ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 335 by ph10, Sat Apr 12 14:36:14 2008 UTC revision 1236 by ph10, Sun Dec 30 17:09:59 2012 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4  Version 7.7 05-Mar-08  Version 8.33 xx-xxxx-201x
5    -------------------------
7    1.  Added 'U' to some constants that are compared to unsigned integers, to
8        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
9        variables that are added to signed variables, to ensure the result is
10        signed and can be negated.
12    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
13        CMake config files.
15    3.  Revise the creation of config.h.generic so that all boolean macros are
16        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
17        overriding via -D on the command line possible.
19    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
20        to unsigned int is reported to make a quite noticeable speed difference in
21        a specific Windows environment. Testing on Linux did also appear to show
22        some benefit (and it is clearly not harmful). Also fixed the definition of
23        Xop which should be unsigned.
26    Version 8.32 30-November-2012
27    -----------------------------
29    1.  Improved JIT compiler optimizations for first character search and single
30        character iterators.
32    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
33        Patch by Daniel Richard G.
35    3.  Single character iterator optimizations in the JIT compiler.
37    4.  Improved JIT compiler optimizations for character ranges.
39    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
40        Reported by Giuseppe D'Angelo.
42    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
43        of a line, was being set incorrectly in cases where .* appeared inside
44        atomic brackets at the start of a pattern, or where there was a subsequent
45        *PRUNE or *SKIP.
47    7.  Improved instruction cache flush for POWER/PowerPC.
48        Patch by Daniel Richard G.
50    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
51        grep:
53        (a) There is now no limit to the number of patterns to be matched.
55        (b) An error is given if a pattern is too long.
57        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
58            are now supported.
60        (d) --exclude-from and --include-from (multiple use) have been added.
62        (e) Exclusions and inclusions now apply to all files and directories, not
63            just to those obtained from scanning a directory recursively.
65        (f) Multiple uses of -f and --file-list are now supported.
67        (g) In a Windows environment, the default for -d has been changed from
68            "read" (the GNU grep default) to "skip", because otherwise the presence
69            of a directory in the file list provokes an error.
71        (h) The documentation has been revised and clarified in places.
73    9.  Improve the matching speed of capturing brackets.
75    10. Changed the meaning of \X so that it now matches a Unicode extended
76        grapheme cluster.
78    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
79        out POSIX threads when JIT support is configured.
81    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
83    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
84        structure could go wrong in environments where size_t is not the same size
85        as int.
87    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
88        set.
90    15. The EBCDIC support had decayed; later updates to the code had included
91        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
92        general tidy up of EBCDIC-related issues, and the documentation was also
93        not quite right. There is now a test that can be run on ASCII systems to
94        check some of the EBCDIC-related things (but is it not a full test).
96    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
97        in a small tidy to the code.
99    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
101    18. If the --only-matching (-o) option in pcregrep is specified multiple
102        times, each one causes appropriate output. For example, -o1 -o2 outputs the
103        substrings matched by the 1st and 2nd capturing parentheses. A separating
104        string can be specified by --om-separator (default empty).
106    19. Improving the first n character searches.
108    20. Turn case lists for horizontal and vertical white space into macros so that
109        they are defined only once.
111    21. This set of changes together give more compatible Unicode case-folding
112        behaviour for characters that have more than one other case when UCP
113        support is available.
115        (a) The Unicode property table now has offsets into a new table of sets of
116            three or more characters that are case-equivalent. The MultiStage2.py
117            script that generates these tables (the pcre_ucd.c file) now scans
118            CaseFolding.txt instead of UnicodeData.txt for character case
119            information.
121        (b) The code for adding characters or ranges of characters to a character
122            class has been abstracted into a generalized function that also handles
123            case-independence. In UTF-mode with UCP support, this uses the new data
124            to handle characters with more than one other case.
126        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
127            whose other case is greater than 256 are now correctly matched
128            caselessly. Previously, the high codepoint matched the low one, but not
129            vice versa.
131        (d) The processing of \h, \H, \v, and \ in character classes now makes use
132            of the new class addition function, using character lists defined as
133            macros alongside the case definitions of 20 above.
135        (e) Caseless back references now work with characters that have more than
136            one other case.
138        (f) General caseless matching of characters with more than one other case
139            is supported.
141    22. Unicode character properties were updated from Unicode 6.2.0
143    23. Improved CMake support under Windows. Patch by Daniel Richard G.
145    24. Add support for 32-bit character strings, and UTF-32
147    25. Major JIT compiler update (code refactoring and bugfixing).
148        Experimental Sparc 32 support is added.
150    26. Applied a modified version of Daniel Richard G's patch to create
151        pcre.h.generic and config.h.generic by "make" instead of in the
152        PrepareRelease script.
154    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
155        pcre_compile.c when checking for a zero character.
157    28. Introducing a native interface for JIT. Through this interface, the compiled
158        machine code can be directly executed. The purpose of this interface is to
159        provide fast pattern matching, so several sanity checks are not performed.
160        However, feature tests are still performed. The new interface provides
161        1.4x speedup compared to the old one.
163    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
164        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
165        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
167    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
168        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
169        the "old" RFC 2279). Instead, it ended up passing a negative length to
170        pcre_exec().
172    31. Add support for GCC's visibility feature to hide internal functions.
174    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
175        "unknown -C option" after outputting 0 or 1.
177    33. There is now support for generating a code coverage report for the test
178        suite in environments where gcc is the compiler and lcov is installed. This
179        is mainly for the benefit of the developers.
181    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
182        unaddressable using valgrind annotations, allowing valgrind to detect
183        invalid memory accesses. This is mainly for the benefit of the developers.
185    25. (*UTF) can now be used to start a pattern in any of the three libraries.
187    26. Give configure error if --enable-cpp but no C++ compiler found.
190    Version 8.31 06-July-2012
191    -------------------------
193    1.  Fixing a wrong JIT test case and some compiler warnings.
195    2.  Removed a bashism from the RunTest script.
197    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
198        to unsigned type, result still unsigned" that was given by an MS compiler
199        on encountering the code "-sizeof(xxx)".
201    4.  Partial matching support is added to the JIT compiler.
203    5.  Fixed several bugs concerned with partial matching of items that consist
204        of more than one character:
206        (a) /^(..)\1/ did not partially match "aba" because checking references was
207            done on an "all or nothing" basis. This also applied to repeated
208            references.
210        (b) \R did not give a hard partial match if \r was found at the end of the
211            subject.
213        (c) \X did not give a hard partial match after matching one or more
214            characters at the end of the subject.
216        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
217            a partial match for the string "\r".
219        (e) When newline was set to CRLF, the metacharacter "." did not recognize
220            a partial match for a CR character at the end of the subject string.
222    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
223        running pcretest, the text "(JIT)" added to the output whenever JIT is
224        actually used to run the match.
226    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
227        or /S+[+] with a digit between 1 and 7.
229    8.  OP_NOT now supports any UTF character not just single-byte ones.
231    9.  (*MARK) control verb is now supported by the JIT compiler.
233    10. The command "./RunTest list" lists the available tests without actually
234        running any of them. (Because I keep forgetting what they all are.)
238    12. Applied a (slightly modified) user-supplied patch that improves performance
239        when the heap is used for recursion (compiled with --disable-stack-for-
240        recursion). Instead of malloc and free for each heap frame each time a
241        logical recursion happens, frames are retained on a chain and re-used where
242        possible. This sometimes gives as much as 30% improvement.
244    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
245        call.
247    14. As documented, (*COMMIT) is now confined to within a positive assertion.
249    15. It is now possible to link pcretest with libedit as an alternative to
250        libreadline.
