ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 91 by nigel, Sat Feb 24 21:41:34 2007 UTC revision 1050 by chpe, Sun Sep 30 18:20:10 2012 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.32
5    ------------
7    1.  Improved JIT compiler optimizations for first character search and single
8        character iterators.
10    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
11        Patch by Daniel Richard G.
13    3.  Single character iterator optimizations in the JIT compiler.
15    4.  Improved JIT compiler optimizations for character ranges.
17    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
18        Reported by Giuseppe D'Angelo.
20    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
21        of a line, was being set incorrectly in cases where .* appeared inside
22        atomic brackets at the start of a pattern, or where there was a subsequent
23        *PRUNE or *SKIP.
25    7.  Improved instruction cache flush for POWER/PowerPC.
26        Patch by Daniel Richard G.
28    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
29        grep:
31        (a) There is now no limit to the number of patterns to be matched.
33        (b) An error is given if a pattern is too long.
35        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
36            are now supported.
38        (d) --exclude-from and --include-from (multiple use) have been added.
40        (e) Exclusions and inclusions now apply to all files and directories, not
41            just to those obtained from scanning a directory recursively.
43        (f) Multiple uses of -f and --file-list are now supported.
45        (g) In a Windows environment, the default for -d has been changed from
46            "read" (the GNU grep default) to "skip", because otherwise the presence
47            of a directory in the file list provokes an error.
49        (h) The documentation has been revised and clarified in places.
51    9.  Improve the matching speed of capturing brackets.
53    10. Changed the meaning of \X so that it now matches a Unicode extended
54        grapheme cluster.
56    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
57        out POSIX threads when JIT support is configured.
59    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
61    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
62        structure could go wrong in environments where size_t is not the same size
63        as int.
65    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
66        set.
68    15. The EBCDIC support had decayed; later updates to the code had included
69        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
70        general tidy up of EBCDIC-related issues, and the documentation was also
71        not quite right. There is now a test that can be run on ASCII systems to
72        check some of the EBCDIC-related things (but is it not a full test).
74    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
75        in a small tidy to the code.
77    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
79    18. If the --only-matching (-o) option in pcregrep is specified multiple
80        times, each one causes appropriate output. For example, -o1 -o2 outputs the
81        substrings matched by the 1st and 2nd capturing parentheses. A separating
82        string can be specified by --om-separator (default empty).
84    19. Improving the first n character searches.
86    20. Turn case lists for horizontal and vertical white space into macros so that
87        they are defined only once.
89    21. This set of changes together give more compatible Unicode case-folding
90        behaviour for characters that have more than one other case when UCP
91        support is available.
93        (a) The Unicode property table now has offsets into a new table of sets of
94            three or more characters that are case-equivalent. The MultiStage2.py
95            script that generates these tables (the pcre_ucd.c file) now scans
96            CaseFolding.txt instead of UnicodeData.txt for character case
97            information.
99        (b) The code for adding characters or ranges of characters to a character
100            class has been abstracted into a generalized function that also handles
101            case-independence. In UTF-mode with UCP support, this uses the new data
102            to handle characters with more than one other case.
104        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
105            whose other case is greater than 256 are now correctly matched
106            caselessly. Previously, the high codepoint matched the low one, but not
107            vice versa.
109        (d) The processing of \h, \H, \v, and \ in character classes now makes use
110            of the new class addition function, using character lists defined as
111            macros alongside the case definitions of 20 above.
113        (e) Caseless back references now work with characters that have more than
114            one other case.
116        (f) General caseless matching of characters with more than one other case
117            is supported.
119    22. Unicode character properties were updated from Unicode 6.2.0
121    Version 8.31 06-July-2012
122    -------------------------
124    1.  Fixing a wrong JIT test case and some compiler warnings.
126    2.  Removed a bashism from the RunTest script.
128    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
129        to unsigned type, result still unsigned" that was given by an MS compiler
130        on encountering the code "-sizeof(xxx)".
132    4.  Partial matching support is added to the JIT compiler.
134    5.  Fixed several bugs concerned with partial matching of items that consist
135        of more than one character:
137        (a) /^(..)\1/ did not partially match "aba" because checking references was
138            done on an "all or nothing" basis. This also applied to repeated
139            references.
141        (b) \R did not give a hard partial match if \r was found at the end of the
142            subject.
144        (c) \X did not give a hard partial match after matching one or more
145            characters at the end of the subject.
147        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
148            a partial match for the string "\r".
150        (e) When newline was set to CRLF, the metacharacter "." did not recognize
151            a partial match for a CR character at the end of the subject string.
153    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
154        running pcretest, the text "(JIT)" added to the output whenever JIT is
155        actually used to run the match.
157    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
158        or /S+[+] with a digit between 1 and 7.
160    8.  OP_NOT now supports any UTF character not just single-byte ones.
162    9.  (*MARK) control verb is now supported by the JIT compiler.
164    10. The command "./RunTest list" lists the available tests without actually
165        running any of them. (Because I keep forgetting what they all are.)
169    12. Applied a (slightly modified) user-supplied patch that improves performance
170        when the heap is used for recursion (compiled with --disable-stack-for-
171        recursion). Instead of malloc and free for each heap frame each time a
172        logical recursion happens, frames are retained on a chain and re-used where
173        possible. This sometimes gives as much as 30% improvement.
175    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
176        call.
178    14. As documented, (*COMMIT) is now confined to within a positive assertion.
180    15. It is now possible to link pcretest with libedit as an alternative to
181        libreadline.
183    16. (*COMMIT) control verb is now supported by the JIT compiler.
185    17. The Unicode data tables have been updated to Unicode 6.1.0.
187    18. Added --file-list option to pcregrep.
189    19. Added binary file support to pcregrep, including the -a, --binary-files,
190        -I, and --text options.
192    20. The madvise function is renamed for posix_madvise for QNX compatibility
193        reasons. Fixed by Giuseppe D'Angelo.
195    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
197    22. Changed the option for creating symbolic links for 16-bit man pages from
198        -s to -sf so that re-installing does not cause issues.
200    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
202    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
203        matches in certain environments (the workspace was not being correctly
204        retained). Also added to pcre_dfa_exec() a simple plausibility check on
205        some of the workspace data at the beginning of a restart.
207    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
208        was not doing so when it should - probably a typo introduced by SVN 528
209        (change 8.10/14).
211    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
212        \w+ when the character tables indicated that \x{c4} was a word character.
213        There were several related cases, all because the tests for doing a table
214        lookup were testing for characters less than 127 instead of 255.
216    27. If a pattern contains capturing parentheses that are not used in a match,
217        their slots in the ovector are set to -1. For those that are higher than
218        any matched groups, this happens at the end of processing. In the case when
219        there were back references that the ovector was too small to contain
220        (causing temporary malloc'd memory to be used during matching), and the
221        highest capturing number was not used, memory off the end of the ovector
222        was incorrectly being set to -1. (It was using the size of the temporary
223        memory instead of the true size.)
225    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
226        ovector size, it uses memory at the end of the block that it has got.
228    29. Check for an overlong MARK name and give an error at compile time. The
229        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
231    30. JIT compiler update.
233    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
234        Rill for the patch.
236    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
238    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
240    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
241        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
242        was enabled.
244    35. Improve JIT code generation for greedy plus quantifier.
246    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
247        1 to "aa" instead of to an empty string. The bug affected repeated groups
248        that could potentially match an empty string.
250    37. Optimizing single character iterators in JIT.
252    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
253        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
254        codepoints that are too big for the mode are faulted, and in a UTF mode,
255        disallowed codepoints are also faulted.
257    39. If PCRE was compiled with UTF support, in three places in the DFA
258        matcher there was code that should only have been obeyed in UTF mode, but
259        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
260        processing when bytes with values greater than 127 were present. In 16-bit
261        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
262        both cases the values are those that cannot be the first data item in a UTF
263        character. The three items that might have provoked this were recursions,
264        possessively repeated groups, and atomic groups.
266    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
267        and pcregrep, because some OS require shared objects to be explicitly
268        passed to ld, causing the link step to fail if they are not.
270    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
271        mode, patterns that started with \h* or \R* might be incorrectly matched.
274    Version 8.30 04-February-2012
275    -----------------------------
277    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
278        name is defined in ctype.h.
280    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
281        only in quite long subpatterns.
283    3.  Removed the function pcre_info(), which has been obsolete and deprecated
284        since it was replaced by pcre_fullinfo() in February 2000.
286    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
287        match a (*MARK), and the match failed at the start of the subject, a
288        reference to memory before the start of the subject could occur. This bug
289        was introduced by fix 17 of release 8.21.
291    5.  A reference to an unset group with zero minimum repetition was giving
292        totally wrong answers (in non-JavaScript-compatibility mode). For example,
293        /(another)?(\1?)test/ matched against "hello world test". This bug was
294        introduced in release 8.13.
296    6.  Add support for 16-bit character strings (a large amount of work involving
297        many changes and refactorings).
299    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
300        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
301        from a file.
303    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
304        rounding is not applied in this particular case).
306    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
307        if they appear, or are escaped, in patterns.
309    10. Get rid of a number of -Wunused-but-set-variable warnings.
311    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
312        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
313        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
314        also returns the mark "x". This bug applied to capturing parentheses,
315        non-capturing parentheses, and atomic parentheses. It also applied to some
316        assertions.
318    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
319        information out of configure.ac instead of relying on pcre.h.generic, which
320        is not stored in the repository.
322    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
323        -lreadline.
325    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
327    15. Applied Graycode's patch to put the top-level frame on the stack rather
328        than the heap when not using the stack for recursion. This gives a
329        performance improvement in many cases when recursion is not deep.
331    16. Experimental code added to "pcretest -C" to output the stack frame size.
334    Version 8.21 12-Dec-2011
335    ------------------------
337    1.  Updating the JIT compiler.
339    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
340        are added as well.
342    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
343        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
344        calling _pcre_jit_exec. Some extra comments are added.
346    4.  (*MARK) settings inside atomic groups that do not contain any capturing
347        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
348        was introduced by change 18 for 8.20.
350    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
351        ECMA-262 standard.
353    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
354        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
355        This bug was probably introduced by change 9 of 8.13.
357    7.  While fixing 6 above, I noticed that a number of other items were being
358        incorrectly rejected as "not fixed length". This arose partly because newer
359        opcodes had not been added to the fixed-length checking code. I have (a)
360        corrected the bug and added tests for these items, and (b) arranged for an
361        error to occur if an unknown opcode is encountered while checking for fixed
362        length instead of just assuming "not fixed length". The items that were
363        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
364        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
365        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
367    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
368        being incorrectly compiled and would have given unpredicatble results.
370    9.  A possessively repeated subpattern with minimum repeat count greater than
371        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
372        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
373        the first (A) could occur when it should not.
375    10. Add a cast and remove a redundant test from the code.
377    11. JIT should use pcre_malloc/pcre_free for allocation.
379    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
380        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
381        is anything other than /usr, -L is still shown).
