assertion or \b or \B (\K is not supported by
1495        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1496        matching functions compatible, and I think it's the right thing to do.
1498    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1499        so that the demonstration program is easily available in environments where
1500        PCRE has not been installed from source.
1502    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1503        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1504        library.
1506    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1507        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1508        is not the first non-POSIX option to be added. Clearly some people find
1509        these options useful.
1511    17. If a caller to the POSIX matching function regexec() passes a non-zero
1512        value for nmatch with a NULL value for pmatch, the value of
1513        nmatch is forced to zero.
1515    18. RunGrepTest did not have a test for the availability of the -u option of
1516        the diff command, as RunTest does. It now checks in the same way as
1517        RunTest, and also checks for the -b option.
1519    19. If an odd number of negated classes containing just a single character
1520        interposed, within parentheses, between a forward reference to a named
1521        subpattern and the definition of the subpattern, compilation crashed with
1522        an internal error, complaining that it could not find the referenced
1523        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1524        [The bug was that it was starting one character too far in when skipping
1525        over the character class, thus treating the ] as data rather than
1526        terminating the class. This meant it could skip too much.]
1528    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1529        /g option in pcretest when the pattern contains \K, which makes it possible
1530        to have an empty string match not at the start, even when the pattern is
1531        anchored. Updated pcretest and pcredemo to use this option.
1533    21. If the maximum number of capturing subpatterns in a recursion was greater
1534        than the maximum at the outer level, the higher number was returned, but
1535        with unset values at the outer level. The correct (outer level) value is
1536        now given.
1538    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1539        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1540        make it do so. The string so far is captured, making this feature
1541        compatible with Perl.
1543    23. The tests have been re-organized, adding tests 11 and 12, to make it
1544        possible to check the Perl 5.10 features against Perl 5.10.
1546    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1547        pattern matches a fixed length string. PCRE did not allow this; now it
1548        does. Neither allows recursion.
1550    25. I finally figured out how to implement a request to provide the minimum
1551        length of subject string that was needed in order to match a given pattern.
1552        (It was back references and recursion that I had previously got hung up
1553        on.) This code has now been added to pcre_study(); it finds a lower bound
1554        to the length of subject needed. It is not necessarily the greatest lower
1555        bound, but using it to avoid searching strings that are too short does give
1556        some useful speed-ups. The value is available to calling programs via
1557        pcre_fullinfo().
1559    26. While implementing 25, I discovered to my embarrassment that pcretest had
1560        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1561        study optimizations had never been tested with that matching function.
1562        Oops. What is worse, even when it was passed study data, there was a bug in
1563        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1564        were also very few tests of studied patterns with pcre_dfa_exec().
1566    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1567        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1568        on the other side of the coin, they are no longer allowed to have different
1569        names, because these cannot be distinguished in PCRE, and this has caused
1570        confusion. (This is a difference from Perl.)
1572    28. When duplicate subpattern names are present (necessarily with different
1573        numbers, as required by 27 above), and a test is made by name in a
1574        conditional pattern, either for a subpattern having been matched, or for
1575        recursion in such a pattern, all the associated numbered subpatterns are
1576        tested, and the overall condition is true if the condition is true for any
1577        one of them. This is the way Perl works, and is also more like the way
1578        testing by number works.
1581    Version 7.9 11-Apr-09
1582    ---------------------
1584    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1585        (pcretest), all targets were linked against these libraries. This included
1586        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1587        libraries. This caused unwanted dependencies to be created. This problem
1588        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1589        pcretest is linked with readline.
1591    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1592        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1593        moved outside it again, because FALSE and TRUE are already defined in AIX,
1594        but BOOL is not.
1596    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1597        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1599    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1600        hyphens) following file names and line numbers when outputting matching
1601        lines. This is not true; no spaces are inserted. I have also clarified the
1602        wording for the --colour (or --color) option.
1604    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1605        was not coloured; this is different to GNU grep, so I have changed it to be
1606        the same.
1608    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1609        each matching line was coloured. Now it goes on to look for further matches
1610        of any of the test patterns, which is the same behaviour as GNU grep.
1612    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1613        doesn't make sense to accept an empty string match in pcregrep, so I have
1614        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1615        seems to be how GNU grep behaves.