252    16. (*COMMIT) control verb is now supported by the JIT compiler.
254    17. The Unicode data tables have been updated to Unicode 6.1.0.
256    18. Added --file-list option to pcregrep.
258    19. Added binary file support to pcregrep, including the -a, --binary-files,
259        -I, and --text options.
261    20. The madvise function is renamed for posix_madvise for QNX compatibility
262        reasons. Fixed by Giuseppe D'Angelo.
264    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
266    22. Changed the option for creating symbolic links for 16-bit man pages from
267        -s to -sf so that re-installing does not cause issues.
269    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
271    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
272        matches in certain environments (the workspace was not being correctly
273        retained). Also added to pcre_dfa_exec() a simple plausibility check on
274        some of the workspace data at the beginning of a restart.
276    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
277        was not doing so when it should - probably a typo introduced by SVN 528
278        (change 8.10/14).
280    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
281        \w+ when the character tables indicated that \x{c4} was a word character.
282        There were several related cases, all because the tests for doing a table
283        lookup were testing for characters less than 127 instead of 255.
285    27. If a pattern contains capturing parentheses that are not used in a match,
286        their slots in the ovector are set to -1. For those that are higher than
287        any matched groups, this happens at the end of processing. In the case when
288        there were back references that the ovector was too small to contain
289        (causing temporary malloc'd memory to be used during matching), and the
290        highest capturing number was not used, memory off the end of the ovector
291        was incorrectly being set to -1. (It was using the size of the temporary
292        memory instead of the true size.)
294    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
295        ovector size, it uses memory at the end of the block that it has got.
297    29. Check for an overlong MARK name and give an error at compile time. The
298        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
300    30. JIT compiler update.
302    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
303        Rill for the patch.
305    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
307    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
309    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
310        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
311        was enabled.
313    35. Improve JIT code generation for greedy plus quantifier.
315    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
316        1 to "aa" instead of to an empty string. The bug affected repeated groups
317        that could potentially match an empty string.
319    37. Optimizing single character iterators in JIT.
321    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
322        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
323        codepoints that are too big for the mode are faulted, and in a UTF mode,
324        disallowed codepoints are also faulted.
326    39. If PCRE was compiled with UTF support, in three places in the DFA
327        matcher there was code that should only have been obeyed in UTF mode, but
328        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
329        processing when bytes with values greater than 127 were present. In 16-bit
330        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
331        both cases the values are those that cannot be the first data item in a UTF
332        character. The three items that might have provoked this were recursions,
333        possessively repeated groups, and atomic groups.
335    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
336        and pcregrep, because some OS require shared objects to be explicitly
337        passed to ld, causing the link step to fail if they are not.
339    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
340        mode, patterns that started with \h* or \R* might be incorrectly matched.
343    Version 8.30 04-February-2012
344    -----------------------------
346    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
347        name is defined in ctype.h.
349    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
350        only in quite long subpatterns.
352    3.  Removed the function pcre_info(), which has been obsolete and deprecated
353        since it was replaced by pcre_fullinfo() in February 2000.
355    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
356        match a (*MARK), and the match failed at the start of the subject, a
357        reference to memory before the start of the subject could occur. This bug
358        was introduced by fix 17 of release 8.21.
360    5.  A reference to an unset group with zero minimum repetition was giving
361        totally wrong answers (in non-JavaScript-compatibility mode). For example,
362        /(another)?(\1?)test/ matched against "hello world test". This bug was
363        introduced in release 8.13.
365    6.  Add support for 16-bit character strings (a large amount of work involving
366        many changes and refactorings).
368    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
369        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
370        from a file.
372    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
373        rounding is not applied in this particular case).
375    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
376        if they appear, or are escaped, in patterns.
378    10. Get rid of a number of -Wunused-but-set-variable warnings.
380    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
381        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
382        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
383        also returns the mark "x". This bug applied to capturing parentheses,
384        non-capturing parentheses, and atomic parentheses. It also applied to some
385        assertions.
387    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
388        information out of configure.ac instead of relying on pcre.h.generic, which
389        is not stored in the repository.
391    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
392        -lreadline.
394    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
396    15. Applied Graycode's patch to put the top-level frame on the stack rather
397        than the heap when not using the stack for recursion. This gives a
398        performance improvement in many cases when recursion is not deep.
400    16. Experimental code added to "pcretest -C" to output the stack frame size.
403    Version 8.21 12-Dec-2011
404    ------------------------
406    1.  Updating the JIT compiler.
408    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
409        are added as well.
411    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
412        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
413        calling _pcre_jit_exec. Some extra comments are added.
415    4.  (*MARK) settings inside atomic groups that do not contain any capturing
416        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
417        was introduced by change 18 for 8.20.
419    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
420        ECMA-262 standard.
422    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
423        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
424        This bug was probably introduced by change 9 of 8.13.
426    7.  While fixing 6 above, I noticed that a number of other items were being
427        incorrectly rejected as "not fixed length". This arose partly because newer
428        opcodes had not been added to the fixed-length checking code. I have (a)
429        corrected the bug and added tests for these items, and (b) arranged for an
430        error to occur if an unknown opcode is encountered while checking for fixed
431        length instead of just assuming "not fixed length". The items that were
432        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
433        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
434        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
436    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
437        being incorrectly compiled and would have given unpredicatble results.
439    9.  A possessively repeated subpattern with minimum repeat count greater than
440        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
441        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
442        the first (A) could occur when it should not.
444    10. Add a cast and remove a redundant test from the code.
446    11. JIT should use pcre_malloc/pcre_free for allocation.
448    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
449        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
450        is anything other than /usr, -L is still shown).
452    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
454    14. Perl does not support \N without a following name in a [] class; PCRE now
455        also gives an error.
457    15. If a forward reference was repeated with an upper limit of around 2000,
458        it caused the error "internal error: overran compiling workspace". The
459        maximum number of forward references (including repeats) was limited by the
460        internal workspace, and dependent on the LINK_SIZE. The code has been
461        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
462        the default depends on LINK_SIZE. There is a new upper limit (for safety)
463        of around 200,000 forward references. While doing this, I also speeded up
464        the filling in of repeated forward references.
466    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
467        incorrectly expecting the subject to contain another "a" after the start.
469    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
470        in the match, the SKIP should be ignored. This was not happening; instead
471        the SKIP was being treated as NOMATCH. For patterns such as
472        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
473        tested.
475    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
476        now much more compatible with Perl, in particular in cases where the result
477        is a non-match for a non-anchored pattern. For example, if
478        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
479        "m", where previously it did not return a name. A side effect of this
480        change is that for partial matches, the last encountered mark name is
481        returned, as for non matches. A number of tests that were previously not
482        Perl-compatible have been moved into the Perl-compatible test files. The
483        refactoring has had the pleasing side effect of removing one argument from
484        the match() function, thus reducing its stack requirements.
486    19. If the /S+ option was used in pcretest to study a pattern using JIT,
487        subsequent uses of /S (without +) incorrectly behaved like /S+.
489    21. Retrieve executable code size support for the JIT compiler and fixing
490        some warnings.
492    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
493        not work when the shorter character appeared right at the end of the
494        subject string.
496    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
497        systems.
499    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
500        output it when the /M option is used in pcretest.
502    25. The CheckMan script was not being included in the distribution. Also, added
503        an explicit "perl" to run Perl scripts from the PrepareRelease script
504        because this is reportedly needed in Windows.
506    26. If study data was being save in a file and studying had not found a set of
507        "starts with" bytes for the pattern, the data written to the file (though
508        never used) was taken from uninitialized memory and so caused valgrind to
509        complain.
511    27. Updated RunTest.bat as provided by Sheri Pierce.
