ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 167 by ph10, Wed May 9 15:53:54 2007 UTC revision 1298 by ph10, Fri Mar 22 16:13:13 2013 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4  Version 7.2 01-May-07  Version 8.33 xx-xxxx-201x
5    -------------------------
7    1.  Added 'U' to some constants that are compared to unsigned integers, to
8        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
9        variables that are added to signed variables, to ensure the result is
10        signed and can be negated.
12    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
13        CMake config files.
15    3.  Revise the creation of config.h.generic so that all boolean macros are
16        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
17        overriding via -D on the command line possible.
19    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
20        to unsigned int is reported to make a quite noticeable speed difference in
21        a specific Windows environment. Testing on Linux did also appear to show
22        some benefit (and it is clearly not harmful). Also fixed the definition of
23        Xop which should be unsigned.
25    5.  Related to (4), changing the definition of the intermediate variable cc
26        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
27        improvements.
29    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
30        unnecessary spaces.
32    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
33        and later.
35    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
37    9.  Optimizing fast_forward_start_bits in JIT.
39    10. Adding experimental support for callouts in JIT, and fixing some
40        issues revealed during this work. Namely:
42        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
44        (b) Minimum length was not checked before the matching is started.
46    11. The value of capture_last that is passed to callouts was incorrect in some
47        cases when there was a capture on one path that was subsequently abandoned
48        after a backtrack. Also, the capture_last value is now reset after a
49        recursion, since all captures are also reset in this case.
51    12. The interpreter no longer returns the "too many substrings" error in the
52        case when an overflowing capture is in a branch that is subsequently
53        abandoned after a backtrack.
55    13. In the pathological case when an offset vector of size 2 is used, pcretest
56        now prints out the matched string after a yield of 0 or 1.
58    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
59        Only supported by the JIT compiler at the moment.
61    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
63    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
64        offset of the starting point of the matching process, provided the offsets
65        vector is large enough.
67    17. The \A escape now records a lookbehind value of 1, though its execution
68        does not actually inspect the previous character. This is to ensure that,
69        in partial multi-segment matching, at least one character from the old
70        segment is retained when a new segment is processed. Otherwise, if there
71        are no lookbehinds in the pattern, \A might match incorrectly at the start
72        of a new segment.
74    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
76    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
77        gives some modest performance improvement in 8-bit mode.
79    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
80        be expressed in certain programming languages using Universal Character
81        Names.
83    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
84        which points out that "non characters" are not "characters that may not
85        appear in Unicode strings" but rather "characters that are reserved for
86        internal use and have only local meaning".
88    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
89        there was a conditional group that depended on an assertion, if the
90        assertion was false, the callout that immediately followed the alternation
91        in the condition was skipped when pcre_exec() was used for matching.
93    23. Allow an explicit callout to be inserted before an assertion that is the
94        condition for a conditional group, for compatibility with automatic
95        callouts, which always insert a callout at this point.
97    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
98        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
100    25. (*PRUNE) is now supported by the JIT compiler.
102    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
104    27. Fix the case where there are two or more SKIPs with arguments that may be
105        ignored.
107    28. Experimental support of (*SKIP) backtracking verb in the JIT compiler.
109    29. Experimental support of (*THEN) backtracking verb in the JIT compiler.
111    30. Update RunTest with additional test selector options.
113    31. The way PCRE handles backtracking verbs has been changed in to ways.
115        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
116        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
117        backtracking. In some cases this makes it more Perl-compatible, but Perl's
118        rather obscure rules do not always do the same thing.
120        (2) Previously, backtracking verbs were confined within assertions. This is
121        no longer the case, except for (*ACCEPT). Again, this sometimes improves
122        Perl compatibility, and sometimes does not.
125    Version 8.32 30-November-2012
126    -----------------------------
128    1.  Improved JIT compiler optimizations for first character search and single
129        character iterators.
131    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
132        Patch by Daniel Richard G.
134    3.  Single character iterator optimizations in the JIT compiler.
136    4.  Improved JIT compiler optimizations for character ranges.
138    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
139        Reported by Giuseppe D'Angelo.
141    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
142        of a line, was being set incorrectly in cases where .* appeared inside
143        atomic brackets at the start of a pattern, or where there was a subsequent
144        *PRUNE or *SKIP.