1617    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1618        start or after a newline", because the conditional assertion was not being
1619        correctly handled. The rule now is that both the assertion and what follows
1620        in the first alternative must satisfy the test.
1622    9.  If auto-callout was enabled in a pattern with a conditional group whose
1623        condition was an assertion, PCRE could crash during matching, both with
1624        pcre_exec() and pcre_dfa_exec().
1626    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1627        used for matching.
1629    11. Unicode property support in character classes was not working for
1630        characters (bytes) greater than 127 when not in UTF-8 mode.
1632    12. Added the -M command line option to pcretest.
1634    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1636    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1638    16. Added comments and documentation about mis-use of no_arg in the C++
1639        wrapper.
1641    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1642        from Martin Jerabek that uses macro names for all relevant character and
1643        string constants.
1645    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1646        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1647        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1648        these, but not everybody uses configure.
1650    19. A conditional group that had only one branch was not being correctly
1651        recognized as an item that could match an empty string. This meant that an
1652        enclosing group might also not be so recognized, causing infinite looping
1653        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1654        with the subject "ab", where knowledge that the repeated group can match
1655        nothing is needed in order to break the loop.
1657    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1658        exec(), but without supplying a callout function, matching went wrong.
1660    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1661        leak if the size of the offset vector was greater than 30. When the vector
1662        is smaller, the saved offsets during recursion go onto a local stack
1663        vector, but for larger vectors malloc() is used. It was failing to free
1664        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1665        error, in fact).
1667    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1668        heapframe that is used only when UTF-8 support is enabled. This caused no
1669        problem, but was untidy.
1671    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1672        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1673        included within another project.
1675    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1676        slightly modified by me:
1678          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1679              not building pcregrep.
1681          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1682              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1684    25. Forward references, both numeric and by name, in patterns that made use of
1685        duplicate group numbers, could behave incorrectly or give incorrect errors,
1686        because when scanning forward to find the reference group, PCRE was not
1687        taking into account the duplicate group numbers. A pattern such as
1688        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1690    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1691        the feature of a custom pointer more persuasive (as requested by a user).
1693    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1694        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1695        pre-defined.
1697    28. Added support for (*UTF8) at the start of a pattern.
1699    29. Arrange for flags added by the "release type" setting in CMake to be shown
1700        in the configuration summary.
1703    Version 7.8 05-Sep-08
1704    ---------------------
1706    1.  Replaced UCP searching code with optimized version as implemented for Ad
1707        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1708        stage table and inline lookup instead of a function, giving speed ups of 2
1709        to 5 times on some simple patterns that I tested. Permission was given to
1710        distribute the MultiStage2.py script that generates the tables (it's not in
1711        the tarball, but is in the Subversion repository).
1713    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1714        scripts.
1716    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1717        a group with a zero qualifier. The result of the study could be incorrect,
1718        or the function might crash, depending on the pattern.
1720    4.  Caseless matching was not working for non-ASCII characters in back
1721        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1722        It now works when Unicode Property Support is available.
1724    5.  In pcretest, an escape such as \x{de} in the data was always generating
1725        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1726        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1727        truncation.
1729    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1731    7.  Added two (int) casts to pcregrep when printing the difference of two
1732        pointers, in case they are 64-bit values.
1734    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1735        test 2 if it fails.
1737    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1738        and a #define of that name to empty if it is not externally set. This is to
1739        allow users of MSVC to set it if necessary.
1741    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1742        the convenience functions in the pcre_get.c source file.
1744    11. An option change at the start of a pattern that had top-level alternatives
1745        could cause overwriting and/or a crash. This command provoked a crash in
1746        some environments:
1748          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1750        This potential security problem was recorded as CVE-2008-2371.
1752    12. For a pattern where the match had to start at the beginning or immediately
1753        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1754        pcre_dfa_exec() could read past the end of the passed subject if there was
1755        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1756        pcretest so that it places the subject at the end of its malloc-ed buffer.
1758    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1759        exec() might read past the end of the data buffer in UTF-8 mode.
1761    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1762        the data contained the byte 0x85 as part of a UTF-8 character within its
1763        first line. This applied both to normal and DFA matching.