383    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
385    14. Perl does not support \N without a following name in a [] class; PCRE now
386        also gives an error.
388    15. If a forward reference was repeated with an upper limit of around 2000,
389        it caused the error "internal error: overran compiling workspace". The
390        maximum number of forward references (including repeats) was limited by the
391        internal workspace, and dependent on the LINK_SIZE. The code has been
392        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
393        the default depends on LINK_SIZE. There is a new upper limit (for safety)
394        of around 200,000 forward references. While doing this, I also speeded up
395        the filling in of repeated forward references.
397    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
398        incorrectly expecting the subject to contain another "a" after the start.
400    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
401        in the match, the SKIP should be ignored. This was not happening; instead
402        the SKIP was being treated as NOMATCH. For patterns such as
403        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
404        tested.
406    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
407        now much more compatible with Perl, in particular in cases where the result
408        is a non-match for a non-anchored pattern. For example, if
409        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
410        "m", where previously it did not return a name. A side effect of this
411        change is that for partial matches, the last encountered mark name is
412        returned, as for non matches. A number of tests that were previously not
413        Perl-compatible have been moved into the Perl-compatible test files. The
414        refactoring has had the pleasing side effect of removing one argument from
415        the match() function, thus reducing its stack requirements.
417    19. If the /S+ option was used in pcretest to study a pattern using JIT,
418        subsequent uses of /S (without +) incorrectly behaved like /S+.
420    21. Retrieve executable code size support for the JIT compiler and fixing
421        some warnings.
423    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
424        not work when the shorter character appeared right at the end of the
425        subject string.
427    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
428        systems.
430    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
431        output it when the /M option is used in pcretest.
433    25. The CheckMan script was not being included in the distribution. Also, added
434        an explicit "perl" to run Perl scripts from the PrepareRelease script
435        because this is reportedly needed in Windows.
437    26. If study data was being save in a file and studying had not found a set of
438        "starts with" bytes for the pattern, the data written to the file (though
439        never used) was taken from uninitialized memory and so caused valgrind to
440        complain.
442    27. Updated RunTest.bat as provided by Sheri Pierce.
444    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
446    29. Computation of memory usage for the table of capturing group names was
447        giving an unnecessarily large value.
450    Version 8.20 21-Oct-2011
451    ------------------------
453    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
454        a POSIX class. After further experiments with Perl, which convinced me that
455        Perl has bugs and confusions, a closing square bracket is no longer allowed
456        in a POSIX name. This bug also affected patterns with classes that started
457        with full stops.
459    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
460        captured substring, but while checking the failing first alternative,
461        substring 1 is temporarily captured. If the output vector supplied to
462        pcre_exec() was not big enough for this capture, the yield of the function
463        was still zero ("insufficient space for captured substrings"). This cannot
464        be totally fixed without adding another stack variable, which seems a lot
465        of expense for a edge case. However, I have improved the situation in cases
466        such as /(a)(b)x|abc/ matched against "abc", where the return code
467        indicates that fewer than the maximum number of slots in the ovector have
468        been set.
470    3.  Related to (2) above: when there are more back references in a pattern than
471        slots in the output vector, pcre_exec() uses temporary memory during
472        matching, and copies in the captures as far as possible afterwards. It was
473        using the entire output vector, but this conflicts with the specification
474        that only 2/3 is used for passing back captured substrings. Now it uses
475        only the first 2/3, for compatibility. This is, of course, another edge
476        case.
478    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
479        main code base, and can be used by building with --enable-jit. When this is
480        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
481        runtime --no-jit option is given.
483    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
484        ovector, the return from the function was zero, implying that there were
485        other matches that did not fit. The correct "exactly full" value is now
486        returned.
488    6.  If a subpattern that was called recursively or as a subroutine contained
489        (*PRUNE) or any other control that caused it to give a non-standard return,
490        invalid errors such as "Error -26 (nested recursion at the same subject
491        position)" or even infinite loops could occur.
493    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
494        computing the minimum length on reaching *ACCEPT, and so ended up with the
495        wrong value of 1 rather than 0. Further investigation indicates that
496        computing a minimum subject length in the presence of *ACCEPT is difficult
497        (think back references, subroutine calls), and so I have changed the code
498        so that no minimum is registered for a pattern that contains *ACCEPT.
500    8.  If (*THEN) was present in the first (true) branch of a conditional group,
501        it was not handled as intended. [But see 16 below.]
503    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
504        Sheri Pierce.
506    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
507        the first byte in a match must be "a".
509    11. Change 17 for 8.13 increased the recursion depth for patterns like
510        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
511        pattern contains any instances of (*THEN). If it does not, the old
512        optimizations are restored. It would be nice to do this on a per-group
513        basis, but at the moment that is not feasible.
515    12. In some environments, the output of pcretest -C is CRLF terminated. This
516        broke RunTest's code that checks for the link size. A single white space
517        character after the value is now allowed for.
519    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
520        For "fr", it uses the Windows-specific input and output files.
522    14. If (*THEN) appeared in a group that was called recursively or as a
523        subroutine, it did not work as intended. [But see next item.]
525    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
526        pattern fragments (but not containing any | characters). If A and B are
527        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
528        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
529        D. In other words, Perl considers parentheses that do not contain any |
530        characters to be part of a surrounding alternative, whereas PCRE was
531        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
532        differently. PCRE now behaves in the same way as Perl, except in the case
533        of subroutine/recursion calls such as (?1) which have in any case always
534        been different (but PCRE had them first :-).
536    16. Related to 15 above: Perl does not treat the | in a conditional group as
537        creating alternatives. Such a group is treated in the same way as an
538        ordinary group without any | characters when processing (*THEN). PCRE has
539        been changed to match Perl's behaviour.
541    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
542        RunGrepTest script failed.
544    18. Change 22 for version 13 caused atomic groups to use more stack. This is
545        inevitable for groups that contain captures, but it can lead to a lot of
546        stack use in large patterns. The old behaviour has been restored for atomic
547        groups that do not contain any capturing parentheses.
549    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
550        suppress the check for a minimum subject length at run time. (If it was
551        given to pcre_exec() or pcre_dfa_exec() it did work.)
553    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
554        fail to work when decoding hex characters in data strings in EBCDIC
555        environments.
557    21. It appears that in at least one Mac OS environment, the isxdigit() function
558        is implemented as a macro that evaluates to its argument more than once,
559        contravening the C 90 Standard (I haven't checked a later standard). There
560        was an instance in pcretest which caused it to go wrong when processing
561        \x{...} escapes in subject strings. The has been rewritten to avoid using
562        things like p++ in the argument of isxdigit().
565    Version 8.13 16-Aug-2011
566    ------------------------
568    1.  The Unicode data tables have been updated to Unicode 6.0.0.
570    2.  Two minor typos in pcre_internal.h have been fixed.
572    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
573        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
574        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
576    4.  There were a number of related bugs in the code for matching backrefences
577        caselessly in UTF-8 mode when codes for the characters concerned were
578        different numbers of bytes. For example, U+023A and U+2C65 are an upper
579        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
580        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
581        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
582        2-byte code at the end of the subject (it thought there wasn't enough data
583        left).
585    5.  Comprehensive information about what went wrong is now returned by
586        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
587        as the output vector has at least 2 elements. The offset of the start of
588        the failing character and a reason code are placed in the vector.
590    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
591        now returned is for the first byte of the failing character, instead of the
592        last byte inspected. This is an incompatible change, but I hope it is small
593        enough not to be a problem. It makes the returned offset consistent with
594        pcre_exec() and pcre_dfa_exec().
596    7.  pcretest now gives a text phrase as well as the error number when
597        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
598        failure, the offset and reason code are output.
600    8.  When \R was used with a maximizing quantifier it failed to skip backwards
601        over a \r\n pair if the subsequent match failed. Instead, it just skipped
602        back over a single character (\n). This seems wrong (because it treated the
603        two characters as a single entity when going forwards), conflicts with the
604        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
605        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
606        has been changed.
608    9.  Some internal refactoring has changed the processing so that the handling
609        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
610        time (the PCRE_DOTALL option was changed this way some time ago: version
611        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
612        which was always a bit of a fudge. It also means that there is one less
613        argument for the match() function, which reduces its stack requirements
614        slightly. This change also fixes an incompatibility with Perl: the pattern
615        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
617    10. More internal refactoring has drastically reduced the number of recursive
618        calls to match() for possessively repeated groups such as (abc)++ when
619        using pcre_exec().
621    11. While implementing 10, a number of bugs in the handling of groups were
622        discovered and fixed:
624        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
625        (a|)*(?1) gave a compile-time internal error.
626        ((a|)+)+  did not notice that the outer group could match an empty string.
627        (^a|^)+   was not marked as anchored.
628        (.*a|.*)+ was not marked as matching at start or after a newline.
630    12. Yet more internal refactoring has removed another argument from the match()
631        function. Special calls to this function are now indicated by setting a
632        value in a variable in the "match data" data block.
634    13. Be more explicit in pcre_study() instead of relying on "default" for
635        opcodes that mean there is no starting character; this means that when new
636        ones are added and accidentally left out of pcre_study(), testing should
637        pick them up.
639    14. The -s option of pcretest has been documented for ages as being an old
640        synonym of -m (show memory usage). I have changed it to mean "force study
641        for every regex", that is, assume /S for every regex. This is similar to -i
642        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
643        using it. It makes it easier to run collections of tests with and without
644        study enabled, and thereby test pcre_study() more easily. All the standard
645        tests are now run with and without -s (but some patterns can be marked as
646        "never study" - see 20 below).
648    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
649        restoration of the capturing data to the outer values was not happening
650        correctly.
652    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
653        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
654        pattern had matched an empty string, and so incorrectly returned a no
655        match.
657    17. There was optimizing code for the last branch of non-capturing parentheses,
658        and also for the obeyed branch of a conditional subexpression, which used
659        tail recursion to cut down on stack usage. Unfortunately, now that there is
660        the possibility of (*THEN) occurring in these branches, tail recursion is
661        no longer possible because the return has to be checked for (*THEN). These
662        two optimizations have therefore been removed. [But see 8.20/11 above.]
664    18. If a pattern containing \R was studied, it was assumed that \R always
665        matched two bytes, thus causing the minimum subject length to be
666        incorrectly computed because \R can also match just one byte.
668    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
669        was incorrectly computed.
671    20. If /S is present twice on a test pattern in pcretest input, it now
672        *disables* studying, thereby overriding the use of -s on the command line
673        (see 14 above). This is necessary for one or two tests to keep the output
674        identical in both cases.
676    21. When (*ACCEPT) was used in an assertion that matched an empty string and
677        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
679    22. When an atomic group that contained a capturing parenthesis was
680        successfully matched, but the branch in which it appeared failed, the
681        capturing was not being forgotten if a higher numbered group was later
682        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
683        group 1 to "a", when in fact it should be unset. This applied to multi-
684        branched capturing and non-capturing groups, repeated or not, and also to
685        positive assertions (capturing in negative assertions does not happen
686        in PCRE) and also to nested atomic groups.