146    7.  Improved instruction cache flush for POWER/PowerPC.
147        Patch by Daniel Richard G.
149    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
150        grep:
152        (a) There is now no limit to the number of patterns to be matched.
154        (b) An error is given if a pattern is too long.
156        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
157            are now supported.
159        (d) --exclude-from and --include-from (multiple use) have been added.
161        (e) Exclusions and inclusions now apply to all files and directories, not
162            just to those obtained from scanning a directory recursively.
164        (f) Multiple uses of -f and --file-list are now supported.
166        (g) In a Windows environment, the default for -d has been changed from
167            "read" (the GNU grep default) to "skip", because otherwise the presence
168            of a directory in the file list provokes an error.
170        (h) The documentation has been revised and clarified in places.
172    9.  Improve the matching speed of capturing brackets.
174    10. Changed the meaning of \X so that it now matches a Unicode extended
175        grapheme cluster.
177    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
178        out POSIX threads when JIT support is configured.
180    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
182    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
183        structure could go wrong in environments where size_t is not the same size
184        as int.
186    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
187        set.
189    15. The EBCDIC support had decayed; later updates to the code had included
190        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
191        general tidy up of EBCDIC-related issues, and the documentation was also
192        not quite right. There is now a test that can be run on ASCII systems to
193        check some of the EBCDIC-related things (but is it not a full test).
195    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
196        in a small tidy to the code.
198    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
200    18. If the --only-matching (-o) option in pcregrep is specified multiple
201        times, each one causes appropriate output. For example, -o1 -o2 outputs the
202        substrings matched by the 1st and 2nd capturing parentheses. A separating
203        string can be specified by --om-separator (default empty).
205    19. Improving the first n character searches.
207    20. Turn case lists for horizontal and vertical white space into macros so that
208        they are defined only once.
210    21. This set of changes together give more compatible Unicode case-folding
211        behaviour for characters that have more than one other case when UCP
212        support is available.
214        (a) The Unicode property table now has offsets into a new table of sets of
215            three or more characters that are case-equivalent. The MultiStage2.py
216            script that generates these tables (the pcre_ucd.c file) now scans
217            CaseFolding.txt instead of UnicodeData.txt for character case
218            information.
220        (b) The code for adding characters or ranges of characters to a character
221            class has been abstracted into a generalized function that also handles
222            case-independence. In UTF-mode with UCP support, this uses the new data
223            to handle characters with more than one other case.
225        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
226            whose other case is greater than 256 are now correctly matched
227            caselessly. Previously, the high codepoint matched the low one, but not
228            vice versa.
230        (d) The processing of \h, \H, \v, and \ in character classes now makes use
231            of the new class addition function, using character lists defined as
232            macros alongside the case definitions of 20 above.
234        (e) Caseless back references now work with characters that have more than
235            one other case.
237        (f) General caseless matching of characters with more than one other case
238            is supported.
240    22. Unicode character properties were updated from Unicode 6.2.0
242    23. Improved CMake support under Windows. Patch by Daniel Richard G.
244    24. Add support for 32-bit character strings, and UTF-32
246    25. Major JIT compiler update (code refactoring and bugfixing).
247        Experimental Sparc 32 support is added.
249    26. Applied a modified version of Daniel Richard G's patch to create
250        pcre.h.generic and config.h.generic by "make" instead of in the
251        PrepareRelease script.
253    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
254        pcre_compile.c when checking for a zero character.
256    28. Introducing a native interface for JIT. Through this interface, the compiled
257        machine code can be directly executed. The purpose of this interface is to
258        provide fast pattern matching, so several sanity checks are not performed.
259        However, feature tests are still performed. The new interface provides
260        1.4x speedup compared to the old one.
262    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
263        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
264        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
266    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
267        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
268        the "old" RFC 2279). Instead, it ended up passing a negative length to
269        pcre_exec().
271    31. Add support for GCC's visibility feature to hide internal functions.
273    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
274        "unknown -C option" after outputting 0 or 1.
276    33. There is now support for generating a code coverage report for the test
277        suite in environments where gcc is the compiler and lcov is installed. This
278        is mainly for the benefit of the developers.
280    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
281        unaddressable using valgrind annotations, allowing valgrind to detect
282        invalid memory accesses. This is mainly for the benefit of the developers.