1765    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1766        /^[^d]*?$/8 failed to match "abc".
1768    16. Added a missing copyright notice to pcrecpp_internal.h.
1770    17. Make it more clear in the documentation that values returned from
1771        pcre_exec() in ovector are byte offsets, not character counts.
1773    18. Tidied a few places to stop certain compilers from issuing warnings.
1775    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1776        supplied by Stefan Weber. I made a further small update for 7.8 because
1777        there is a change of source arrangements: the pcre_searchfuncs.c module is
1778        replaced by pcre_ucd.c.
1781    Version 7.7 07-May-08
1782    ---------------------
1784    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1785        a string to a long long, pretend we don't even have a long long." This is
1786        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1788    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1789        pre-7.6 versions, which defined a global no_arg variable instead of putting
1790        it in the RE class. (See also #8 below.)
1792    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1793        Lopes.
1795    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1797        (1) The include/exclude patterns were being applied to the whole pathnames
1798            of files, instead of just to the final components.
1800        (2) If there was more than one level of directory, the subdirectories were
1801            skipped unless they satisfied the include/exclude conditions. This is
1802            inconsistent with GNU grep (and could even be seen as contrary to the
1803            pcregrep specification - which I improved to make it absolutely clear).
1804            The action now is always to scan all levels of directory, and just
1805            apply the include/exclude patterns to regular files.
1807    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1808        --exclude_dir in the tests to avoid scanning .svn directories.
1810    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1811        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1812        doesn't support NULs in patterns.
1814    7.  Added some missing "const"s to declarations of static tables in
1815        pcre_compile.c and pcre_dfa_exec.c.
1817    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1818        caused by fix #2  above. (Subsequently also a second patch to fix the
1819        first patch. And a third patch - this was a messy problem.)
1821    9.  Applied Craig's patch to remove the use of push_back().
1823    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1824        matching function regexec().
1826    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1827        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1828        references. PCRE supports relative numbers with this syntax (I don't think
1829        Oniguruma does).
1831    12. Previously, a group with a zero repeat such as (...){0} was completely
1832        omitted from the compiled regex. However, this means that if the group
1833        was called as a subroutine from elsewhere in the pattern, things went wrong
1834        (an internal error was given). Such groups are now left in the compiled
1835        pattern, with a new opcode that causes them to be skipped at execution
1836        time.
1838    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1839        to the way PCRE behaves:
1841        (a) A lone ] character is dis-allowed (Perl treats it as data).
1843        (b) A back reference to an unmatched subpattern matches an empty string
1844            (Perl fails the current match path).
1846        (c) A data ] in a character class must be notated as \] because if the
1847            first data character in a class is ], it defines an empty class. (In
1848            Perl it is not possible to have an empty class.) The empty class []
1849            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1850            The negative empty class [^] matches any one character, independently
1851            of the DOTALL setting.
1853    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1854        non-existent subpattern following a character class starting with ']' and
1855        containing () gave an internal compiling error instead of "reference to
1856        non-existent subpattern". Fortunately, when the pattern did exist, the
1857        compiled code was correct. (When scanning forwards to check for the
1858        existencd of the subpattern, it was treating the data ']' as terminating
1859        the class, so got the count wrong. When actually compiling, the reference
1860        was subsequently set up correctly.)
1862    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1863        it was being rejected as not supported by pcre_dfa_exec(), even though
1864        other assertions are supported. I have made pcre_dfa_exec() support
1865        (*FAIL).
1867    16. The implementation of 13c above involved the invention of a new opcode,
1868        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1869        cannot be changed at match time, I realized I could make a small
1870        improvement to matching performance by compiling OP_ALLANY instead of
1871        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1872        on the OP_ANY path.
1874    17. Compiling pcretest on Windows with readline support failed without the
1875        following two fixes: (1) Make the unistd.h include conditional on
1876        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1878    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1879        ncurses library to be included for pcretest when ReadLine support is
1880        requested, but also to allow for it to be overridden. This patch came from
1881        Daniel Bergström.
1883    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1884        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1885        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1886        spotting this.
1889    Version 7.6 28-Jan-08
1890    ---------------------
1892    1.  A character class containing a very large number of characters with
1893        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1894        overflow.