688    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
689        subject after a captured substring, to make it easier to tell which of a
690        number of identical substrings has been captured.
692    24. The way atomic groups are processed by pcre_exec() has been changed so that
693        if they are repeated, backtracking one repetition now resets captured
694        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
695        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
696        "aaa". Previously, it would have been "a". As part of this code
697        refactoring, the way recursive calls are handled has also been changed.
699    25. If an assertion condition captured any substrings, they were not passed
700        back unless some other capturing happened later. For example, if
701        (?(?=(a))a) was matched against "a", no capturing was returned.
703    26. When studying a pattern that contained subroutine calls or assertions,
704        the code for finding the minimum length of a possible match was handling
705        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
706        group 1 called group 2 while simultaneously a separate group 2 called group
707        1). A stack overflow occurred in this case. I have fixed this by limiting
708        the recursion depth to 10.
710    27. Updated RunTest.bat in the distribution to the version supplied by Tom
711        Fortmann. This supports explicit test numbers on the command line, and has
712        argument validation and error reporting.
714    28. An instance of \X with an unlimited repeat could fail if at any point the
715        first character it looked at was a mark character.
717    29. Some minor code refactoring concerning Unicode properties and scripts
718        should reduce the stack requirement of match() slightly.
720    30. Added the '=' option to pcretest to check the setting of unused capturing
721        slots at the end of the pattern, which are documented as being -1, but are
722        not included in the return count.
724    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
725        compiled something random. Now it gives a compile-time error (as does
726        Perl).
728    32. A *MARK encountered during the processing of a positive assertion is now
729        recorded and passed back (compatible with Perl).
731    33. If --only-matching or --colour was set on a pcregrep call whose pattern
732        had alternative anchored branches, the search for a second match in a line
733        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
734        matched the line "0102" twice. The same bug affected patterns that started
735        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
736        twice.
738    34. Previously, PCRE did not allow quantification of assertions. However, Perl
739        does, and because of capturing effects, quantifying parenthesized
740        assertions may at times be useful. Quantifiers are now allowed for
741        parenthesized assertions.
743    35. A minor code tidy in pcre_compile() when checking options for \R usage.
745    36. \g was being checked for fancy things in a character class, when it should
746        just be a literal "g".
748    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
749        appearance of a nested POSIX class supersedes an apparent external class.
750        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
751        unescaped square brackets may also appear as part of class names. For
752        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
753        more like Perl. (But see 8.20/1 above.)
755    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
756        was because it thought it was \N{name}, which is not supported).
758    39. Add minix to OS list not supporting the -S option in pcretest.
760    40. PCRE tries to detect cases of infinite recursion at compile time, but it
761        cannot analyze patterns in sufficient detail to catch mutual recursions
762        such as ((?1))((?2)). There is now a runtime test that gives an error if a
763        subgroup is called recursively as a subpattern for a second time at the
764        same position in the subject string. In previous releases this might have
765        been caught by the recursion limit, or it might have run out of stack.
767    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
768        happen only once. PCRE was, however incorrectly giving a compile time error
769        "recursive call could loop indefinitely" because it cannot analyze the
770        pattern in sufficient detail. The compile time test no longer happens when
771        PCRE is compiling a conditional subpattern, but actual runaway loops are
772        now caught at runtime (see 40 above).
774    42. It seems that Perl allows any characters other than a closing parenthesis
775        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
776        has been changed to be the same.
778    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
779        as not to get warnings when autogen.sh is called. Also changed
780        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
782    44. To help people who use pcregrep to scan files containing exceedingly long
783        lines, the following changes have been made:
785        (a) The default value of the buffer size parameter has been increased from
786            8K to 20K. (The actual buffer used is three times this size.)
788        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
789            PCRE is built.
791        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
792            to be set at run time.
794        (d) Numerical values in pcregrep options can be followed by K or M, for
795            example --buffer-size=50K.
797        (e) If a line being scanned overflows pcregrep's buffer, an error is now
798            given and the return code is set to 2.
800    45. Add a pointer to the latest mark to the callout data block.
802    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
803        partial match of an empty string instead of no match. This was specific to
804        the use of ".".
806    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
807        complete match instead of a partial match. This bug was dependent on both
808        the PCRE_UTF8 and PCRE_DOTALL options being set.
810    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
811        starting byte set, because \b was not being ignored.
814    Version 8.12 15-Jan-2011
815    ------------------------
817    1.  Fixed some typos in the markup of the man pages, and wrote a script that
818        checks for such things as part of the documentation building process.
820    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
821        --match-limit and --recursion-limit options (added for 8.11). In
822        particular, this made one of the standard tests fail. (The integer value
823        went into the wrong half of a long int.)
825    3.  If the --colour option was given to pcregrep with -v (invert match), it
826        did strange things, either producing crazy output, or crashing. It should,
827        of course, ignore a request for colour when reporting lines that do not
828        match.
830    4.  Another pcregrep bug caused similar problems if --colour was specified with
831        -M (multiline) and the pattern match finished with a line ending.
833    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
834        matched in multiline mode, the following line was shown as part of the
835        match. This seems wrong, so I have changed it.
837    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
838        the check for further matches in the same line (so they could be coloured)
839        to overrun the end of the current line. If another match was found, it was
840        incorrectly shown (and then shown again when found in the next line).
842    7.  If pcregrep was compiled under Windows, there was a reference to the
843        function pcregrep_exit() before it was defined. I am assuming this was
844        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
845        reported by a user. I've moved the definition above the reference.
848    Version 8.11 10-Dec-2010
849    ------------------------
851    1.  (*THEN) was not working properly if there were untried alternatives prior
852        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
853        backtracked to try for "b" instead of moving to the next alternative branch
854        at the same level (in this case, to look for "c"). The Perl documentation
855        is clear that when (*THEN) is backtracked onto, it goes to the "next
856        alternative in the innermost enclosing group".
858    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
859        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
860        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
861        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
862        (*THEN).
864    3.  If \s appeared in a character class, it removed the VT character from
865        the class, even if it had been included by some previous item, for example
866        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
867        of \s, but is part of the POSIX "space" class.)
869    4.  A partial match never returns an empty string (because you can always
870        match an empty string at the end of the subject); however the checking for
871        an empty string was starting at the "start of match" point. This has been
872        changed to the "earliest inspected character" point, because the returned
873        data for a partial match starts at this character. This means that, for
874        example, /(?<=abc)def/ gives a partial match for the subject "abc"
875        (previously it gave "no match").
877    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
878        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
879        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
880        has an implication that the given string is incomplete (because a partial
881        match is preferred over a full match). For this reason, these items now
882        give a partial match in this situation. [Aside: previously, the one case
883        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
884        match rather than a full match, which was wrong by the old rules, but is
885        now correct.]
887    6.  There was a bug in the handling of #-introduced comments, recognized when
888        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
889        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
890        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
891        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
892        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
893        places in pcre_compile().
895    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
896        comments when looking ahead for named forward references to subpatterns,
897        the only newline sequence it recognized was NL. It now handles newlines
898        according to the set newline convention.
900    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
901        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
902        cater for a lack of strerror(). These oversights have been fixed.
904    9.  Added --match-limit and --recursion-limit to pcregrep.
906    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
908    11. When the -o option was used, pcregrep was setting a return code of 1, even
909        when matches were found, and --line-buffered was not being honoured.
911    12. Added an optional parentheses number to the -o and --only-matching options
912        of pcregrep.
914    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
915        can match an empty string. The code to do it in pcretest and pcredemo
916        needed fixing:
918        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
919            only one byte after an empty string match just before CRLF (this case
920            just got forgotten; "any" and "anycrlf" were OK).
922        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
923            mode when an empty string match preceded an ASCII character followed by
924            a non-ASCII character. (The code for advancing by one character rather
925            than one byte was nonsense.)
927        (c) The pcredemo.c sample program did not have any code at all to handle
928            the cases when CRLF is a valid newline sequence.
930    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
931        as a starting offset was within the subject string. There is now a new
932        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
933        negative or greater than the length of the string. In order to test this,
934        pcretest is extended to allow the setting of negative starting offsets.
936    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
937        starting offset points to the beginning of a UTF-8 character was
938        unnecessarily clumsy. I tidied it up.
940    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
941        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
943    17. Nobody had reported that the --include_dir option, which was added in
944        release 7.7 should have been called --include-dir (hyphen, not underscore)
945        for compatibility with GNU grep. I have changed it to --include-dir, but
946        left --include_dir as an undocumented synonym, and the same for
947        --exclude-dir, though that is not available in GNU grep, at least as of
948        release 2.5.4.
950    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
951        characters from a string of bytes) have been redefined so as not to use
952        loops, in order to improve performance in some environments. At the same
953        time, I abstracted some of the common code into auxiliary macros to save
954        repetition (this should not affect the compiled code).
956    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
957        compile-time error is now given if \c is not followed by an ASCII
958        character, that is, a byte less than 128. (In EBCDIC mode, the code is
959        different, and any byte value is allowed.)
961    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
962        START_OPTIMIZE option, which is now allowed at compile time - but just
963        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
964        to pcregrep and other applications that have no direct access to PCRE
965        options. The new /Y option in pcretest sets this option when calling
966        pcre_compile().
968    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
969        back references. Groups containing recursive back references were forced to
970        be atomic by that change, but in the case of named groups, the amount of
971        memory required was incorrectly computed, leading to "Failed: internal
972        error: code overflow". This has been fixed.
974    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
975        pcretest.c, to avoid build problems in some Borland environments.
978    Version 8.10 25-Jun-2010
979    ------------------------
981    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
982        THEN.
984    2.  (*ACCEPT) was not working when inside an atomic group.
986    3.  Inside a character class, \B is treated as a literal by default, but
987        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
988        causes the error). The code is unchanged, but I tidied the documentation.
990    4.  Inside a character class, PCRE always treated \R and \X as literals,
991        whereas Perl faults them if its -w option is set. I have changed PCRE so
992        that it faults them when PCRE_EXTRA is set.
994    5.  Added support for \N, which always matches any character other than
995        newline. (It is the same as "." when PCRE_DOTALL is not set.)
997    6.  When compiling pcregrep with newer versions of gcc which may have
998        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
999        declared with attribute warn_unused_result" were given. Just casting the
1000        result to (void) does not stop the warnings; a more elaborate fudge is
1001        needed. I've used a macro to implement this.
1003    7.  Minor change to pcretest.c to avoid a compiler warning.
1005    8.  Added four artifical Unicode properties to help with an option to make
1006        \s etc use properties (see next item). The new properties are: Xan
1007        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1009    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1010        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1011        this option. Modified pcretest to add /W to test this facility. Added
1012        REG_UCP to make it available via the POSIX interface.
1014    10. Added --line-buffered to pcregrep.
1016    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1017        studied, and the match started with a letter with a code point greater than
1018        127 whose first byte was different to the first byte of the other case of
1019        the letter, the other case of this starting letter was not recognized
1020        (#976).