284    25. (*UTF) can now be used to start a pattern in any of the three libraries.
286    26. Give configure error if --enable-cpp but no C++ compiler found.
289    Version 8.31 06-July-2012
290    -------------------------
292    1.  Fixing a wrong JIT test case and some compiler warnings.
294    2.  Removed a bashism from the RunTest script.
296    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
297        to unsigned type, result still unsigned" that was given by an MS compiler
298        on encountering the code "-sizeof(xxx)".
300    4.  Partial matching support is added to the JIT compiler.
302    5.  Fixed several bugs concerned with partial matching of items that consist
303        of more than one character:
305        (a) /^(..)\1/ did not partially match "aba" because checking references was
306            done on an "all or nothing" basis. This also applied to repeated
307            references.
309        (b) \R did not give a hard partial match if \r was found at the end of the
310            subject.
312        (c) \X did not give a hard partial match after matching one or more
313            characters at the end of the subject.
315        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
316            a partial match for the string "\r".
318        (e) When newline was set to CRLF, the metacharacter "." did not recognize
319            a partial match for a CR character at the end of the subject string.
321    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
322        running pcretest, the text "(JIT)" added to the output whenever JIT is
323        actually used to run the match.
325    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
326        or /S+[+] with a digit between 1 and 7.
328    8.  OP_NOT now supports any UTF character not just single-byte ones.
330    9.  (*MARK) control verb is now supported by the JIT compiler.
332    10. The command "./RunTest list" lists the available tests without actually
333        running any of them. (Because I keep forgetting what they all are.)
337    12. Applied a (slightly modified) user-supplied patch that improves performance
338        when the heap is used for recursion (compiled with --disable-stack-for-
339        recursion). Instead of malloc and free for each heap frame each time a
340        logical recursion happens, frames are retained on a chain and re-used where
341        possible. This sometimes gives as much as 30% improvement.
343    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
344        call.
346    14. As documented, (*COMMIT) is now confined to within a positive assertion.
348    15. It is now possible to link pcretest with libedit as an alternative to
349        libreadline.
351    16. (*COMMIT) control verb is now supported by the JIT compiler.
353    17. The Unicode data tables have been updated to Unicode 6.1.0.
355    18. Added --file-list option to pcregrep.
357    19. Added binary file support to pcregrep, including the -a, --binary-files,
358        -I, and --text options.
360    20. The madvise function is renamed for posix_madvise for QNX compatibility
361        reasons. Fixed by Giuseppe D'Angelo.
363    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
365    22. Changed the option for creating symbolic links for 16-bit man pages from
366        -s to -sf so that re-installing does not cause issues.
368    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
370    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
371        matches in certain environments (the workspace was not being correctly
372        retained). Also added to pcre_dfa_exec() a simple plausibility check on
373        some of the workspace data at the beginning of a restart.
375    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
376        was not doing so when it should - probably a typo introduced by SVN 528
377        (change 8.10/14).
379    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
380        \w+ when the character tables indicated that \x{c4} was a word character.
381        There were several related cases, all because the tests for doing a table
382        lookup were testing for characters less than 127 instead of 255.
384    27. If a pattern contains capturing parentheses that are not used in a match,
385        their slots in the ovector are set to -1. For those that are higher than
386        any matched groups, this happens at the end of processing. In the case when
387        there were back references that the ovector was too small to contain
388        (causing temporary malloc'd memory to be used during matching), and the
389        highest capturing number was not used, memory off the end of the ovector
390        was incorrectly being set to -1. (It was using the size of the temporary
391        memory instead of the true size.)
393    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
394        ovector size, it uses memory at the end of the block that it has got.
396    29. Check for an overlong MARK name and give an error at compile time. The
397        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
399    30. JIT compiler update.
401    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
402        Rill for the patch.
404    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
406    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
408    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
409        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
410        was enabled.
412    35. Improve JIT code generation for greedy plus quantifier.
414    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
415        1 to "aa" instead of to an empty string. The bug affected repeated groups
416        that could potentially match an empty string.
418    37. Optimizing single character iterators in JIT.
420    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
421        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
422        codepoints that are too big for the mode are faulted, and in a UTF mode,
423        disallowed codepoints are also faulted.