1896    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1897        HAVE_LONG_LONG is not defined.
1899    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1900        bring them up to date and include new features. This patch includes:
1902        - Fixed PH's badly added libz and libbz2 support.
1903        - Fixed a problem with static linking.
1904        - Added pcredemo. [But later removed - see 7 below.]
1905        - Fixed dftables problem and added an option.
1906        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1907            HAVE_LONG_LONG.
1908        - Added readline support for pcretest.
1909        - Added an listing of the option settings after cmake has run.
1911    4.  A user submitted a patch to Makefile that makes it easy to create
1912        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1913        Makefile.am that cause it to include this special target, without
1914        affecting anything else. Note that the same mingw target plus all
1915        the other distribution libraries and programs are now supported
1916        when configuring with CMake (see 6 below) instead of with
1917        Configure/Make.
1919    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1920        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1921        exported in the Windows port". It has not yet been confirmed that the patch
1922        solves the problem, but it does no harm.
1924    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1925        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1926        with CMake, and also correct the comment about stack recursion.
1928    7.  Remove the automatic building of pcredemo from the ./configure system and
1929        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1930        of a program that users should build themselves after PCRE is installed, so
1931        building it automatically is not really right. What is more, it gave
1932        trouble in some build environments.
1934    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1937    Version 7.5 10-Jan-08
1938    ---------------------
1940    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1941        values in parens when parsing an RE using the C++ wrapper."
1943    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1944        Characters greater than 255 were excluded from the class instead of being
1945        included.
1947    3.  The same bug as (2) above applied to negated POSIX classes such as
1948        [:^space:].
1950    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1951        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1952        I have changed it.
1954    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1955        first named subpattern) and a construct such as (?&a) would reference the
1956        first named subpattern whose name started with "a" (in other words, the
1957        length check was missing). Both these problems are fixed. "Subpattern name
1958        expected" is now given for (?&) (a zero-length name), and this patch also
1959        makes it give the same error for \k'' (previously it complained that that
1960        was a reference to a non-existent subpattern).
1962    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1963        this is right because (?- can be followed by option settings as well as by
1964        digits. I have, however, made the messages clearer.
1966    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1967        than the number used in the conditional) now cause a compile-time error.
1968        This is actually not compatible with Perl, which accepts such patterns, but
1969        treats the conditional as always being FALSE (as PCRE used to), but it
1970        seems to me that giving a diagnostic is better.
1972    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1973        and messages.
1975    9.  Fix two occurrences of "backslash" in comments that should have been
1976        "backspace".
1978    10. Remove two redundant lines of code that can never be obeyed (their function
1979        was moved elsewhere).
1981    11. The program that makes PCRE's Unicode character property table had a bug
1982        which caused it to generate incorrect table entries for sequences of
1983        characters that have the same character type, but are in different scripts.
1984        It amalgamated them into a single range, with the script of the first of
1985        them. In other words, some characters were in the wrong script. There were
1986        thirteen such cases, affecting characters in the following ranges:
1988          U+002b0 - U+002c1
1989          U+0060c - U+0060d
1990          U+0061e - U+00612
1991          U+0064b - U+0065e
1992          U+0074d - U+0076d
1993          U+01800 - U+01805
1994          U+01d00 - U+01d77
1995          U+01d9b - U+01dbf
1996          U+0200b - U+0200f
1997          U+030fc - U+030fe
1998          U+03260 - U+0327f
1999          U+0fb46 - U+0fbb1
2000          U+10450 - U+1049d
2002    12. The -o option (show only the matching part of a line) for pcregrep was not
2003        compatible with GNU grep in that, if there was more than one match in a
2004        line, it showed only the first of them. It now behaves in the same way as
2005        GNU grep.
2007    13. If the -o and -v options were combined for pcregrep, it printed a blank
2008        line for every non-matching line. GNU grep prints nothing, and pcregrep now
2009        does the same. The return code can be used to tell if there were any
2010        non-matching lines.
2012    14. Added --file-offsets and --line-offsets to pcregrep.
2014    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
2015        infinitely looping recursion. The bug was that positive lookaheads were not
2016        being skipped when checking for a possible empty match (negative lookaheads
2017        and both kinds of lookbehind were skipped).