513    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
515    29. Computation of memory usage for the table of capturing group names was
516        giving an unnecessarily large value.
519    Version 8.20 21-Oct-2011
520    ------------------------
522    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
523        a POSIX class. After further experiments with Perl, which convinced me that
524        Perl has bugs and confusions, a closing square bracket is no longer allowed
525        in a POSIX name. This bug also affected patterns with classes that started
526        with full stops.
528    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
529        captured substring, but while checking the failing first alternative,
530        substring 1 is temporarily captured. If the output vector supplied to
531        pcre_exec() was not big enough for this capture, the yield of the function
532        was still zero ("insufficient space for captured substrings"). This cannot
533        be totally fixed without adding another stack variable, which seems a lot
534        of expense for a edge case. However, I have improved the situation in cases
535        such as /(a)(b)x|abc/ matched against "abc", where the return code
536        indicates that fewer than the maximum number of slots in the ovector have
537        been set.
539    3.  Related to (2) above: when there are more back references in a pattern than
540        slots in the output vector, pcre_exec() uses temporary memory during
541        matching, and copies in the captures as far as possible afterwards. It was
542        using the entire output vector, but this conflicts with the specification
543        that only 2/3 is used for passing back captured substrings. Now it uses
544        only the first 2/3, for compatibility. This is, of course, another edge
545        case.
547    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
548        main code base, and can be used by building with --enable-jit. When this is
549        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
550        runtime --no-jit option is given.
552    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
553        ovector, the return from the function was zero, implying that there were
554        other matches that did not fit. The correct "exactly full" value is now
555        returned.
557    6.  If a subpattern that was called recursively or as a subroutine contained
558        (*PRUNE) or any other control that caused it to give a non-standard return,
559        invalid errors such as "Error -26 (nested recursion at the same subject
560        position)" or even infinite loops could occur.
562    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
563        computing the minimum length on reaching *ACCEPT, and so ended up with the
564        wrong value of 1 rather than 0. Further investigation indicates that
565        computing a minimum subject length in the presence of *ACCEPT is difficult
566        (think back references, subroutine calls), and so I have changed the code
567        so that no minimum is registered for a pattern that contains *ACCEPT.
569    8.  If (*THEN) was present in the first (true) branch of a conditional group,
570        it was not handled as intended. [But see 16 below.]
572    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
573        Sheri Pierce.
575    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
576        the first byte in a match must be "a".
578    11. Change 17 for 8.13 increased the recursion depth for patterns like
579        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
580        pattern contains any instances of (*THEN). If it does not, the old
581        optimizations are restored. It would be nice to do this on a per-group
582        basis, but at the moment that is not feasible.
584    12. In some environments, the output of pcretest -C is CRLF terminated. This
585        broke RunTest's code that checks for the link size. A single white space
586        character after the value is now allowed for.
588    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
589        For "fr", it uses the Windows-specific input and output files.
591    14. If (*THEN) appeared in a group that was called recursively or as a
592        subroutine, it did not work as intended. [But see next item.]
594    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
595        pattern fragments (but not containing any | characters). If A and B are
596        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
597        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
598        D. In other words, Perl considers parentheses that do not contain any |
599        characters to be part of a surrounding alternative, whereas PCRE was
600        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
601        differently. PCRE now behaves in the same way as Perl, except in the case
602        of subroutine/recursion calls such as (?1) which have in any case always
603        been different (but PCRE had them first :-).
605    16. Related to 15 above: Perl does not treat the | in a conditional group as
606        creating alternatives. Such a group is treated in the same way as an
607        ordinary group without any | characters when processing (*THEN). PCRE has
608        been changed to match Perl's behaviour.
610    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
611        RunGrepTest script failed.
613    18. Change 22 for version 13 caused atomic groups to use more stack. This is
614        inevitable for groups that contain captures, but it can lead to a lot of
615        stack use in large patterns. The old behaviour has been restored for atomic
616        groups that do not contain any capturing parentheses.
618    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
619        suppress the check for a minimum subject length at run time. (If it was
620        given to pcre_exec() or pcre_dfa_exec() it did work.)
622    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
623        fail to work when decoding hex characters in data strings in EBCDIC
624        environments.
626    21. It appears that in at least one Mac OS environment, the isxdigit() function
627        is implemented as a macro that evaluates to its argument more than once,
628        contravening the C 90 Standard (I haven't checked a later standard). There
629        was an instance in pcretest which caused it to go wrong when processing
630        \x{...} escapes in subject strings. The has been rewritten to avoid using
631        things like p++ in the argument of isxdigit().
634    Version 8.13 16-Aug-2011
635    ------------------------
637    1.  The Unicode data tables have been updated to Unicode 6.0.0.
639    2.  Two minor typos in pcre_internal.h have been fixed.
641    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
642        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
643        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
645    4.  There were a number of related bugs in the code for matching backrefences
646        caselessly in UTF-8 mode when codes for the characters concerned were
647        different numbers of bytes. For example, U+023A and U+2C65 are an upper
648        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
649        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
650        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
651        2-byte code at the end of the subject (it thought there wasn't enough data
652        left).
654    5.  Comprehensive information about what went wrong is now returned by
655        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
656        as the output vector has at least 2 elements. The offset of the start of
657        the failing character and a reason code are placed in the vector.
659    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
660        now returned is for the first byte of the failing character, instead of the
661        last byte inspected. This is an incompatible change, but I hope it is small
662        enough not to be a problem. It makes the returned offset consistent with
663        pcre_exec() and pcre_dfa_exec().
665    7.  pcretest now gives a text phrase as well as the error number when
666        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
667        failure, the offset and reason code are output.
669    8.  When \R was used with a maximizing quantifier it failed to skip backwards
670        over a \r\n pair if the subsequent match failed. Instead, it just skipped
671        back over a single character (\n). This seems wrong (because it treated the
672        two characters as a single entity when going forwards), conflicts with the
673        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
674        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
675        has been changed.
677    9.  Some internal refactoring has changed the processing so that the handling
678        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
679        time (the PCRE_DOTALL option was changed this way some time ago: version
680        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
681        which was always a bit of a fudge. It also means that there is one less
682        argument for the match() function, which reduces its stack requirements
683        slightly. This change also fixes an incompatibility with Perl: the pattern
684        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
686    10. More internal refactoring has drastically reduced the number of recursive
687        calls to match() for possessively repeated groups such as (abc)++ when
688        using pcre_exec().
690    11. While implementing 10, a number of bugs in the handling of groups were
691        discovered and fixed:
693        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
694        (a|)*(?1) gave a compile-time internal error.
695        ((a|)+)+  did not notice that the outer group could match an empty string.
696        (^a|^)+   was not marked as anchored.
697        (.*a|.*)+ was not marked as matching at start or after a newline.
699    12. Yet more internal refactoring has removed another argument from the match()
700        function. Special calls to this function are now indicated by setting a
701        value in a variable in the "match data" data block.
703    13. Be more explicit in pcre_study() instead of relying on "default" for
704        opcodes that mean there is no starting character; this means that when new
705        ones are added and accidentally left out of pcre_study(), testing should
706        pick them up.
708    14. The -s option of pcretest has been documented for ages as being an old
709        synonym of -m (show memory usage). I have changed it to mean "force study
710        for every regex", that is, assume /S for every regex. This is similar to -i
711        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
712        using it. It makes it easier to run collections of tests with and without
713        study enabled, and thereby test pcre_study() more easily. All the standard
714        tests are now run with and without -s (but some patterns can be marked as
715        "never study" - see 20 below).
717    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
718        restoration of the capturing data to the outer values was not happening
719        correctly.