1022    12. If a pattern that was studied started with a repeated Unicode property
1023        test, for example, \p{Nd}+, there was the theoretical possibility of
1024        setting up an incorrect bitmap of starting bytes, but fortunately it could
1025        not have actually happened in practice until change 8 above was made (it
1026        added property types that matched character-matching opcodes).
1028    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1029        possible starting bytes for non-anchored patterns.
1031    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1032        \R, and also a number of cases that involve Unicode properties, both
1033        explicit and implicit when PCRE_UCP is set.
1035    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1036        input, it could crash or give wrong results if characters with values
1037        greater than 0xc0 were present in the subject string. (Detail: it assumed
1038        UTF-8 input when processing these items.)
1040    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1041        size_t is 64-bit (#991).
1043    17. Added a check for running out of memory when PCRE is compiled with
1044        --disable-stack-for-recursion (#990).
1046    18. If the last data line in a file for pcretest does not have a newline on
1047        the end, a newline was missing in the output.
1049    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1050        less than 128) in its various bitmaps. However, there is a facility for
1051        generating tables according to the current locale when PCRE is compiled. It
1052        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1053        characters, are recognized by isspace() and therefore were getting set in
1054        these tables, and indeed these tables seem to approximate to ISO 8859. This
1055        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1056        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1057        which of course cannot start UTF-8 characters. I have changed the code so
1058        that only real ASCII characters (less than 128) and the correct starting
1059        bytes for UTF-8 encodings are set for characters greater than 127 when in
1060        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1061        altogether.)
1063    20. Added the /T option to pcretest so as to be able to run tests with non-
1064        standard character tables, thus making it possible to include the tests
1065        used for 19 above in the standard set of tests.
1067    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1068        reference to a subpattern the other side of a comment that contains an
1069        opening parenthesis caused either an internal compiling error, or a
1070        reference to the wrong subpattern.
1073    Version 8.02 19-Mar-2010
1074    ------------------------
1076    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1078    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1079        configured.
1081    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1082        original author of that file, following a query about its status.
1084    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1085        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1087    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1088        quantifier applied to a forward-referencing subroutine call, could compile
1089        incorrect code or give the error "internal error: previously-checked
1090        referenced subpattern not found".
1092    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1093        variables to point to external functions. For these systems, therefore,
1094        pcre_malloc etc. are now initialized to local functions that call the
1095        relevant global functions.
1097    7.  There were two entries missing in the vectors called coptable and poptable
1098        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1099        I've fixed the data, and added a kludgy way of testing at compile time that
1100        the lengths are correct (equal to the number of opcodes).
1102    8.  Following on from 7, I added a similar kludge to check the length of the
1103        eint vector in pcreposix.c.
1105    9.  Error texts for pcre_compile() are held as one long string to avoid too
1106        much relocation at load time. To find a text, the string is searched,
1107        counting zeros. There was no check for running off the end of the string,
1108        which could happen if a new error number was added without updating the
1109        string.
1111    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1113    11. \K was not working if it appeared in an atomic group or in a group that
1114        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1115        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1116        the assertion is positive, but not if it is negative.
1118    12. Change 11 fortuitously reduced the size of the stack frame used in the
1119        "match()" function of pcre_exec.c by one pointer. Forthcoming
1120        implementation of support for (*MARK) will need an extra pointer on the
1121        stack; I have reserved it now, so that the stack frame size does not
1122        decrease.
1124    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1125        item in branch that calls a recursion is a subroutine call - as in the
1126        second branch in the above example - was incorrectly given the compile-
1127        time error "recursive call could loop indefinitely" because pcre_compile()
1128        was not correctly checking the subroutine for matching a non-empty string.
1130    14. The checks for overrunning compiling workspace could trigger after an
1131        overrun had occurred. This is a "should never occur" error, but it can be
1132        triggered by pathological patterns such as hundreds of nested parentheses.
1133        The checks now trigger 100 bytes before the end of the workspace.
1135    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1138    Version 8.01 19-Jan-2010
1139    ------------------------
1141    1.  If a pattern contained a conditional subpattern with only one branch (in
1142        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1143        computed the wrong minimum data length (which is of course zero for such
1144        subpatterns). This could cause incorrect "no match" results.
1146    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1147        the pattern is reset in the first branch, pcre_compile() failed with
1148        "internal error: code overflow at offset...". This happened only when
1149        the reset was to the original external option setting. (An optimization
1150        abstracts leading options settings into an external setting, which was the
1151        cause of this.)
1153    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1154        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1155        assertion pattern did not match (meaning that the assertion was true), it
1156        was incorrectly treated as false if the SKIP had been reached during the
1157        matching. This also applied to assertions used as conditions.
1159    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1160        assertion subpattern, including such a pattern used as a condition,
1161        unpredictable results occurred, instead of the error return
1164    5.  The C++ GlobalReplace function was not working like Perl for the special
1165        situation when an empty string is matched. It now does the fancy magic
1166        stuff that is necessary.
1168    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1169        removed. (These were left over from very, very early versions of PCRE.)
1171    7.  Some cosmetic changes to the code to make life easier when compiling it
1172        as part of something else:
1174        (a) Change DEBUG to PCRE_DEBUG.
1176        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1177            called "current" as "current_branch", to prevent a collision with the
1178            Linux macro when compiled as a kernel module.
1180        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1181            prevent a collision with the Linux macro when compiled as a kernel
1182            module.
1184    8.  In pcre_compile() there are some checks for integer overflows that used to
1185        cast potentially large values to (double). This has been changed to that
1186        when building, a check for int64_t is made, and if it is found, it is used
1187        instead, thus avoiding the use of floating point arithmetic. (There is no
1188        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1189        double.
1191    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1192        2005 (difference between two addresses compared to an unsigned value).
1194    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1195        custom one, because of the following reported problem in Windows:
1197          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1198              under Win32.
1199          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1200              therefore missing the function definition.
1201          - The compiler thus generates a "C" signature for the test function.
1202          - The linker fails to find the "C" function.
1203          - PCRE fails to configure if asked to do so against libbz2.
1205    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1206        messages were output:
1208          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1209          rerunning libtoolize, to keep the correct libtool macros in-tree.
1210          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1212        I have done both of these things.
1214    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1215        most of the time, it *can* run out if it is given a pattern that contains a
1216        runaway infinite recursion. I updated the discussion in the pcrestack man
1217        page.
1219    13. Now that we have gone to the x.xx style of version numbers, the minor
1220        version may start with zero. Using 08 or 09 is a bad idea because users
1221        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1222        interpreted as invalid octal numbers. I've updated the previous comment in
1223        configure.ac, and also added a check that gives an error if 08 or 09 are
1224        used.
1226    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1227        causing partial matching to fail when the end of the subject matched \W
1228        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1230    15. There were some discrepancies between the declarations in pcre_internal.h
1231        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1232        their definitions. The declarations used "const uschar *" and the
1233        definitions used USPTR. Even though USPTR is normally defined as "const
1234        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1235        reported that: "This difference in casting confuses some C++ compilers, for
1236        example, SunCC recognizes above declarations as different functions and
1237        generates broken code for hbpcre." I have changed the declarations to use
1238        USPTR.
1240    16. GNU libtool is named differently on some systems. The autogen.sh script now
1241        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1242        (FreeBSD).
1244    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1245        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1246        comment: "Figure out how to create a longlong from a string: strtoll and
1247        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1248        instance, but it only takes 2 args instead of 3!"
1250    18. A subtle bug concerned with back references has been fixed by a change of
1251        specification, with a corresponding code fix. A pattern such as
1252        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1253        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1254        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1255        same bug. Such groups have to be quantified to be useful, or contained
1256        inside another quantified group. (If there's no repetition, the reference
1257        can never match.) The problem arises because, having left the group and
1258        moved on to the rest of the pattern, a later failure that backtracks into
1259        the group uses the captured value from the final iteration of the group
1260        rather than the correct earlier one. I have fixed this in PCRE by forcing
1261        any group that contains a reference to itself to be an atomic group; that
1262        is, there cannot be any backtracking into it once it has completed. This is
1263        similar to recursive and subroutine calls.
1266    Version 8.00 19-Oct-09
1267    ----------------------
1269    1.  The table for translating pcre_compile() error codes into POSIX error codes
1270        was out-of-date, and there was no check on the pcre_compile() error code
1271        being within the table. This could lead to an OK return being given in
1272        error.
1274    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1275        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1276        in a Windows environment.
1278    3.  The pcregrep --count option prints the count for each file even when it is
1279        zero, as does GNU grep. However, pcregrep was also printing all files when
1280        --files-with-matches was added. Now, when both options are given, it prints
1281        counts only for those files that have at least one match. (GNU grep just
1282        prints the file name in this circumstance, but including the count seems
1283        more useful - otherwise, why use --count?) Also ensured that the
1284        combination -clh just lists non-zero counts, with no names.
1286    4.  The long form of the pcregrep -F option was incorrectly implemented as
1287        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1288        but it seems right to fix it, and I didn't think it was worth preserving
1289        the old behaviour.
1291    5.  The command line items --regex=pattern and --regexp=pattern were not
1292        recognized by pcregrep, which required --regex pattern or --regexp pattern
1293        (with a space rather than an '='). The man page documented the '=' forms,
1294        which are compatible with GNU grep; these now work.
1296    6.  No libpcreposix.pc file was created for pkg-config; there was just
1297        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1299    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1300        when UCP support is not needed, by modifying the Python script that
1301        generates it from Unicode data files. This should not matter if the module
1302        is correctly used as a library, but I received one complaint about 50K of
1303        unwanted data. My guess is that the person linked everything into his
1304        program rather than using a library. Anyway, it does no harm.
1306    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1307        was a minimum greater than 1 for a wide character in a possessive
1308        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1309        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1310        character. Chaos in the form of incorrect output or a compiling loop could
1311        result.
1313    9.  The restrictions on what a pattern can contain when partial matching is
1314        requested for pcre_exec() have been removed. All patterns can now be
1315        partially matched by this function. In addition, if there are at least two
1316        slots in the offset vector, the offset of the earliest inspected character
1317        for the match and the offset of the end of the subject are set in them when
1318        PCRE_ERROR_PARTIAL is returned.
1320    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1321        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1322        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1323        and may be more useful for multi-segment matching.
1325    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1326        used to be given if ever the end of the subject was reached; now it is
1327        given only if matching could not proceed because another character was
1328        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1329        string "Z", which now yields "no match" instead of "partial match". In the
1330        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1331        final character ended with (*FAIL).
1333    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1334        if the pattern had a "must contain" character that was already found in the
1335        earlier partial match, unless partial matching was again requested. For
1336        example, with the pattern /dog.(body)?/, the "must contain" character is
1337        "g". If the first part-match was for the string "dog", restarting with
1338        "sbody" failed. This bug has been fixed.