2019    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
2020        inclusion of <windows.h> to before rather than after the definition of
2021        INVALID_FILE_ATTRIBUTES (patch from David Byron).
2023    17. Specifying a possessive quantifier with a specific limit for a Unicode
2024        character property caused pcre_compile() to compile bad code, which led at
2025        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
2026        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
2027        caused the error; without that there was no problem.
2029    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
2031    19. Added --enable-pcretest-libreadline.
2033    20. In pcrecpp.cc, the variable 'count' was incremented twice in
2034        RE::GlobalReplace(). As a result, the number of replacements returned was
2035        double what it should be. I removed one of the increments, but Craig sent a
2036        later patch that removed the other one (the right fix) and added unit tests
2037        that check the return values (which was not done before).
2039    21. Several CMake things:
2041        (1) Arranged that, when cmake is used on Unix, the libraries end up with
2042            the names libpcre and libpcreposix, not just pcre and pcreposix.
2044        (2) The above change means that pcretest and pcregrep are now correctly
2045            linked with the newly-built libraries, not previously installed ones.
2049    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
2050        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
2051        UTF-8 newline character). The key issue is that the pattern starts .*;
2052        this means that the match must be either at the beginning, or after a
2053        newline. The bug was in the code for advancing after a failed match and
2054        checking that the new position followed a newline. It was not taking
2055        account of UTF-8 characters correctly.
2057    23. PCRE was behaving differently from Perl in the way it recognized POSIX
2058        character classes. PCRE was not treating the sequence [:...:] as a
2059        character class unless the ... were all letters. Perl, however, seems to
2060        allow any characters between [: and :], though of course it rejects as
2061        unknown any "names" that contain non-letters, because all the known class
2062        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
2063        for example, whereas PCRE did not - it did not recognize a POSIX character
2064        class. This seemed a bit dangerous, so the code has been changed to be
2065        closer to Perl. The behaviour is not identical to Perl, because PCRE will
2066        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
2067        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
2068        Perl does, and where it didn't before.
2070    24. Rewrite so as to remove the single use of %n from pcregrep because in some
2071        Windows environments %n is disabled by default.
2074    Version 7.4 21-Sep-07
2075    ---------------------
2077    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
2078        means that a class such as [\s] counted as "explicit reference to CR or
2079        LF". That isn't really right - the whole point of the change was to try to
2080        help when there was an actual mention of one of the two characters. So now
2081        the change happens only if \r or \n (or a literal CR or LF) character is
2082        encountered.
2084    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
2085        of both had grown to the point where there were only 3 bits left.
2086        Fortunately, there was spare space in the data structure, and so I have
2087        moved the internal flags into a new 16-bit field to free up more option
2088        bits.
2090    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
2091        but did not set the internal JCHANGED flag - either of these is enough to
2092        control the way the "get" function works - but the PCRE_INFO_JCHANGED
2093        facility is supposed to tell if (?J) was ever used, so now (?J) at the
2094        start sets both bits.
2096    4.  Added options (at build time, compile time, exec time) to change \R from
2097        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
2099    5.  doc/pcresyntax.html was missing from the distribution.
2101    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
2102        compatibility, even though it is no longer used.
2104    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
2105        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
2106        windows.h file is present (where different names are used). [This was
2107        reversed later after testing - see 16 below.]
2109    8.  Changed all #include <config.h> to #include "config.h". There were also
2110        some further <pcre.h> cases that I changed to "pcre.h".
2112    9.  When pcregrep was used with the --colour option, it missed the line ending
2113        sequence off the lines that it output.
2115    10. It was pointed out to me that arrays of string pointers cause lots of
2116        relocations when a shared library is dynamically loaded. A technique of
2117        using a single long string with a table of offsets can drastically reduce
2118        these. I have refactored PCRE in four places to do this. The result is
2119        dramatic:
2121          Originally:                          290
2122          After changing UCP table:            187
2123          After changing error message table:   43
2124          After changing table of "verbs"       36
2125          After changing table of Posix names   22
2127        Thanks to the folks working on Gregex for glib for this insight.
2129    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
2130        unicode-properties was also set.