721    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
722        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
723        pattern had matched an empty string, and so incorrectly returned a no
724        match.
726    17. There was optimizing code for the last branch of non-capturing parentheses,
727        and also for the obeyed branch of a conditional subexpression, which used
728        tail recursion to cut down on stack usage. Unfortunately, now that there is
729        the possibility of (*THEN) occurring in these branches, tail recursion is
730        no longer possible because the return has to be checked for (*THEN). These
731        two optimizations have therefore been removed. [But see 8.20/11 above.]
733    18. If a pattern containing \R was studied, it was assumed that \R always
734        matched two bytes, thus causing the minimum subject length to be
735        incorrectly computed because \R can also match just one byte.
737    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
738        was incorrectly computed.
740    20. If /S is present twice on a test pattern in pcretest input, it now
741        *disables* studying, thereby overriding the use of -s on the command line
742        (see 14 above). This is necessary for one or two tests to keep the output
743        identical in both cases.
745    21. When (*ACCEPT) was used in an assertion that matched an empty string and
746        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
748    22. When an atomic group that contained a capturing parenthesis was
749        successfully matched, but the branch in which it appeared failed, the
750        capturing was not being forgotten if a higher numbered group was later
751        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
752        group 1 to "a", when in fact it should be unset. This applied to multi-
753        branched capturing and non-capturing groups, repeated or not, and also to
754        positive assertions (capturing in negative assertions does not happen
755        in PCRE) and also to nested atomic groups.
757    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
758        subject after a captured substring, to make it easier to tell which of a
759        number of identical substrings has been captured.
761    24. The way atomic groups are processed by pcre_exec() has been changed so that
762        if they are repeated, backtracking one repetition now resets captured
763        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
764        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
765        "aaa". Previously, it would have been "a". As part of this code
766        refactoring, the way recursive calls are handled has also been changed.
768    25. If an assertion condition captured any substrings, they were not passed
769        back unless some other capturing happened later. For example, if
770        (?(?=(a))a) was matched against "a", no capturing was returned.
772    26. When studying a pattern that contained subroutine calls or assertions,
773        the code for finding the minimum length of a possible match was handling
774        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
775        group 1 called group 2 while simultaneously a separate group 2 called group
776        1). A stack overflow occurred in this case. I have fixed this by limiting
777        the recursion depth to 10.
779    27. Updated RunTest.bat in the distribution to the version supplied by Tom
780        Fortmann. This supports explicit test numbers on the command line, and has
781        argument validation and error reporting.
783    28. An instance of \X with an unlimited repeat could fail if at any point the
784        first character it looked at was a mark character.
786    29. Some minor code refactoring concerning Unicode properties and scripts
787        should reduce the stack requirement of match() slightly.
789    30. Added the '=' option to pcretest to check the setting of unused capturing
790        slots at the end of the pattern, which are documented as being -1, but are
791        not included in the return count.
793    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
794        compiled something random. Now it gives a compile-time error (as does
795        Perl).
797    32. A *MARK encountered during the processing of a positive assertion is now
798        recorded and passed back (compatible with Perl).
800    33. If --only-matching or --colour was set on a pcregrep call whose pattern
801        had alternative anchored branches, the search for a second match in a line
802        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
803        matched the line "0102" twice. The same bug affected patterns that started
804        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
805        twice.
807    34. Previously, PCRE did not allow quantification of assertions. However, Perl
808        does, and because of capturing effects, quantifying parenthesized
809        assertions may at times be useful. Quantifiers are now allowed for
810        parenthesized assertions.
812    35. A minor code tidy in pcre_compile() when checking options for \R usage.
814    36. \g was being checked for fancy things in a character class, when it should
815        just be a literal "g".
817    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
818        appearance of a nested POSIX class supersedes an apparent external class.
819        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
820        unescaped square brackets may also appear as part of class names. For
821        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
822        more like Perl. (But see 8.20/1 above.)
824    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
825        was because it thought it was \N{name}, which is not supported).
827    39. Add minix to OS list not supporting the -S option in pcretest.
829    40. PCRE tries to detect cases of infinite recursion at compile time, but it
830        cannot analyze patterns in sufficient detail to catch mutual recursions
831        such as ((?1))((?2)). There is now a runtime test that gives an error if a
832        subgroup is called recursively as a subpattern for a second time at the
833        same position in the subject string. In previous releases this might have
834        been caught by the recursion limit, or it might have run out of stack.
836    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
837        happen only once. PCRE was, however incorrectly giving a compile time error
838        "recursive call could loop indefinitely" because it cannot analyze the
839        pattern in sufficient detail. The compile time test no longer happens when
840        PCRE is compiling a conditional subpattern, but actual runaway loops are
841        now caught at runtime (see 40 above).
843    42. It seems that Perl allows any characters other than a closing parenthesis
844        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
845        has been changed to be the same.
847    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
848        as not to get warnings when autogen.sh is called. Also changed
849        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
851    44. To help people who use pcregrep to scan files containing exceedingly long
852        lines, the following changes have been made:
854        (a) The default value of the buffer size parameter has been increased from
855            8K to 20K. (The actual buffer used is three times this size.)
857        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
858            PCRE is built.
860        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
861            to be set at run time.
863        (d) Numerical values in pcregrep options can be followed by K or M, for
864            example --buffer-size=50K.
866        (e) If a line being scanned overflows pcregrep's buffer, an error is now
867            given and the return code is set to 2.
869    45. Add a pointer to the latest mark to the callout data block.
871    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
872        partial match of an empty string instead of no match. This was specific to
873        the use of ".".
875    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
876        complete match instead of a partial match. This bug was dependent on both
877        the PCRE_UTF8 and PCRE_DOTALL options being set.
879    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
880        starting byte set, because \b was not being ignored.
883    Version 8.12 15-Jan-2011
884    ------------------------
886    1.  Fixed some typos in the markup of the man pages, and wrote a script that
887        checks for such things as part of the documentation building process.
889    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
890        --match-limit and --recursion-limit options (added for 8.11). In
891        particular, this made one of the standard tests fail. (The integer value
892        went into the wrong half of a long int.)
894    3.  If the --colour option was given to pcregrep with -v (invert match), it
895        did strange things, either producing crazy output, or crashing. It should,
896        of course, ignore a request for colour when reporting lines that do not
897        match.
899    4.  Another pcregrep bug caused similar problems if --colour was specified with
900        -M (multiline) and the pattern match finished with a line ending.
902    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
903        matched in multiline mode, the following line was shown as part of the
904        match. This seems wrong, so I have changed it.
906    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
907        the check for further matches in the same line (so they could be coloured)
908        to overrun the end of the current line. If another match was found, it was
909        incorrectly shown (and then shown again when found in the next line).
911    7.  If pcregrep was compiled under Windows, there was a reference to the
912        function pcregrep_exit() before it was defined. I am assuming this was
913        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
914        reported by a user. I've moved the definition above the reference.
917    Version 8.11 10-Dec-2010
918    ------------------------
920    1.  (*THEN) was not working properly if there were untried alternatives prior
921        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
922        backtracked to try for "b" instead of moving to the next alternative branch
923        at the same level (in this case, to look for "c"). The Perl documentation
924        is clear that when (*THEN) is backtracked onto, it goes to the "next
925        alternative in the innermost enclosing group".
927    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
928        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
929        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
930        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
931        (*THEN).
933    3.  If \s appeared in a character class, it removed the VT character from
934        the class, even if it had been included by some previous item, for example
935        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
936        of \s, but is part of the POSIX "space" class.)