425    39. If PCRE was compiled with UTF support, in three places in the DFA
426        matcher there was code that should only have been obeyed in UTF mode, but
427        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
428        processing when bytes with values greater than 127 were present. In 16-bit
429        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
430        both cases the values are those that cannot be the first data item in a UTF
431        character. The three items that might have provoked this were recursions,
432        possessively repeated groups, and atomic groups.
434    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
435        and pcregrep, because some OS require shared objects to be explicitly
436        passed to ld, causing the link step to fail if they are not.
438    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
439        mode, patterns that started with \h* or \R* might be incorrectly matched.
442    Version 8.30 04-February-2012
443    -----------------------------
445    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
446        name is defined in ctype.h.
448    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
449        only in quite long subpatterns.
451    3.  Removed the function pcre_info(), which has been obsolete and deprecated
452        since it was replaced by pcre_fullinfo() in February 2000.
454    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
455        match a (*MARK), and the match failed at the start of the subject, a
456        reference to memory before the start of the subject could occur. This bug
457        was introduced by fix 17 of release 8.21.
459    5.  A reference to an unset group with zero minimum repetition was giving
460        totally wrong answers (in non-JavaScript-compatibility mode). For example,
461        /(another)?(\1?)test/ matched against "hello world test". This bug was
462        introduced in release 8.13.
464    6.  Add support for 16-bit character strings (a large amount of work involving
465        many changes and refactorings).
467    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
468        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
469        from a file.
471    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
472        rounding is not applied in this particular case).
474    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
475        if they appear, or are escaped, in patterns.
477    10. Get rid of a number of -Wunused-but-set-variable warnings.
479    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
480        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
481        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
482        also returns the mark "x". This bug applied to capturing parentheses,
483        non-capturing parentheses, and atomic parentheses. It also applied to some
484        assertions.
486    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
487        information out of configure.ac instead of relying on pcre.h.generic, which
488        is not stored in the repository.
490    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
491        -lreadline.
493    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
495    15. Applied Graycode's patch to put the top-level frame on the stack rather
496        than the heap when not using the stack for recursion. This gives a
497        performance improvement in many cases when recursion is not deep.
499    16. Experimental code added to "pcretest -C" to output the stack frame size.
502    Version 8.21 12-Dec-2011
503    ------------------------
505    1.  Updating the JIT compiler.
507    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
508        are added as well.
510    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
511        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
512        calling _pcre_jit_exec. Some extra comments are added.
514    4.  (*MARK) settings inside atomic groups that do not contain any capturing
515        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
516        was introduced by change 18 for 8.20.
518    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
519        ECMA-262 standard.
521    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
522        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
523        This bug was probably introduced by change 9 of 8.13.
525    7.  While fixing 6 above, I noticed that a number of other items were being
526        incorrectly rejected as "not fixed length". This arose partly because newer
527        opcodes had not been added to the fixed-length checking code. I have (a)
528        corrected the bug and added tests for these items, and (b) arranged for an
529        error to occur if an unknown opcode is encountered while checking for fixed
530        length instead of just assuming "not fixed length". The items that were
531        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
532        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
533        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
535    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
536        being incorrectly compiled and would have given unpredicatble results.
538    9.  A possessively repeated subpattern with minimum repeat count greater than
539        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
540        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
541        the first (A) could occur when it should not.
543    10. Add a cast and remove a redundant test from the code.
545    11. JIT should use pcre_malloc/pcre_free for allocation.
547    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
548        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
549        is anything other than /usr, -L is still shown).
551    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
553    14. Perl does not support \N without a following name in a [] class; PCRE now
554        also gives an error.
556    15. If a forward reference was repeated with an upper limit of around 2000,
557        it caused the error "internal error: overran compiling workspace". The
558        maximum number of forward references (including repeats) was limited by the
559        internal workspace, and dependent on the LINK_SIZE. The code has been
560        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
561        the default depends on LINK_SIZE. There is a new upper limit (for safety)
562        of around 200,000 forward references. While doing this, I also speeded up
563        the filling in of repeated forward references.
565    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
566        incorrectly expecting the subject to contain another "a" after the start.
568    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
569        in the match, the SKIP should be ignored. This was not happening; instead
570        the SKIP was being treated as NOMATCH. For patterns such as
571        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
572        tested.