2132    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
2134    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
2135        checked only for CRLF.
2137    14. Added casts to pcretest.c to avoid compiler warnings.
2139    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
2141    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
2142        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2143        entirely. This removes changes made in 7 above.
2145    17. The CMake files have been updated, and there is now more information about
2146        building with CMake in the NON-UNIX-USE document.
2149    Version 7.3 28-Aug-07
2150    ---------------------
2152     1. In the rejigging of the build system that eventually resulted in 7.1, the
2153        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
2154        brackets there is not right, since it causes compilers to look for an
2155        installed pcre.h, not the version that is in the source that is being
2156        compiled (which of course may be different). I have changed it back to:
2158          #include "pcre.h"
2160        I have a vague recollection that the change was concerned with compiling in
2161        different directories, but in the new build system, that is taken care of
2162        by the VPATH setting the Makefile.
2164     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
2165        when the subject happened to end in the byte 0x85 (e.g. if the last
2166        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
2167        characters but of course it shouldn't be taken as a newline when it is part
2168        of another character. The bug was that, for an unlimited repeat of . in
2169        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
2170        characters when looking for a newline.
2172     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
2174     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
2175        in debug output.
2177     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
2178        long printing in the pcrecpp unittest when running under MinGW.
2180     6. ESC_K was left out of the EBCDIC table.
2182     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
2183        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
2184        limit also applies to "virtual nesting" when a pattern is recursive, and in
2185        this case 1000 isn't so big. I have been able to remove this limit at the
2186        expense of backing off one optimization in certain circumstances. Normally,
2187        when pcre_exec() would call its internal match() function recursively and
2188        immediately return the result unconditionally, it uses a "tail recursion"
2189        feature to save stack. However, when a subpattern that can match an empty
2190        string has an unlimited repetition quantifier, it no longer makes this
2191        optimization. That gives it a stack frame in which to save the data for
2192        checking that an empty string has been matched. Previously this was taken
2193        from the 1000-entry workspace that had been reserved. So now there is no
2194        explicit limit, but more stack is used.
2196     8. Applied Daniel's patches to solve problems with the import/export magic
2197        syntax that is required for Windows, and which was going wrong for the
2198        pcreposix and pcrecpp parts of the library. These were overlooked when this
2199        problem was solved for the main library.
2201     9. There were some crude static tests to avoid integer overflow when computing
2202        the size of patterns that contain repeated groups with explicit upper
2203        limits. As the maximum quantifier is 65535, the maximum group length was
2204        set at 30,000 so that the product of these two numbers did not overflow a
2205        32-bit integer. However, it turns out that people want to use groups that
2206        are longer than 30,000 bytes (though not repeat them that many times).
2207        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
2208        made it possible to implement the integer overflow checks in a much more
2209        dynamic way, which I have now done. The artificial limitation on group
2210        length has been removed - we now have only the limit on the total length of
2211        the compiled pattern, which depends on the LINK_SIZE setting.
2213    10. Fixed a bug in the documentation for get/copy named substring when
2214        duplicate names are permitted. If none of the named substrings are set, the
2215        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
2216        empty string.
2218    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
2219        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
2220        because the ] is interpreted as the first data character and the
2221        terminating ] is not found. PCRE has been made compatible with Perl in this
2222        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
2223        cause memory overwriting.
2225    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
2226        string has been matched (to stop an infinite loop). It was not recognizing
2227        a conditional subpattern that could match an empty string if that
2228        subpattern was within another subpattern. For example, it looped when
2229        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
2230        condition was not nested. This bug has been fixed.
2232    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
2233        past the start of the subject in the presence of bytes with the top bit
2234        set, for example "\x8aBCD".
2236    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
2237        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
2239    14. Optimized (?!) to (*FAIL).
2241    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
2242        This restricts code points to be within the range 0 to 0x10FFFF, excluding
2243        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
2244        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
2245        does: it's just the validity check that is more restrictive.
2247    16. Inserted checks for integer overflows during escape sequence (backslash)
2248        processing, and also fixed erroneous offset values for syntax errors during
2249        backslash processing.
2251    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
2252        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
2254    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
2255        caused an overrun.