938    4.  A partial match never returns an empty string (because you can always
939        match an empty string at the end of the subject); however the checking for
940        an empty string was starting at the "start of match" point. This has been
941        changed to the "earliest inspected character" point, because the returned
942        data for a partial match starts at this character. This means that, for
943        example, /(?<=abc)def/ gives a partial match for the subject "abc"
944        (previously it gave "no match").
946    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
947        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
948        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
949        has an implication that the given string is incomplete (because a partial
950        match is preferred over a full match). For this reason, these items now
951        give a partial match in this situation. [Aside: previously, the one case
952        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
953        match rather than a full match, which was wrong by the old rules, but is
954        now correct.]
956    6.  There was a bug in the handling of #-introduced comments, recognized when
957        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
958        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
959        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
960        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
961        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
962        places in pcre_compile().
964    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
965        comments when looking ahead for named forward references to subpatterns,
966        the only newline sequence it recognized was NL. It now handles newlines
967        according to the set newline convention.
969    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
970        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
971        cater for a lack of strerror(). These oversights have been fixed.
973    9.  Added --match-limit and --recursion-limit to pcregrep.
975    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
977    11. When the -o option was used, pcregrep was setting a return code of 1, even
978        when matches were found, and --line-buffered was not being honoured.
980    12. Added an optional parentheses number to the -o and --only-matching options
981        of pcregrep.
983    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
984        can match an empty string. The code to do it in pcretest and pcredemo
985        needed fixing:
987        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
988            only one byte after an empty string match just before CRLF (this case
989            just got forgotten; "any" and "anycrlf" were OK).
991        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
992            mode when an empty string match preceded an ASCII character followed by
993            a non-ASCII character. (The code for advancing by one character rather
994            than one byte was nonsense.)
996        (c) The pcredemo.c sample program did not have any code at all to handle
997            the cases when CRLF is a valid newline sequence.
999    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1000        as a starting offset was within the subject string. There is now a new
1001        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1002        negative or greater than the length of the string. In order to test this,
1003        pcretest is extended to allow the setting of negative starting offsets.
1005    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1006        starting offset points to the beginning of a UTF-8 character was
1007        unnecessarily clumsy. I tidied it up.
1009    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1010        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1012    17. Nobody had reported that the --include_dir option, which was added in
1013        release 7.7 should have been called --include-dir (hyphen, not underscore)
1014        for compatibility with GNU grep. I have changed it to --include-dir, but
1015        left --include_dir as an undocumented synonym, and the same for
1016        --exclude-dir, though that is not available in GNU grep, at least as of
1017        release 2.5.4.
1019    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1020        characters from a string of bytes) have been redefined so as not to use
1021        loops, in order to improve performance in some environments. At the same
1022        time, I abstracted some of the common code into auxiliary macros to save
1023        repetition (this should not affect the compiled code).
1025    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1026        compile-time error is now given if \c is not followed by an ASCII
1027        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1028        different, and any byte value is allowed.)
1030    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1031        START_OPTIMIZE option, which is now allowed at compile time - but just
1032        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1033        to pcregrep and other applications that have no direct access to PCRE
1034        options. The new /Y option in pcretest sets this option when calling
1035        pcre_compile().
1037    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1038        back references. Groups containing recursive back references were forced to
1039        be atomic by that change, but in the case of named groups, the amount of
1040        memory required was incorrectly computed, leading to "Failed: internal
1041        error: code overflow". This has been fixed.
1043    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1044        pcretest.c, to avoid build problems in some Borland environments.
1047    Version 8.10 25-Jun-2010
1048    ------------------------
1050    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1051        THEN.
1053    2.  (*ACCEPT) was not working when inside an atomic group.
1055    3.  Inside a character class, \B is treated as a literal by default, but
1056        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1057        causes the error). The code is unchanged, but I tidied the documentation.
1059    4.  Inside a character class, PCRE always treated \R and \X as literals,
1060        whereas Perl faults them if its -w option is set. I have changed PCRE so
1061        that it faults them when PCRE_EXTRA is set.
1063    5.  Added support for \N, which always matches any character other than
1064        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1066    6.  When compiling pcregrep with newer versions of gcc which may have
1067        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1068        declared with attribute warn_unused_result" were given. Just casting the
1069        result to (void) does not stop the warnings; a more elaborate fudge is
1070        needed. I've used a macro to implement this.
1072    7.  Minor change to pcretest.c to avoid a compiler warning.
1074    8.  Added four artifical Unicode properties to help with an option to make
1075        \s etc use properties (see next item). The new properties are: Xan
1076        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1078    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1079        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1080        this option. Modified pcretest to add /W to test this facility. Added
1081        REG_UCP to make it available via the POSIX interface.
1083    10. Added --line-buffered to pcregrep.
1085    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1086        studied, and the match started with a letter with a code point greater than
1087        127 whose first byte was different to the first byte of the other case of
1088        the letter, the other case of this starting letter was not recognized
1089        (#976).
1091    12. If a pattern that was studied started with a repeated Unicode property
1092        test, for example, \p{Nd}+, there was the theoretical possibility of
1093        setting up an incorrect bitmap of starting bytes, but fortunately it could
1094        not have actually happened in practice until change 8 above was made (it
1095        added property types that matched character-matching opcodes).
1097    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1098        possible starting bytes for non-anchored patterns.
1100    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1101        \R, and also a number of cases that involve Unicode properties, both
1102        explicit and implicit when PCRE_UCP is set.
1104    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1105        input, it could crash or give wrong results if characters with values
1106        greater than 0xc0 were present in the subject string. (Detail: it assumed
1107        UTF-8 input when processing these items.)
1109    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1110        size_t is 64-bit (#991).
1112    17. Added a check for running out of memory when PCRE is compiled with
1113        --disable-stack-for-recursion (#990).
1115    18. If the last data line in a file for pcretest does not have a newline on
1116        the end, a newline was missing in the output.
1118    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1119        less than 128) in its various bitmaps. However, there is a facility for
1120        generating tables according to the current locale when PCRE is compiled. It
1121        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1122        characters, are recognized by isspace() and therefore were getting set in
1123        these tables, and indeed these tables seem to approximate to ISO 8859. This
1124        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1125        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1126        which of course cannot start UTF-8 characters. I have changed the code so
1127        that only real ASCII characters (less than 128) and the correct starting
1128        bytes for UTF-8 encodings are set for characters greater than 127 when in
1129        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1130        altogether.)
1132    20. Added the /T option to pcretest so as to be able to run tests with non-
1133        standard character tables, thus making it possible to include the tests
1134        used for 19 above in the standard set of tests.
1136    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1137        reference to a subpattern the other side of a comment that contains an
1138        opening parenthesis caused either an internal compiling error, or a
1139        reference to the wrong subpattern.
1142    Version 8.02 19-Mar-2010
1143    ------------------------
1145    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1147    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1148        configured.
1150    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1151        original author of that file, following a query about its status.
1153    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1154        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1156    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1157        quantifier applied to a forward-referencing subroutine call, could compile
1158        incorrect code or give the error "internal error: previously-checked
1159        referenced subpattern not found".
1161    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1162        variables to point to external functions. For these systems, therefore,
1163        pcre_malloc etc. are now initialized to local functions that call the
1164        relevant global functions.
1166    7.  There were two entries missing in the vectors called coptable and poptable
1167        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1168        I've fixed the data, and added a kludgy way of testing at compile time that
1169        the lengths are correct (equal to the number of opcodes).
1171    8.  Following on from 7, I added a similar kludge to check the length of the
1172        eint vector in pcreposix.c.
1174    9.  Error texts for pcre_compile() are held as one long string to avoid too
1175        much relocation at load time. To find a text, the string is searched,
1176        counting zeros. There was no check for running off the end of the string,
1177        which could happen if a new error number was added without updating the
1178        string.