1340    13. The string returned by pcre_dfa_exec() after a partial match has been
1341        changed so that it starts at the first inspected character rather than the
1342        first character of the match. This makes a difference only if the pattern
1343        starts with a lookbehind assertion or \b or \B (\K is not supported by
1344        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1345        matching functions compatible, and I think it's the right thing to do.
1347    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1348        so that the demonstration program is easily available in environments where
1349        PCRE has not been installed from source.
1351    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1352        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1353        library.
1355    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1356        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1357        is not the first non-POSIX option to be added. Clearly some people find
1358        these options useful.
1360    17. If a caller to the POSIX matching function regexec() passes a non-zero
1361        value for nmatch with a NULL value for pmatch, the value of
1362        nmatch is forced to zero.
1364    18. RunGrepTest did not have a test for the availability of the -u option of
1365        the diff command, as RunTest does. It now checks in the same way as
1366        RunTest, and also checks for the -b option.
1368    19. If an odd number of negated classes containing just a single character
1369        interposed, within parentheses, between a forward reference to a named
1370        subpattern and the definition of the subpattern, compilation crashed with
1371        an internal error, complaining that it could not find the referenced
1372        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1373        [The bug was that it was starting one character too far in when skipping
1374        over the character class, thus treating the ] as data rather than
1375        terminating the class. This meant it could skip too much.]
1377    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1378        /g option in pcretest when the pattern contains \K, which makes it possible
1379        to have an empty string match not at the start, even when the pattern is
1380        anchored. Updated pcretest and pcredemo to use this option.
1382    21. If the maximum number of capturing subpatterns in a recursion was greater
1383        than the maximum at the outer level, the higher number was returned, but
1384        with unset values at the outer level. The correct (outer level) value is
1385        now given.
1387    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1388        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1389        make it do so. The string so far is captured, making this feature
1390        compatible with Perl.
1392    23. The tests have been re-organized, adding tests 11 and 12, to make it
1393        possible to check the Perl 5.10 features against Perl 5.10.
1395    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1396        pattern matches a fixed length string. PCRE did not allow this; now it
1397        does. Neither allows recursion.
1399    25. I finally figured out how to implement a request to provide the minimum
1400        length of subject string that was needed in order to match a given pattern.
1401        (It was back references and recursion that I had previously got hung up
1402        on.) This code has now been added to pcre_study(); it finds a lower bound
1403        to the length of subject needed. It is not necessarily the greatest lower
1404        bound, but using it to avoid searching strings that are too short does give
1405        some useful speed-ups. The value is available to calling programs via
1406        pcre_fullinfo().
1408    26. While implementing 25, I discovered to my embarrassment that pcretest had
1409        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1410        study optimizations had never been tested with that matching function.
1411        Oops. What is worse, even when it was passed study data, there was a bug in
1412        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1413        were also very few tests of studied patterns with pcre_dfa_exec().
1415    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1416        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1417        on the other side of the coin, they are no longer allowed to have different
1418        names, because these cannot be distinguished in PCRE, and this has caused
1419        confusion. (This is a difference from Perl.)
1421    28. When duplicate subpattern names are present (necessarily with different
1422        numbers, as required by 27 above), and a test is made by name in a
1423        conditional pattern, either for a subpattern having been matched, or for
1424        recursion in such a pattern, all the associated numbered subpatterns are
1425        tested, and the overall condition is true if the condition is true for any
1426        one of them. This is the way Perl works, and is also more like the way
1427        testing by number works.
1430    Version 7.9 11-Apr-09
1431    ---------------------
1433    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1434        (pcretest), all targets were linked against these libraries. This included
1435        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1436        libraries. This caused unwanted dependencies to be created. This problem
1437        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1438        pcretest is linked with readline.
1440    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1441        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1442        moved outside it again, because FALSE and TRUE are already defined in AIX,
1443        but BOOL is not.
1445    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1446        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1448    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1449        hyphens) following file names and line numbers when outputting matching
1450        lines. This is not true; no spaces are inserted. I have also clarified the
1451        wording for the --colour (or --color) option.
1453    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1454        was not coloured; this is different to GNU grep, so I have changed it to be
1455        the same.
1457    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1458        each matching line was coloured. Now it goes on to look for further matches
1459        of any of the test patterns, which is the same behaviour as GNU grep.
1461    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1462        doesn't make sense to accept an empty string match in pcregrep, so I have
1463        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1464        seems to be how GNU grep behaves.
1466    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1467        start or after a newline", because the conditional assertion was not being
1468        correctly handled. The rule now is that both the assertion and what follows
1469        in the first alternative must satisfy the test.
1471    9.  If auto-callout was enabled in a pattern with a conditional group whose
1472        condition was an assertion, PCRE could crash during matching, both with
1473        pcre_exec() and pcre_dfa_exec().
1475    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1476        used for matching.
1478    11. Unicode property support in character classes was not working for
1479        characters (bytes) greater than 127 when not in UTF-8 mode.
1481    12. Added the -M command line option to pcretest.
1483    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1485    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1487    16. Added comments and documentation about mis-use of no_arg in the C++
1488        wrapper.
1490    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1491        from Martin Jerabek that uses macro names for all relevant character and
1492        string constants.
1494    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1495        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1496        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1497        these, but not everybody uses configure.
1499    19. A conditional group that had only one branch was not being correctly
1500        recognized as an item that could match an empty string. This meant that an
1501        enclosing group might also not be so recognized, causing infinite looping
1502        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1503        with the subject "ab", where knowledge that the repeated group can match
1504        nothing is needed in order to break the loop.
1506    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1507        exec(), but without supplying a callout function, matching went wrong.
1509    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1510        leak if the size of the offset vector was greater than 30. When the vector
1511        is smaller, the saved offsets during recursion go onto a local stack
1512        vector, but for larger vectors malloc() is used. It was failing to free
1513        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1514        error, in fact).
1516    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1517        heapframe that is used only when UTF-8 support is enabled. This caused no
1518        problem, but was untidy.
1520    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1521        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1522        included within another project.
1524    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1525        slightly modified by me:
1527          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1528              not building pcregrep.
1530          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1531              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1533    25. Forward references, both numeric and by name, in patterns that made use of
1534        duplicate group numbers, could behave incorrectly or give incorrect errors,
1535        because when scanning forward to find the reference group, PCRE was not
1536        taking into account the duplicate group numbers. A pattern such as
1537        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1539    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1540        the feature of a custom pointer more persuasive (as requested by a user).
1542    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1543        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1544        pre-defined.
1546    28. Added support for (*UTF8) at the start of a pattern.
1548    29. Arrange for flags added by the "release type" setting in CMake to be shown
1549        in the configuration summary.
1552    Version 7.8 05-Sep-08
1553    ---------------------
1555    1.  Replaced UCP searching code with optimized version as implemented for Ad
1556        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1557        stage table and inline lookup instead of a function, giving speed ups of 2
1558        to 5 times on some simple patterns that I tested. Permission was given to
1559        distribute the MultiStage2.py script that generates the tables (it's not in
1560        the tarball, but is in the Subversion repository).
1562    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1563        scripts.
1565    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1566        a group with a zero qualifier. The result of the study could be incorrect,
1567        or the function might crash, depending on the pattern.
1569    4.  Caseless matching was not working for non-ASCII characters in back
1570        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1571        It now works when Unicode Property Support is available.
1573    5.  In pcretest, an escape such as \x{de} in the data was always generating
1574        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1575        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1576        truncation.
1578    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1580    7.  Added two (int) casts to pcregrep when printing the difference of two
1581        pointers, in case they are 64-bit values.
1583    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1584        test 2 if it fails.
1586    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1587        and a #define of that name to empty if it is not externally set. This is to
1588        allow users of MSVC to set it if necessary.
1590    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1591        the convenience functions in the pcre_get.c source file.
1593    11. An option change at the start of a pattern that had top-level alternatives
1594        could cause overwriting and/or a crash. This command provoked a crash in
1595        some environments:
1597          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1599        This potential security problem was recorded as CVE-2008-2371.
1601    12. For a pattern where the match had to start at the beginning or immediately
1602        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1603        pcre_dfa_exec() could read past the end of the passed subject if there was
1604        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1605        pcretest so that it places the subject at the end of its malloc-ed buffer.
1607    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1608        exec() might read past the end of the data buffer in UTF-8 mode.
1610    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1611        the data contained the byte 0x85 as part of a UTF-8 character within its
1612        first line. This applied both to normal and DFA matching.
1614    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1615        /^[^d]*?$/8 failed to match "abc".
1617    16. Added a missing copyright notice to pcrecpp_internal.h.
1619    17. Make it more clear in the documentation that values returned from
1620        pcre_exec() in ovector are byte offsets, not character counts.
1622    18. Tidied a few places to stop certain compilers from issuing warnings.
1624    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1625        supplied by Stefan Weber. I made a further small update for 7.8 because
1626        there is a change of source arrangements: the pcre_searchfuncs.c module is
1627        replaced by pcre_ucd.c.
1630    Version 7.7 07-May-08
1631    ---------------------
1633    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1634        a string to a long long, pretend we don't even have a long long." This is
1635        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1637    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1638        pre-7.6 versions, which defined a global no_arg variable instead of putting
1639        it in the RE class. (See also #8 below.)
1641    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1642        Lopes.
1644    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1646        (1) The include/exclude patterns were being applied to the whole pathnames
1647            of files, instead of just to the final components.
1649        (2) If there was more than one level of directory, the subdirectories were
1650            skipped unless they satisfied the include/exclude conditions. This is
1651            inconsistent with GNU grep (and could even be seen as contrary to the
1652            pcregrep specification - which I improved to make it absolutely clear).
1653            The action now is always to scan all levels of directory, and just
1654            apply the include/exclude patterns to regular files.
1656    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1657        --exclude_dir in the tests to avoid scanning .svn directories.
1659    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1660        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1661        doesn't support NULs in patterns.
1663    7.  Added some missing "const"s to declarations of static tables in
1664        pcre_compile.c and pcre_dfa_exec.c.
1666    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1667        caused by fix #2  above. (Subsequently also a second patch to fix the
1668        first patch. And a third patch - this was a messy problem.)
1670    9.  Applied Craig's patch to remove the use of push_back().
1672    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1673        matching function regexec().
1675    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1676        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1677        references. PCRE supports relative numbers with this syntax (I don't think
1678        Oniguruma does).
1680    12. Previously, a group with a zero repeat such as (...){0} was completely
1681        omitted from the compiled regex. However, this means that if the group
1682        was called as a subroutine from elsewhere in the pattern, things went wrong
1683        (an internal error was given). Such groups are now left in the compiled
1684        pattern, with a new opcode that causes them to be skipped at execution
1685        time.
1687    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1688        to the way PCRE behaves:
1690        (a) A lone ] character is dis-allowed (Perl treats it as data).
1692        (b) A back reference to an unmatched subpattern matches an empty string
1693            (Perl fails the current match path).