2257    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
2258        something other than just ASCII characters) inside a group that had an
2259        unlimited repeat caused a loop at compile time (while checking to see
2260        whether the group could match an empty string).
2262    20. Debugging a pattern containing \p or \P could cause a crash. For example,
2263        [\P{Any}] did so. (Error in the code for printing property names.)
2265    21. An orphan \E inside a character class could cause a crash.
2267    22. A repeated capturing bracket such as (A)? could cause a wild memory
2268        reference during compilation.
2270    23. There are several functions in pcre_compile() that scan along a compiled
2271        expression for various reasons (e.g. to see if it's fixed length for look
2272        behind). There were bugs in these functions when a repeated \p or \P was
2273        present in the pattern. These operators have additional parameters compared
2274        with \d, etc, and these were not being taken into account when moving along
2275        the compiled data. Specifically:
2277        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2278            length.
2280        (b) An item such as \pL+ within a repeated group could cause crashes or
2281            loops.
2283        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2284            "reference to non-existent subpattern" error.
2286        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2288    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2289        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2291    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2293    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2294        character were causing crashes (broken optimization).
2296    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2297        \p or \P) caused a compile-time loop.
2299    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2300        break. For example, the unstudied pattern [\r\n]A does not match the string
2301        "\r\nA" because change 7.0/46 below moves the current point on by two
2302        characters after failing to match at the start. However, the pattern \nA
2303        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2304        the same is true. There doesn't seem any very clean way out of this, but
2305        what I have chosen to do makes the common cases work: PCRE now takes note
2306        of whether there can be an explicit match for \r or \n anywhere in the
2307        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2308        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2309        pattern has explicit CR or LF references.
2311    29. Added (*CR) etc for changing newline setting at start of pattern.
2314    Version 7.2 19-Jun-07
2315  ---------------------  ---------------------
2317   1. If the fr_FR locale cannot be found for test 3, try the "french" locale,   1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
# Line 21  Version 7.2 01-May-07 Line 2331  Version 7.2 01-May-07
2331      stack recursion. This gives a massive performance boost under BSD, but just      stack recursion. This gives a massive performance boost under BSD, but just
2332      a small improvement under Linux. However, it saves one field in the frame      a small improvement under Linux. However, it saves one field in the frame
2333      in all cases.      in all cases.
2335   6. Added more features from the forthcoming Perl 5.10:   6. Added more features from the forthcoming Perl 5.10:
2337      (a) (?-n) (where n is a string of digits) is a relative subroutine or      (a) (?-n) (where n is a string of digits) is a relative subroutine or
2338          recursion call. It refers to the nth most recently opened parentheses.          recursion call. It refers to the nth most recently opened parentheses.
2340      (b) (?+n) is also a relative subroutine call; it refers to the nth next      (b) (?+n) is also a relative subroutine call; it refers to the nth next
2341          to be opened parentheses.          to be opened parentheses.
2343      (c) Conditions that refer to capturing parentheses can be specified      (c) Conditions that refer to capturing parentheses can be specified
2344          relatively, for example, (?(-2)... or (?(+3)...          relatively, for example, (?(-2)... or (?(+3)...
2346        (d) \K resets the start of the current match so that everything before
2347            is not part of it.
2349        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2351        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2352            reference syntax.
2354        (g) (?| introduces a group in which the numbering of parentheses in each
2355            alternative starts with the same number.
2357        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2359     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2362     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2363        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2364        for detecting groups that can match an empty string.
2366     9. A pattern with a very large number of alternatives (more than several
2367        hundred) was running out of internal workspace during the pre-compile
2368        phase, where pcre_compile() figures out how much memory will be needed. A
2369        bit of new cunning has reduced the workspace needed for groups with
2370        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2371        workspace instead of running out of the 4096 that are available.
2373    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2375    11. Applied patch from Google to remove an optimization that didn't quite work.
2376        The report of the bug said:
2378          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2379          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2380          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2382    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2383        it matched the wrong number of bytes.
2386  Version 7.1 24-Apr-07  Version 7.1 24-Apr-07
2387  ---------------------  ---------------------

Removed from v.167  
changed lines
  Added in v.1274

  ViewVC Help
Powered by ViewVC 1.1.5