1180    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1182    11. \K was not working if it appeared in an atomic group or in a group that
1183        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1184        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1185        the assertion is positive, but not if it is negative.
1187    12. Change 11 fortuitously reduced the size of the stack frame used in the
1188        "match()" function of pcre_exec.c by one pointer. Forthcoming
1189        implementation of support for (*MARK) will need an extra pointer on the
1190        stack; I have reserved it now, so that the stack frame size does not
1191        decrease.
1193    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1194        item in branch that calls a recursion is a subroutine call - as in the
1195        second branch in the above example - was incorrectly given the compile-
1196        time error "recursive call could loop indefinitely" because pcre_compile()
1197        was not correctly checking the subroutine for matching a non-empty string.
1199    14. The checks for overrunning compiling workspace could trigger after an
1200        overrun had occurred. This is a "should never occur" error, but it can be
1201        triggered by pathological patterns such as hundreds of nested parentheses.
1202        The checks now trigger 100 bytes before the end of the workspace.
1204    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1207    Version 8.01 19-Jan-2010
1208    ------------------------
1210    1.  If a pattern contained a conditional subpattern with only one branch (in
1211        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1212        computed the wrong minimum data length (which is of course zero for such
1213        subpatterns). This could cause incorrect "no match" results.
1215    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1216        the pattern is reset in the first branch, pcre_compile() failed with
1217        "internal error: code overflow at offset...". This happened only when
1218        the reset was to the original external option setting. (An optimization
1219        abstracts leading options settings into an external setting, which was the
1220        cause of this.)
1222    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1223        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1224        assertion pattern did not match (meaning that the assertion was true), it
1225        was incorrectly treated as false if the SKIP had been reached during the
1226        matching. This also applied to assertions used as conditions.
1228    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1229        assertion subpattern, including such a pattern used as a condition,
1230        unpredictable results occurred, instead of the error return
1233    5.  The C++ GlobalReplace function was not working like Perl for the special
1234        situation when an empty string is matched. It now does the fancy magic
1235        stuff that is necessary.
1237    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1238        removed. (These were left over from very, very early versions of PCRE.)
1240    7.  Some cosmetic changes to the code to make life easier when compiling it
1241        as part of something else:
1243        (a) Change DEBUG to PCRE_DEBUG.
1245        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1246            called "current" as "current_branch", to prevent a collision with the
1247            Linux macro when compiled as a kernel module.
1249        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1250            prevent a collision with the Linux macro when compiled as a kernel
1251            module.
1253    8.  In pcre_compile() there are some checks for integer overflows that used to
1254        cast potentially large values to (double). This has been changed to that
1255        when building, a check for int64_t is made, and if it is found, it is used
1256        instead, thus avoiding the use of floating point arithmetic. (There is no
1257        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1258        double.
1260    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1261        2005 (difference between two addresses compared to an unsigned value).
1263    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1264        custom one, because of the following reported problem in Windows:
1266          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1267              under Win32.
1268          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1269              therefore missing the function definition.
1270          - The compiler thus generates a "C" signature for the test function.
1271          - The linker fails to find the "C" function.
1272          - PCRE fails to configure if asked to do so against libbz2.
1274    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1275        messages were output:
1277          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1278          rerunning libtoolize, to keep the correct libtool macros in-tree.
1279          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1281        I have done both of these things.
1283    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1284        most of the time, it *can* run out if it is given a pattern that contains a
1285        runaway infinite recursion. I updated the discussion in the pcrestack man
1286        page.
1288    13. Now that we have gone to the x.xx style of version numbers, the minor
1289        version may start with zero. Using 08 or 09 is a bad idea because users
1290        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1291        interpreted as invalid octal numbers. I've updated the previous comment in
1292        configure.ac, and also added a check that gives an error if 08 or 09 are
1293        used.
1295    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1296        causing partial matching to fail when the end of the subject matched \W
1297        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1299    15. There were some discrepancies between the declarations in pcre_internal.h
1300        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1301        their definitions. The declarations used "const uschar *" and the
1302        definitions used USPTR. Even though USPTR is normally defined as "const
1303        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1304        reported that: "This difference in casting confuses some C++ compilers, for
1305        example, SunCC recognizes above declarations as different functions and
1306        generates broken code for hbpcre." I have changed the declarations to use
1307        USPTR.
1309    16. GNU libtool is named differently on some systems. The autogen.sh script now
1310        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1311        (FreeBSD).
1313    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1314        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1315        comment: "Figure out how to create a longlong from a string: strtoll and
1316        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1317        instance, but it only takes 2 args instead of 3!"
1319    18. A subtle bug concerned with back references has been fixed by a change of
1320        specification, with a corresponding code fix. A pattern such as
1321        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1322        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1323        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1324        same bug. Such groups have to be quantified to be useful, or contained
1325        inside another quantified group. (If there's no repetition, the reference
1326        can never match.) The problem arises because, having left the group and
1327        moved on to the rest of the pattern, a later failure that backtracks into
1328        the group uses the captured value from the final iteration of the group
1329        rather than the correct earlier one. I have fixed this in PCRE by forcing
1330        any group that contains a reference to itself to be an atomic group; that
1331        is, there cannot be any backtracking into it once it has completed. This is
1332        similar to recursive and subroutine calls.
1335    Version 8.00 19-Oct-09
1336    ----------------------
1338    1.  The table for translating pcre_compile() error codes into POSIX error codes
1339        was out-of-date, and there was no check on the pcre_compile() error code
1340        being within the table. This could lead to an OK return being given in
1341        error.
1343    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1344        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1345        in a Windows environment.
1347    3.  The pcregrep --count option prints the count for each file even when it is
1348        zero, as does GNU grep. However, pcregrep was also printing all files when
1349        --files-with-matches was added. Now, when both options are given, it prints
1350        counts only for those files that have at least one match. (GNU grep just
1351        prints the file name in this circumstance, but including the count seems
1352        more useful - otherwise, why use --count?) Also ensured that the
1353        combination -clh just lists non-zero counts, with no names.
1355    4.  The long form of the pcregrep -F option was incorrectly implemented as
1356        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1357        but it seems right to fix it, and I didn't think it was worth preserving
1358        the old behaviour.
1360    5.  The command line items --regex=pattern and --regexp=pattern were not
1361        recognized by pcregrep, which required --regex pattern or --regexp pattern
1362        (with a space rather than an '='). The man page documented the '=' forms,
1363        which are compatible with GNU grep; these now work.
1365    6.  No libpcreposix.pc file was created for pkg-config; there was just
1366        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1368    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1369        when UCP support is not needed, by modifying the Python script that
1370        generates it from Unicode data files. This should not matter if the module
1371        is correctly used as a library, but I received one complaint about 50K of
1372        unwanted data. My guess is that the person linked everything into his
1373        program rather than using a library. Anyway, it does no harm.
1375    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1376        was a minimum greater than 1 for a wide character in a possessive
1377        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1378        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1379        character. Chaos in the form of incorrect output or a compiling loop could
1380        result.
1382    9.  The restrictions on what a pattern can contain when partial matching is
1383        requested for pcre_exec() have been removed. All patterns can now be
1384        partially matched by this function. In addition, if there are at least two
1385        slots in the offset vector, the offset of the earliest inspected character
1386        for the match and the offset of the end of the subject are set in them when
1387        PCRE_ERROR_PARTIAL is returned.
1389    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1390        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1391        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1392        and may be more useful for multi-segment matching.