1695        (c) A data ] in a character class must be notated as \] because if the
1696            first data character in a class is ], it defines an empty class. (In
1697            Perl it is not possible to have an empty class.) The empty class []
1698            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1699            The negative empty class [^] matches any one character, independently
1700            of the DOTALL setting.
1702    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1703        non-existent subpattern following a character class starting with ']' and
1704        containing () gave an internal compiling error instead of "reference to
1705        non-existent subpattern". Fortunately, when the pattern did exist, the
1706        compiled code was correct. (When scanning forwards to check for the
1707        existencd of the subpattern, it was treating the data ']' as terminating
1708        the class, so got the count wrong. When actually compiling, the reference
1709        was subsequently set up correctly.)
1711    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1712        it was being rejected as not supported by pcre_dfa_exec(), even though
1713        other assertions are supported. I have made pcre_dfa_exec() support
1714        (*FAIL).
1716    16. The implementation of 13c above involved the invention of a new opcode,
1717        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1718        cannot be changed at match time, I realized I could make a small
1719        improvement to matching performance by compiling OP_ALLANY instead of
1720        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1721        on the OP_ANY path.
1723    17. Compiling pcretest on Windows with readline support failed without the
1724        following two fixes: (1) Make the unistd.h include conditional on
1725        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1727    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1728        ncurses library to be included for pcretest when ReadLine support is
1729        requested, but also to allow for it to be overridden. This patch came from
1730        Daniel Bergström.
1732    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1733        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1734        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1735        spotting this.
1738    Version 7.6 28-Jan-08
1739    ---------------------
1741    1.  A character class containing a very large number of characters with
1742        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1743        overflow.
1745    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1746        HAVE_LONG_LONG is not defined.
1748    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1749        bring them up to date and include new features. This patch includes:
1751        - Fixed PH's badly added libz and libbz2 support.
1752        - Fixed a problem with static linking.
1753        - Added pcredemo. [But later removed - see 7 below.]
1754        - Fixed dftables problem and added an option.
1755        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1756            HAVE_LONG_LONG.
1757        - Added readline support for pcretest.
1758        - Added an listing of the option settings after cmake has run.
1760    4.  A user submitted a patch to Makefile that makes it easy to create
1761        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1762        Makefile.am that cause it to include this special target, without
1763        affecting anything else. Note that the same mingw target plus all
1764        the other distribution libraries and programs are now supported
1765        when configuring with CMake (see 6 below) instead of with
1766        Configure/Make.
1768    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1769        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1770        exported in the Windows port". It has not yet been confirmed that the patch
1771        solves the problem, but it does no harm.
1773    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1774        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1775        with CMake, and also correct the comment about stack recursion.
1777    7.  Remove the automatic building of pcredemo from the ./configure system and
1778        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1779        of a program that users should build themselves after PCRE is installed, so
1780        building it automatically is not really right. What is more, it gave
1781        trouble in some build environments.
1783    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1786    Version 7.5 10-Jan-08
1787    ---------------------
1789    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1790        values in parens when parsing an RE using the C++ wrapper."
1792    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1793        Characters greater than 255 were excluded from the class instead of being
1794        included.
1796    3.  The same bug as (2) above applied to negated POSIX classes such as
1797        [:^space:].
1799    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1800        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1801        I have changed it.
1803    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1804        first named subpattern) and a construct such as (?&a) would reference the
1805        first named subpattern whose name started with "a" (in other words, the
1806        length check was missing). Both these problems are fixed. "Subpattern name
1807        expected" is now given for (?&) (a zero-length name), and this patch also
1808        makes it give the same error for \k'' (previously it complained that that
1809        was a reference to a non-existent subpattern).
1811    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1812        this is right because (?- can be followed by option settings as well as by
1813        digits. I have, however, made the messages clearer.
1815    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1816        than the number used in the conditional) now cause a compile-time error.
1817        This is actually not compatible with Perl, which accepts such patterns, but
1818        treats the conditional as always being FALSE (as PCRE used to), but it
1819        seems to me that giving a diagnostic is better.
1821    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1822        and messages.
1824    9.  Fix two occurrences of "backslash" in comments that should have been
1825        "backspace".
1827    10. Remove two redundant lines of code that can never be obeyed (their function
1828        was moved elsewhere).
1830    11. The program that makes PCRE's Unicode character property table had a bug
1831        which caused it to generate incorrect table entries for sequences of
1832        characters that have the same character type, but are in different scripts.
1833        It amalgamated them into a single range, with the script of the first of
1834        them. In other words, some characters were in the wrong script. There were
1835        thirteen such cases, affecting characters in the following ranges:
1837          U+002b0 - U+002c1
1838          U+0060c - U+0060d
1839          U+0061e - U+00612
1840          U+0064b - U+0065e
1841          U+0074d - U+0076d
1842          U+01800 - U+01805
1843          U+01d00 - U+01d77
1844          U+01d9b - U+01dbf
1845          U+0200b - U+0200f
1846          U+030fc - U+030fe
1847          U+03260 - U+0327f
1848          U+0fb46 - U+0fbb1
1849          U+10450 - U+1049d
1851    12. The -o option (show only the matching part of a line) for pcregrep was not
1852        compatible with GNU grep in that, if there was more than one match in a
1853        line, it showed only the first of them. It now behaves in the same way as
1854        GNU grep.
1856    13. If the -o and -v options were combined for pcregrep, it printed a blank
1857        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1858        does the same. The return code can be used to tell if there were any
1859        non-matching lines.
1861    14. Added --file-offsets and --line-offsets to pcregrep.
1863    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1864        infinitely looping recursion. The bug was that positive lookaheads were not
1865        being skipped when checking for a possible empty match (negative lookaheads
1866        and both kinds of lookbehind were skipped).
1868    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1869        inclusion of <windows.h> to before rather than after the definition of
1870        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1872    17. Specifying a possessive quantifier with a specific limit for a Unicode
1873        character property caused pcre_compile() to compile bad code, which led at
1874        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1875        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1876        caused the error; without that there was no problem.
1878    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1880    19. Added --enable-pcretest-libreadline.
1882    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1883        RE::GlobalReplace(). As a result, the number of replacements returned was
1884        double what it should be. I removed one of the increments, but Craig sent a
1885        later patch that removed the other one (the right fix) and added unit tests
1886        that check the return values (which was not done before).
1888    21. Several CMake things:
1890        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1891            the names libpcre and libpcreposix, not just pcre and pcreposix.
1893        (2) The above change means that pcretest and pcregrep are now correctly
1894            linked with the newly-built libraries, not previously installed ones.
1898    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1899        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1900        UTF-8 newline character). The key issue is that the pattern starts .*;
1901        this means that the match must be either at the beginning, or after a
1902        newline. The bug was in the code for advancing after a failed match and
1903        checking that the new position followed a newline. It was not taking
1904        account of UTF-8 characters correctly.
1906    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1907        character classes. PCRE was not treating the sequence [:...:] as a
1908        character class unless the ... were all letters. Perl, however, seems to
1909        allow any characters between [: and :], though of course it rejects as
1910        unknown any "names" that contain non-letters, because all the known class
1911        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1912        for example, whereas PCRE did not - it did not recognize a POSIX character
1913        class. This seemed a bit dangerous, so the code has been changed to be
1914        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1915        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1916        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1917        Perl does, and where it didn't before.
1919    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1920        Windows environments %n is disabled by default.
1923    Version 7.4 21-Sep-07
1924    ---------------------
1926    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1927        means that a class such as [\s] counted as "explicit reference to CR or
1928        LF". That isn't really right - the whole point of the change was to try to
1929        help when there was an actual mention of one of the two characters. So now
1930        the change happens only if \r or \n (or a literal CR or LF) character is
1931        encountered.
1933    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1934        of both had grown to the point where there were only 3 bits left.
1935        Fortunately, there was spare space in the data structure, and so I have
1936        moved the internal flags into a new 16-bit field to free up more option
1937        bits.
1939    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1940        but did not set the internal JCHANGED flag - either of these is enough to
1941        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1942        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1943        start sets both bits.
1945    4.  Added options (at build time, compile time, exec time) to change \R from
1946        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1948    5.  doc/pcresyntax.html was missing from the distribution.
1950    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1951        compatibility, even though it is no longer used.
1953    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1954        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1955        windows.h file is present (where different names are used). [This was
1956        reversed later after testing - see 16 below.]
1958    8.  Changed all #include <config.h> to #include "config.h". There were also
1959        some further <pcre.h> cases that I changed to "pcre.h".
1961    9.  When pcregrep was used with the --colour option, it missed the line ending
1962        sequence off the lines that it output.
1964    10. It was pointed out to me that arrays of string pointers cause lots of
1965        relocations when a shared library is dynamically loaded. A technique of
1966        using a single long string with a table of offsets can drastically reduce
1967        these. I have refactored PCRE in four places to do this. The result is
1968        dramatic:
1970          Originally:                          290
1971          After changing UCP table:            187
1972          After changing error message table:   43
1973          After changing table of "verbs"       36
1974          After changing table of Posix names   22
1976        Thanks to the folks working on Gregex for glib for this insight.
1978    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1979        unicode-properties was also set.
1981    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1983    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1984        checked only for CRLF.
1986    14. Added casts to pcretest.c to avoid compiler warnings.
1988    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1990    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1991        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1992        entirely. This removes changes made in 7 above.
1994    17. The CMake files have been updated, and there is now more information about
1995        building with CMake in the NON-UNIX-USE document.
1998    Version 7.3 28-Aug-07
1999    ---------------------
2001     1. In the rejigging of the build system that eventually resulted in 7.1, the
2002        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
2003        brackets there is not right, since it causes compilers to look for an
2004        installed pcre.h, not the version that is in the source that is being
2005        compiled (which of course may be different). I have changed it back to:
2007          #include "pcre.h"
2009        I have a vague recollection that the change was concerned with compiling in
2010        different directories, but in the new build system, that is taken care of
2011        by the VPATH setting the Makefile.
2013     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
2014        when the subject happened to end in the byte 0x85 (e.g. if the last
2015        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
2016        characters but of course it shouldn't be taken as a newline when it is part
2017        of another character. The bug was that, for an unlimited repeat of . in
2018        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
2019        characters when looking for a newline.
2021     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
2023     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
2024        in debug output.
2026     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
2027        long printing in the pcrecpp unittest when running under MinGW.
2029     6. ESC_K was left out of the EBCDIC table.
2031     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
2032        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
2033        limit also applies to "virtual nesting" when a pattern is recursive, and in
2034        this case 1000 isn't so big. I have been able to remove this limit at the
2035        expense of backing off one optimization in certain circumstances. Normally,
2036        when pcre_exec() would call its internal match() function recursively and
2037        immediately return the result unconditionally, it uses a "tail recursion"
2038        feature to save stack. However, when a subpattern that can match an empty
2039        string has an unlimited repetition quantifier, it no longer makes this
2040        optimization. That gives it a stack frame in which to save the data for
2041        checking that an empty string has been matched. Previously this was taken
2042        from the 1000-entry workspace that had been reserved. So now there is no
2043        explicit limit, but more stack is used.