1394    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1395        used to be given if ever the end of the subject was reached; now it is
1396        given only if matching could not proceed because another character was
1397        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1398        string "Z", which now yields "no match" instead of "partial match". In the
1399        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1400        final character ended with (*FAIL).
1402    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1403        if the pattern had a "must contain" character that was already found in the
1404        earlier partial match, unless partial matching was again requested. For
1405        example, with the pattern /dog.(body)?/, the "must contain" character is
1406        "g". If the first part-match was for the string "dog", restarting with
1407        "sbody" failed. This bug has been fixed.
1409    13. The string returned by pcre_dfa_exec() after a partial match has been
1410        changed so that it starts at the first inspected character rather than the
1411        first character of the match. This makes a difference only if the pattern
1412        starts with a lookbehind assertion or \b or \B (\K is not supported by
1413        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1414        matching functions compatible, and I think it's the right thing to do.
1416    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1417        so that the demonstration program is easily available in environments where
1418        PCRE has not been installed from source.
1420    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1421        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1422        library.
1424    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1425        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1426        is not the first non-POSIX option to be added. Clearly some people find
1427        these options useful.
1429    17. If a caller to the POSIX matching function regexec() passes a non-zero
1430        value for nmatch with a NULL value for pmatch, the value of
1431        nmatch is forced to zero.
1433    18. RunGrepTest did not have a test for the availability of the -u option of
1434        the diff command, as RunTest does. It now checks in the same way as
1435        RunTest, and also checks for the -b option.
1437    19. If an odd number of negated classes containing just a single character
1438        interposed, within parentheses, between a forward reference to a named
1439        subpattern and the definition of the subpattern, compilation crashed with
1440        an internal error, complaining that it could not find the referenced
1441        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1442        [The bug was that it was starting one character too far in when skipping
1443        over the character class, thus treating the ] as data rather than
1444        terminating the class. This meant it could skip too much.]
1446    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1447        /g option in pcretest when the pattern contains \K, which makes it possible
1448        to have an empty string match not at the start, even when the pattern is
1449        anchored. Updated pcretest and pcredemo to use this option.
1451    21. If the maximum number of capturing subpatterns in a recursion was greater
1452        than the maximum at the outer level, the higher number was returned, but
1453        with unset values at the outer level. The correct (outer level) value is
1454        now given.
1456    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1457        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1458        make it do so. The string so far is captured, making this feature
1459        compatible with Perl.
1461    23. The tests have been re-organized, adding tests 11 and 12, to make it
1462        possible to check the Perl 5.10 features against Perl 5.10.
1464    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1465        pattern matches a fixed length string. PCRE did not allow this; now it
1466        does. Neither allows recursion.
1468    25. I finally figured out how to implement a request to provide the minimum
1469        length of subject string that was needed in order to match a given pattern.
1470        (It was back references and recursion that I had previously got hung up
1471        on.) This code has now been added to pcre_study(); it finds a lower bound
1472        to the length of subject needed. It is not necessarily the greatest lower
1473        bound, but using it to avoid searching strings that are too short does give
1474        some useful speed-ups. The value is available to calling programs via
1475        pcre_fullinfo().
1477    26. While implementing 25, I discovered to my embarrassment that pcretest had
1478        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1479        study optimizations had never been tested with that matching function.
1480        Oops. What is worse, even when it was passed study data, there was a bug in
1481        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1482        were also very few tests of studied patterns with pcre_dfa_exec().
1484    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1485        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1486        on the other side of the coin, they are no longer allowed to have different
1487        names, because these cannot be distinguished in PCRE, and this has caused
1488        confusion. (This is a difference from Perl.)
1490    28. When duplicate subpattern names are present (necessarily with different
1491        numbers, as required by 27 above), and a test is made by name in a
1492        conditional pattern, either for a subpattern having been matched, or for
1493        recursion in such a pattern, all the associated numbered subpatterns are
1494        tested, and the overall condition is true if the condition is true for any
1495        one of them. This is the way Perl works, and is also more like the way
1496        testing by number works.
1499    Version 7.9 11-Apr-09
1500    ---------------------
1502    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1503        (pcretest), all targets were linked against these libraries. This included
1504        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1505        libraries. This caused unwanted dependencies to be created. This problem
1506        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1507        pcretest is linked with readline.
1509    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1510        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1511        moved outside it again, because FALSE and TRUE are already defined in AIX,
1512        but BOOL is not.
1514    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1515        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1517    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1518        hyphens) following file names and line numbers when outputting matching
1519        lines. This is not true; no spaces are inserted. I have also clarified the
1520        wording for the --colour (or --color) option.
1522    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1523        was not coloured; this is different to GNU grep, so I have changed it to be
1524        the same.
1526    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1527        each matching line was coloured. Now it goes on to look for further matches
1528        of any of the test patterns, which is the same behaviour as GNU grep.
1530    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1531        doesn't make sense to accept an empty string match in pcregrep, so I have
1532        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1533        seems to be how GNU grep behaves.
1535    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1536        start or after a newline", because the conditional assertion was not being
1537        correctly handled. The rule now is that both the assertion and what follows
1538        in the first alternative must satisfy the test.
1540    9.  If auto-callout was enabled in a pattern with a conditional group whose
1541        condition was an assertion, PCRE could crash during matching, both with
1542        pcre_exec() and pcre_dfa_exec().
1544    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1545        used for matching.
1547    11. Unicode property support in character classes was not working for
1548        characters (bytes) greater than 127 when not in UTF-8 mode.
1550    12. Added the -M command line option to pcretest.
1552    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1554    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1556    16. Added comments and documentation about mis-use of no_arg in the C++
1557        wrapper.
1559    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1560        from Martin Jerabek that uses macro names for all relevant character and
1561        string constants.
1563    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1564        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1565        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1566        these, but not everybody uses configure.
1568    19. A conditional group that had only one branch was not being correctly
1569        recognized as an item that could match an empty string. This meant that an
1570        enclosing group might also not be so recognized, causing infinite looping
1571        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1572        with the subject "ab", where knowledge that the repeated group can match
1573        nothing is needed in order to break the loop.
1575    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1576        exec(), but without supplying a callout function, matching went wrong.
1578    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1579        leak if the size of the offset vector was greater than 30. When the vector
1580        is smaller, the saved offsets during recursion go onto a local stack
1581        vector, but for larger vectors malloc() is used. It was failing to free
1582        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1583        error, in fact).
1585    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1586        heapframe that is used only when UTF-8 support is enabled. This caused no
1587        problem, but was untidy.
1589    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1590        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1591        included within another project.
1593    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1594        slightly modified by me:
1596          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1597              not building pcregrep.
1599          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1600              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1602    25. Forward references, both numeric and by name, in patterns that made use of
1603        duplicate group numbers, could behave incorrectly or give incorrect errors,
1604        because when scanning forward to find the reference group, PCRE was not
1605        taking into account the duplicate group numbers. A pattern such as
1606        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1608    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1609        the feature of a custom pointer more persuasive (as requested by a user).
1611    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1612        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1613        pre-defined.
1615    28. Added support for (*UTF8) at the start of a pattern.
1617    29. Arrange for flags added by the "release type" setting in CMake to be shown
1618        in the configuration summary.
1621    Version 7.8 05-Sep-08
1622    ---------------------
1624    1.  Replaced UCP searching code with optimized version as implemented for Ad
1625        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1626        stage table and inline lookup instead of a function, giving speed ups of 2
1627        to 5 times on some simple patterns that I tested. Permission was given to
1628        distribute the MultiStage2.py script that generates the tables (it's not in
1629        the tarball, but is in the Subversion repository).
1631    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1632        scripts.
1634    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1635        a group with a zero qualifier. The result of the study could be incorrect,
1636        or the function might crash, depending on the pattern.
1638    4.  Caseless matching was not working for non-ASCII characters in back
1639        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1640        It now works when Unicode Property Support is available.
1642    5.  In pcretest, an escape such as \x{de} in the data was always generating
1643        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1644        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1645        truncation.
1647    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1649    7.  Added two (int) casts to pcregrep when printing the difference of two
1650        pointers, in case they are 64-bit values.
1652    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1653        test 2 if it fails.
1655    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1656        and a #define of that name to empty if it is not externally set. This is to
1657        allow users of MSVC to set it if necessary.
1659    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1660        the convenience functions in the pcre_get.c source file.
1662    11. An option change at the start of a pattern that had top-level alternatives
1663        could cause overwriting and/or a crash. This command provoked a crash in
1664        some environments:
1666          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1668        This potential security problem was recorded as CVE-2008-2371.
1670    12. For a pattern where the match had to start at the beginning or immediately
1671        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1672        pcre_dfa_exec() could read past the end of the passed subject if there was
1673        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1674        pcretest so that it places the subject at the end of its malloc-ed buffer.
1676    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1677        exec() might read past the end of the data buffer in UTF-8 mode.
1679    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1680        the data contained the byte 0x85 as part of a UTF-8 character within its
1681        first line. This applied both to normal and DFA matching.
1683    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1684        /^[^d]*?$/8 failed to match "abc".
1686    16. Added a missing copyright notice to pcrecpp_internal.h.
1688    17. Make it more clear in the documentation that values returned from
1689        pcre_exec() in ovector are byte offsets, not character counts.
1691    18. Tidied a few places to stop certain compilers from issuing warnings.
1693    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1694        supplied by Stefan Weber. I made a further small update for 7.8 because
1695        there is a change of source arrangements: the pcre_searchfuncs.c module is
1696        replaced by pcre_ucd.c.
1699    Version 7.7 07-May-08
1700  ---------------------  ---------------------
1702  1.  Applied Craig's patch to sort out a long long problem: "If we can't convert  1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1703      a string to a long long, pretend we don't even have a long long." This is      a string to a long long, pretend we don't even have a long long." This is
1704      done by checking for the strtoq, strtoll, and _strtoi64 functions.      done by checking for the strtoq, strtoll, and _strtoi64 functions.
1706  2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with  2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1707      pre-7.6 versions, which defined a global no_arg variable instead of putting      pre-7.6 versions, which defined a global no_arg variable instead of putting
1708      it in the RE class. (See also #8 below.)      it in the RE class. (See also #8 below.)
1710  3.  Remove a line of dead code, identified by coverity and reported by Nuno  3.  Remove a line of dead code, identified by coverity and reported by Nuno
1711      Lopes.      Lopes.
1713  4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:  4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1715      (1) The include/exclude patterns were being applied to the whole pathnames      (1) The include/exclude patterns were being applied to the whole pathnames
1716          of files, instead of just to the final components.          of files, instead of just to the final components.
1718      (2) If there was more than one level of directory, the subdirectories were      (2) If there was more than one level of directory, the subdirectories were
1719          skipped unless they satisfied the include/exclude conditions. This is          skipped unless they satisfied the include/exclude conditions. This is
1720          inconsistent with GNU grep (and could even be seen as contrary to the          inconsistent with GNU grep (and could even be seen as contrary to the
1721          pcregrep specification - which I improved to make it absolutely clear).          pcregrep specification - which I improved to make it absolutely clear).
1722          The action now is always to scan all levels of directory, and just          The action now is always to scan all levels of directory, and just
1723          apply the include/exclude patterns to regular files.          apply the include/exclude patterns to regular files.
1725  5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used  5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1726      --exclude_dir in the tests to avoid scanning .svn directories.      --exclude_dir in the tests to avoid scanning .svn directories.
1728  6.  Applied Craig's patch to the QuoteMeta function so that it escapes the  6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1729      NUL character as backslash + 0 rather than backslash + NUL, because PCRE      NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1730      doesn't support NULs in patterns.      doesn't support NULs in patterns.
1732  7.  Added some missing "const"s to declarations of static tables in  7.  Added some missing "const"s to declarations of static tables in
1733      pcre_compile.c and pcre_dfa_exec.c.      pcre_compile.c and pcre_dfa_exec.c.
1735  8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was  8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1736      caused by fix #2  above. (Subsequently also a second patch to fix the      caused by fix #2  above. (Subsequently also a second patch to fix the
1737      first patch. And a third patch - this was a messy problem.)      first patch. And a third patch - this was a messy problem.)
1739  9.  Applied Craig's patch to remove the use of push_back().  9.  Applied Craig's patch to remove the use of push_back().
1741  10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX  10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1742      matching function regexec().      matching function regexec().
1744  11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',  11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1745      which, however, unlike Perl's \g{...}, are subroutine calls, not back      which, however, unlike Perl's \g{...}, are subroutine calls, not back
1746      references. PCRE supports relative numbers with this syntax (I don't think      references. PCRE supports relative numbers with this syntax (I don't think
1747      Oniguruma does).      Oniguruma does).
1749  12. Previously, a group with a zero repeat such as (...){0} was completely  12. Previously, a group with a zero repeat such as (...){0} was completely
1750      omitted from the compiled regex. However, this means that if the group      omitted from the compiled regex. However, this means that if the group
1751      was called as a subroutine from elsewhere in the pattern, things went wrong      was called as a subroutine from elsewhere in the pattern, things went wrong
1752      (an internal error was given). Such groups are now left in the compiled      (an internal error was given). Such groups are now left in the compiled
1753      pattern, with a new opcode that causes them to be skipped at execution      pattern, with a new opcode that causes them to be skipped at execution
1754      time.      time.
1756    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1757        to the way PCRE behaves:
1759        (a) A lone ] character is dis-allowed (Perl treats it as data).
1761        (b) A back reference to an unmatched subpattern matches an empty string
1762            (Perl fails the current match path).
1764        (c) A data ] in a character class must be notated as \] because if the
1765            first data character in a class is ], it defines an empty class. (In
1766            Perl it is not possible to have an empty class.) The empty class []
1767            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1768            The negative empty class [^] matches any one character, independently
1769            of the DOTALL setting.
1771    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1772        non-existent subpattern following a character class starting with ']' and
1773        containing () gave an internal compiling error instead of "reference to
1774        non-existent subpattern". Fortunately, when the pattern did exist, the
1775        compiled code was correct. (When scanning forwards to check for the
1776        existencd of the subpattern, it was treating the data ']' as terminating
1777        the class, so got the count wrong. When actually compiling, the reference
1778        was subsequently set up correctly.)
1780    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1781        it was being rejected as not supported by pcre_dfa_exec(), even though
1782        other assertions are supported. I have made pcre_dfa_exec() support
1783        (*FAIL).
1785    16. The implementation of 13c above involved the invention of a new opcode,
1786        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1787        cannot be changed at match time, I realized I could make a small
1788        improvement to matching performance by compiling OP_ALLANY instead of
1789        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1790        on the OP_ANY path.
1792    17. Compiling pcretest on Windows with readline support failed without the
1793        following two fixes: (1) Make the unistd.h include conditional on
1794        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1796    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1797        ncurses library to be included for pcretest when ReadLine support is
1798        requested, but also to allow for it to be overridden. This patch came from
1799        Daniel Bergström.
1801    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1802        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1803        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1804        spotting this.
1807  Version 7.6 28-Jan-08  Version 7.6 28-Jan-08
1808  ---------------------  ---------------------

Removed from v.335  
changed lines
  Added in v.1236

  ViewVC Help
Powered by ViewVC 1.1.5