2045     8. Applied Daniel's patches to solve problems with the import/export magic
2046        syntax that is required for Windows, and which was going wrong for the
2047        pcreposix and pcrecpp parts of the library. These were overlooked when this
2048        problem was solved for the main library.
2050     9. There were some crude static tests to avoid integer overflow when computing
2051        the size of patterns that contain repeated groups with explicit upper
2052        limits. As the maximum quantifier is 65535, the maximum group length was
2053        set at 30,000 so that the product of these two numbers did not overflow a
2054        32-bit integer. However, it turns out that people want to use groups that
2055        are longer than 30,000 bytes (though not repeat them that many times).
2056        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
2057        made it possible to implement the integer overflow checks in a much more
2058        dynamic way, which I have now done. The artificial limitation on group
2059        length has been removed - we now have only the limit on the total length of
2060        the compiled pattern, which depends on the LINK_SIZE setting.
2062    10. Fixed a bug in the documentation for get/copy named substring when
2063        duplicate names are permitted. If none of the named substrings are set, the
2064        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
2065        empty string.
2067    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
2068        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
2069        because the ] is interpreted as the first data character and the
2070        terminating ] is not found. PCRE has been made compatible with Perl in this
2071        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
2072        cause memory overwriting.
2074    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
2075        string has been matched (to stop an infinite loop). It was not recognizing
2076        a conditional subpattern that could match an empty string if that
2077        subpattern was within another subpattern. For example, it looped when
2078        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
2079        condition was not nested. This bug has been fixed.
2081    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
2082        past the start of the subject in the presence of bytes with the top bit
2083        set, for example "\x8aBCD".
2085    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
2086        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
2088    14. Optimized (?!) to (*FAIL).
2090    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
2091        This restricts code points to be within the range 0 to 0x10FFFF, excluding
2092        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
2093        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
2094        does: it's just the validity check that is more restrictive.
2096    16. Inserted checks for integer overflows during escape sequence (backslash)
2097        processing, and also fixed erroneous offset values for syntax errors during
2098        backslash processing.
2100    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
2101        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
2103    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
2104        caused an overrun.
2106    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
2107        something other than just ASCII characters) inside a group that had an
2108        unlimited repeat caused a loop at compile time (while checking to see
2109        whether the group could match an empty string).
2111    20. Debugging a pattern containing \p or \P could cause a crash. For example,
2112        [\P{Any}] did so. (Error in the code for printing property names.)
2114    21. An orphan \E inside a character class could cause a crash.
2116    22. A repeated capturing bracket such as (A)? could cause a wild memory
2117        reference during compilation.
2119    23. There are several functions in pcre_compile() that scan along a compiled
2120        expression for various reasons (e.g. to see if it's fixed length for look
2121        behind). There were bugs in these functions when a repeated \p or \P was
2122        present in the pattern. These operators have additional parameters compared
2123        with \d, etc, and these were not being taken into account when moving along
2124        the compiled data. Specifically:
2126        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2127            length.
2129        (b) An item such as \pL+ within a repeated group could cause crashes or
2130            loops.
2132        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2133            "reference to non-existent subpattern" error.
2135        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2137    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2138        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2140    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2142    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2143        character were causing crashes (broken optimization).
2145    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2146        \p or \P) caused a compile-time loop.
2148    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2149        break. For example, the unstudied pattern [\r\n]A does not match the string
2150        "\r\nA" because change 7.0/46 below moves the current point on by two
2151        characters after failing to match at the start. However, the pattern \nA
2152        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2153        the same is true. There doesn't seem any very clean way out of this, but
2154        what I have chosen to do makes the common cases work: PCRE now takes note
2155        of whether there can be an explicit match for \r or \n anywhere in the
2156        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2157        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2158        pattern has explicit CR or LF references.
2160    29. Added (*CR) etc for changing newline setting at start of pattern.
2163    Version 7.2 19-Jun-07
2164    ---------------------
2166     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
2167        which is apparently normally available under Windows.
2169     2. Re-jig the pcregrep tests with different newline settings in an attempt
2170        to make them independent of the local environment's newline setting.
2172     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
2174     4. Some of the "internals" tests were previously cut out when the link size
2175        was not 2, because the output contained actual offsets. The recent new
2176        "Z" feature of pcretest means that these can be cut out, making the tests
2177        usable with all link sizes.
2179     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
2180        stack recursion. This gives a massive performance boost under BSD, but just
2181        a small improvement under Linux. However, it saves one field in the frame
2182        in all cases.
2184     6. Added more features from the forthcoming Perl 5.10:
2186        (a) (?-n) (where n is a string of digits) is a relative subroutine or
2187            recursion call. It refers to the nth most recently opened parentheses.
2189        (b) (?+n) is also a relative subroutine call; it refers to the nth next
2190            to be opened parentheses.
2192        (c) Conditions that refer to capturing parentheses can be specified
2193            relatively, for example, (?(-2)... or (?(+3)...
2195        (d) \K resets the start of the current match so that everything before
2196            is not part of it.
2198        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2200        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2201            reference syntax.
2203        (g) (?| introduces a group in which the numbering of parentheses in each
2204            alternative starts with the same number.
2206        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2208     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2211     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2212        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2213        for detecting groups that can match an empty string.
2215     9. A pattern with a very large number of alternatives (more than several
2216        hundred) was running out of internal workspace during the pre-compile
2217        phase, where pcre_compile() figures out how much memory will be needed. A
2218        bit of new cunning has reduced the workspace needed for groups with
2219        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2220        workspace instead of running out of the 4096 that are available.
2222    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2224    11. Applied patch from Google to remove an optimization that didn't quite work.
2225        The report of the bug said:
2227          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2228          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2229          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2231    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2232        it matched the wrong number of bytes.
2235    Version 7.1 24-Apr-07
2236    ---------------------
2238     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
2239        that is more "standard", making use of automake and other Autotools. There
2240        is some re-arrangement of the files and adjustment of comments consequent
2241        on this.
2243     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
2244        for recursive directory scanning broke on some systems because the files
2245        are not scanned in any specific order and on different systems the order
2246        was different. A call to "sort" has been inserted into RunGrepTest for the
2247        approprate test as a short-term fix. In the longer term there may be an
2248        alternative.
2250     3. I had an email from Eric Raymond about problems translating some of PCRE's
2251        man pages to HTML (despite the fact that I distribute HTML pages, some
2252        people do their own conversions for various reasons). The problems
2253        concerned the use of low-level troff macros .br and .in. I have therefore
2254        removed all such uses from the man pages (some were redundant, some could
2255        be replaced by .nf/.fi pairs). The 132html script that I use to generate
2256        HTML has been updated to handle .nf/.fi and to complain if it encounters
2257        .br or .in.
2259     4. Updated comments in configure.ac that get placed in config.h.in and also
2260        arranged for config.h to be included in the distribution, with the name
2261        config.h.generic, for the benefit of those who have to compile without
2262        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
2264     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
2265        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
2266        makevp.bat for the current PCRE, using the additional files
2267        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
2269     6. A Windows user reported a minor discrepancy with test 2, which turned out
2270        to be caused by a trailing space on an input line that had got lost in his
2271        copy. The trailing space was an accident, so I've just removed it.
2273     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
2274        that is needed.
2276     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
2277        as "const" (a) because they are and (b) because it helps the PHP
2278        maintainers who have recently made a script to detect big data structures
2279        in the php code that should be moved to the .rodata section. I remembered
2280        to update Builducptable as well, so it won't revert if ucptable.h is ever
2281        re-created.
2283     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
2284        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
2285        order to be able to cut out the UTF-8 tables in the latter when UTF-8
2286        support is not required. This saves 1.5-2K of code, which is important in
2287        some applications.
2289        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
2290        so as not to refer to the tables, even though these functions will never be
2291        called when UTF-8 support is disabled. Otherwise there are problems with a
2292        shared library.
2294    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
2296        (a) It was defining its arguments as char * instead of void *.
2298        (b) It was assuming that all moves were upwards in memory; this was true
2299            a long time ago when I wrote it, but is no longer the case.
2301        The emulated memove() is provided for those environments that have neither
2302        memmove() nor bcopy(). I didn't think anyone used it these days, but that
2303        is clearly not the case, as these two bugs were recently reported.
2305    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
2306        and Detrail to create the HTML documentation, the .txt form of the man
2307        pages, and it removes trailing spaces from listed files. It also creates
2308        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
2309        case, it wraps all the #defines with #ifndefs. This script should be run
2310        before "make dist".
2312    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
2313        with Unicode property support.
2315        (a) For a maximizing quantifier, if the two different cases of the
2316            character were of different lengths in their UTF-8 codings (there are
2317            some cases like this - I found 11), and the matching function had to
2318            back up over a mixture of the two cases, it incorrectly assumed they
2319            were both the same length.
2321        (b) When PCRE was configured to use the heap rather than the stack for
2322            recursion during matching, it was not correctly preserving the data for
2323            the other case of a UTF-8 character when checking ahead for a match
2324            while processing a minimizing repeat. If the check also involved
2325            matching a wide character, but failed, corruption could cause an
2326            erroneous result when trying to check for a repeat of the original
2327            character.
2329    13. Some tidying changes to the testing mechanism:
2331        (a) The RunTest script now detects the internal link size and whether there
2332            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
2333            values substituted by "configure". (The RunGrepTest script already did
2334            this for UTF-8.) The configure.ac script no longer substitutes the
2335            relevant variables.
2337        (b) The debugging options /B and /D in pcretest show the compiled bytecode
2338            with length and offset values. This means that the output is different
2339            for different internal link sizes. Test 2 is skipped for link sizes
2340            other than 2 because of this, bypassing the problem. Unfortunately,
2341            there was also a test in test 3 (the locale tests) that used /B and
2342            failed for link sizes other than 2. Rather than cut the whole test out,
2343            I have added a new /Z option to pcretest that replaces the length and
2344            offset values with spaces. This is now used to make test 3 independent
2345            of link size. (Test 2 will be tidied up later.)
2347    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
2348        segmentation fault instead of returning the appropriate error message.
2350    15. In multiline mode when the newline sequence was set to "any", the pattern
2351        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
2352        This doesn't seem right; it now treats the CRLF combination as the line
2353        ending, and so does not match in that case. It's only a pattern such as ^$
2354        that would hit this one: something like ^ABC$ would have failed after \r
2355        and then tried again after \r\n.
2357    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
2358        in an attempt to make files that differ only in their line terminators
2359        compare equal. This works on Linux.
2361    17. Under certain error circumstances pcregrep might try to free random memory
2362        as it exited. This is now fixed, thanks to valgrind.
2364    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
2365        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
2366        was because its rules for how to advance for /g after matching an empty
2367        string at the end of a line did not allow for this case. They now check for
2368        it specially.
2370    20. pcretest is supposed to handle patterns and data of any length, by
2371        extending its buffers when necessary. It was getting this wrong when the
2372        buffer for a data line had to be extended.
2374    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
2375        CRLF as a newline sequence.
2377    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
2378        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
2379        I have nevertheless tidied it up.
2381    23. Added some casts to kill warnings from HP-UX ia64 compiler.
2383    24. Added a man page for pcre-config.
2386    Version 7.0 19-Dec-06
2387    ---------------------
2389     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
2390        moving to gcc 4.1.1.
2392     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
2393        sys/time.h, which is documented as needed for this function. It doesn't
2394        seem to matter on Linux, but it showed up on some releases of OS X.
2396     3. It seems that there are systems where bytes whose values are greater than
2397        127 match isprint() in the "C" locale. The "C" locale should be the
2398        default when a C program starts up. In most systems, only ASCII printing
2399        characters match isprint(). This difference caused the output from pcretest
2400        to vary, making some of the tests fail. I have changed pcretest so that:
2402        (a) When it is outputting text in the compiled version of a pattern, bytes
2403            other than 32-126 are always shown as hex escapes.
2405        (b) When it is outputting text that is a matched part of a subject string,
2406            it does the same, unless a different locale has been set for the match
2407            (using the /L modifier). In this case, it uses isprint() to decide.
2409     4. Fixed a major bug that caused incorrect computation of the amount of memory
2410        required for a compiled pattern when options that changed within the
2411        pattern affected the logic of the preliminary scan that determines the
2412        length. The relevant options are -x, and -i in UTF-8 mode. The result was
2413        that the computed length was too small. The symptoms of this bug were
2414        either the PCRE error "internal error: code overflow" from pcre_compile(),
2415        or a glibc crash with a message such as "pcretest: free(): invalid next
2416        size (fast)". Examples of patterns that provoked this bug (shown in
2417        pcretest format) are:
2419          /(?-x: )/x
2420          /(?x)(?-x: \s*#\s*)/
2421          /((?i)[\x{c0}])/8
2422          /(?i:[\x{c0}])/8
2424        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
2425        is now done differently.
2427     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
2428        wrapper classes; (b) implement a new function in the C++ scanner that is
2429        more efficient than the old way of doing things because it avoids levels of
2430        recursion in the regex matching; (c) add a paragraph to the documentation
2431        for the FullMatch() function.
2433     6. The escape sequence \n was being treated as whatever was defined as
2434        "newline". Not only was this contrary to the documentation, which states
2435        that \n is character 10 (hex 0A), but it also went horribly wrong when
2436        "newline" was defined as CRLF. This has been fixed.
2438     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
2439        was being set to -1 for the "end of line" case (supposedly a value that no
2440        character can have). Though this value is never used (the check for end of
2441        line is "zero bytes in current character"), it caused compiler complaints.
2442        I've changed it to 0xffffffff.
2444     8. In pcre_version.c, the version string was being built by a sequence of
2445        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
2446        string (as it is for production releases) called a macro with an empty
2447        argument. The C standard says the result of this is undefined. The gcc
2448        compiler treats it as an empty string (which was what was wanted) but it is
2449        reported that Visual C gives an error. The source has been hacked around to
2450        avoid this problem.
2452     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
2453        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
2454        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
2455        of them did).
2457    10. Originally, pcretest opened its input and output without "b"; then I was
2458        told that "b" was needed in some environments, so it was added for release
2459        5.0 to both the input and output. (It makes no difference on Unix-like
2460        systems.) Later I was told that it is wrong for the input on Windows. I've
2461        now abstracted the modes into two macros, to make it easier to fiddle with
2462        them, and removed "b" from the input mode under Windows.
2464    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
2466    12. Added -help and --help to pcretest as an official way of being reminded
2467        of the options.
2469    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
2470        and pcrecpp.cc because they annoy compilers at high warning levels.
2472    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
2474    15. Fixed an occurrence of == in configure.ac that should have been = (shell
2475        scripts are not C programs :-) and which was not noticed because it works
2476        on Linux.
2478    16. pcretest is supposed to handle any length of pattern and data line (as one
2479        line or as a continued sequence of lines) by extending its input buffer if
2480        necessary. This feature was broken for very long pattern lines, leading to
2481        a string of junk being passed to pcre_compile() if the pattern was longer
2482        than about 50K.
2484    17. I have done a major re-factoring of the way pcre_compile() computes the
2485        amount of memory needed for a compiled pattern. Previously, there was code
2486        that made a preliminary scan of the pattern in order to do this. That was
2487        OK when PCRE was new, but as the facilities have expanded, it has become
2488        harder and harder to keep it in step with the real compile phase, and there
2489        have been a number of bugs (see for example, 4 above). I have now found a
2490        cunning way of running the real compile function in a "fake" mode that
2491        enables it to compute how much memory it would need, while actually only
2492        ever using a few hundred bytes of working memory and without too many
2493        tests of the mode. This should make future maintenance and development
2494        easier. A side effect of this work is that the limit of 200 on the nesting
2495        depth of parentheses has been removed (though this was never a serious
2496        limitation, I suspect). However, there is a downside: pcre_compile() now
2497        runs more slowly than before (30% or more, depending on the pattern). I
2498        hope this isn't a big issue. There is no effect on runtime performance.
2500    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
2501        newline (only possible for the last line of a file) and it was a
2502        pattern that set a locale (followed by /Lsomething), pcretest crashed.
2504    19. Added additional timing features to pcretest. (1) The -tm option now times
2505        matching only, not compiling. (2) Both -t and -tm can be followed, as a
2506        separate command line item, by a number that specifies the number of
2507        repeats to use when timing. The default is 50000; this gives better
2508        precision, but takes uncomfortably long for very large patterns.
2510    20. Extended pcre_study() to be more clever in cases where a branch of a
2511        subpattern has no definite first character. For example, (a*|b*)[cd] would
2512        previously give no result from pcre_study(). Now it recognizes that the
2513        first character must be a, b, c, or d.
2515    21. There was an incorrect error "recursive call could loop indefinitely" if
2516        a subpattern (or the entire pattern) that was being tested for matching an
2517        empty string contained only one non-empty item after a nested subpattern.
2518        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
2519        incorrectly, because the \d was being skipped in the check.
2521    22. The pcretest program now has a new pattern option /B and a command line
2522        option -b, which is equivalent to adding /B to every pattern. This causes
2523        it to show the compiled bytecode, without the additional information that
2524        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
2525        is the same as /B/I).
2527    23. A new optimization is now able automatically to treat some sequences such
2528        as a*b as a*+b. More specifically, if something simple (such as a character
2529        or a simple class like \d) has an unlimited quantifier, and is followed by
2530        something that cannot possibly match the quantified thing, the quantifier
2531        is automatically "possessified".
2533    24. A recursive reference to a subpattern whose number was greater than 39
2534        went wrong under certain circumstances in UTF-8 mode. This bug could also
2535        have affected the operation of pcre_study().
2537    25. Realized that a little bit of performance could be had by replacing
2538        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
2540    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
2542    27. Possessive quantifiers such as a++ were previously implemented by turning
2543        them into atomic groups such as ($>a+). Now they have their own opcodes,
2544        which improves performance. This includes the automatically created ones
2545        from 23 above.
2547    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
2548        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
2549        the first matched character to be a colon. This applied both to named and
2550        numbered groups.
2552    29. The ucpinternal.h header file was missing its idempotency #ifdef.
2554    30. I was sent a "project" file called libpcre.a.dev which I understand makes
2555        building PCRE on Windows easier, so I have included it in the distribution.
2557    31. There is now a check in pcretest against a ridiculously large number being
2558        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
2559        loop, the loop is abandoned.
2561    32. Forward references to subpatterns in conditions such as (?(2)...) where
2562        subpattern 2 is defined later cause pcre_compile() to search forwards in
2563        the pattern for the relevant set of parentheses. This search went wrong
2564        when there were unescaped parentheses in a character class, parentheses
2565        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2567    33. "Subroutine" calls and backreferences were previously restricted to
2568        referencing subpatterns earlier in the regex. This restriction has now
2569        been removed.
2571    34. Added a number of extra features that are going to be in Perl 5.10. On the
2572        whole, these are just syntactic alternatives for features that PCRE had
2573        previously implemented using the Python syntax or my own invention. The
2574        other formats are all retained for compatibility.
2576        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2577            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2578            also .NET compatible.
2580        (b) A recursion or subroutine call to a named group can now be defined as
2581            (?&name) as well as (?P>name).
2583        (c) A backreference to a named group can now be defined as \k<name> or
2584            \k'name' as well as (?P=name). The new forms, as well as being in Perl
2585            5.10, are also .NET compatible.
2587        (d) A conditional reference to a named group can now use the syntax
2588            (?(<name>) or (?('name') as well as (?(name).
2590        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2591            groups (named and numbered) that are never evaluated inline, but can be
2592            called as "subroutines" from elsewhere. In effect, the DEFINE condition
2593            is always false. There may be only one alternative in such a group.
2595        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2596            as the simple (?(R). The condition is true only if the most recent
2597            recursion is that of the given number or name. It does not search out
2598            through the entire recursion stack.
2600        (g) The escape \gN or \g{N} has been added, where N is a positive or
2601            negative number, specifying an absolute or relative reference.
2603    35. Tidied to get rid of some further signed/unsigned compiler warnings and
2604        some "unreachable code" warnings.
2606    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2607        things, this adds five new scripts.
2609    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2610        There were also incompatibilities regarding the handling of \Q..\E inside
2611        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2612        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2614    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2615        matches an empty string, and forcibly breaks the loop. There were bugs in
2616        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2617        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2618        separate and independent bugs (that affected different cases) have been
2619        fixed.
2621    39. Refactored the code to abolish the use of different opcodes for small
2622        capturing bracket numbers. This is a tidy that I avoided doing when I
2623        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2624        The new approach is not only tidier, it makes it possible to reduce the
2625        memory needed to fix the previous bug (38).
2627    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2628        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2629        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2630        mode.
2632    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2633        report.
2635    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2636        copy construction and assignment in the C++ wrapper.
2638    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2639        couple of bugs that could have given wrong results in the "--newline=crlf"
2640        case.
2642    44. Added a number of casts and did some reorganization of signed/unsigned int
2643        variables following suggestions from Dair Grant. Also renamed the variable
2644        "this" as "item" because it is a C++ keyword.
2646    45. Arranged for dftables to add
2648          #include "pcre_internal.h"
2650        to pcre_chartables.c because without it, gcc 4.x may remove the array
2651        definition from the final binary if PCRE is built into a static library and
2652        dead code stripping is activated.
2654    46. For an unanchored pattern, if a match attempt fails at the start of a
2655        newline sequence, and the newline setting is CRLF or ANY, and the next two
2656        characters are CRLF, advance by two characters instead of one.
2659  Version 6.7 04-Jul-06  Version 6.7 04-Jul-06
2660  ---------------------  ---------------------

Removed from v.91  
changed lines
  Added in v.1050

  ViewVC Help
Powered by ViewVC 1.1.5