574    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
575        now much more compatible with Perl, in particular in cases where the result
576        is a non-match for a non-anchored pattern. For example, if
577        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
578        "m", where previously it did not return a name. A side effect of this
579        change is that for partial matches, the last encountered mark name is
580        returned, as for non matches. A number of tests that were previously not
581        Perl-compatible have been moved into the Perl-compatible test files. The
582        refactoring has had the pleasing side effect of removing one argument from
583        the match() function, thus reducing its stack requirements.
585    19. If the /S+ option was used in pcretest to study a pattern using JIT,
586        subsequent uses of /S (without +) incorrectly behaved like /S+.
588    21. Retrieve executable code size support for the JIT compiler and fixing
589        some warnings.
591    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
592        not work when the shorter character appeared right at the end of the
593        subject string.
595    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
596        systems.
598    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
599        output it when the /M option is used in pcretest.
601    25. The CheckMan script was not being included in the distribution. Also, added
602        an explicit "perl" to run Perl scripts from the PrepareRelease script
603        because this is reportedly needed in Windows.
605    26. If study data was being save in a file and studying had not found a set of
606        "starts with" bytes for the pattern, the data written to the file (though
607        never used) was taken from uninitialized memory and so caused valgrind to
608        complain.
610    27. Updated RunTest.bat as provided by Sheri Pierce.
612    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
614    29. Computation of memory usage for the table of capturing group names was
615        giving an unnecessarily large value.
618    Version 8.20 21-Oct-2011
619    ------------------------
621    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
622        a POSIX class. After further experiments with Perl, which convinced me that
623        Perl has bugs and confusions, a closing square bracket is no longer allowed
624        in a POSIX name. This bug also affected patterns with classes that started
625        with full stops.
627    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
628        captured substring, but while checking the failing first alternative,
629        substring 1 is temporarily captured. If the output vector supplied to
630        pcre_exec() was not big enough for this capture, the yield of the function
631        was still zero ("insufficient space for captured substrings"). This cannot
632        be totally fixed without adding another stack variable, which seems a lot
633        of expense for a edge case. However, I have improved the situation in cases
634        such as /(a)(b)x|abc/ matched against "abc", where the return code
635        indicates that fewer than the maximum number of slots in the ovector have
636        been set.
638    3.  Related to (2) above: when there are more back references in a pattern than
639        slots in the output vector, pcre_exec() uses temporary memory during
640        matching, and copies in the captures as far as possible afterwards. It was
641        using the entire output vector, but this conflicts with the specification
642        that only 2/3 is used for passing back captured substrings. Now it uses
643        only the first 2/3, for compatibility. This is, of course, another edge
644        case.
646    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
647        main code base, and can be used by building with --enable-jit. When this is
648        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
649        runtime --no-jit option is given.
651    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
652        ovector, the return from the function was zero, implying that there were
653        other matches that did not fit. The correct "exactly full" value is now
654        returned.
656    6.  If a subpattern that was called recursively or as a subroutine contained
657        (*PRUNE) or any other control that caused it to give a non-standard return,
658        invalid errors such as "Error -26 (nested recursion at the same subject
659        position)" or even infinite loops could occur.
661    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
662        computing the minimum length on reaching *ACCEPT, and so ended up with the
663        wrong value of 1 rather than 0. Further investigation indicates that
664        computing a minimum subject length in the presence of *ACCEPT is difficult
665        (think back references, subroutine calls), and so I have changed the code
666        so that no minimum is registered for a pattern that contains *ACCEPT.
668    8.  If (*THEN) was present in the first (true) branch of a conditional group,
669        it was not handled as intended. [But see 16 below.]
671    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
672        Sheri Pierce.
674    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
675        the first byte in a match must be "a".
677    11. Change 17 for 8.13 increased the recursion depth for patterns like
678        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
679        pattern contains any instances of (*THEN). If it does not, the old
680        optimizations are restored. It would be nice to do this on a per-group
681        basis, but at the moment that is not feasible.
683    12. In some environments, the output of pcretest -C is CRLF terminated. This
684        broke RunTest's code that checks for the link size. A single white space
685        character after the value is now allowed for.
687    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
688        For "fr", it uses the Windows-specific input and output files.
690    14. If (*THEN) appeared in a group that was called recursively or as a
691        subroutine, it did not work as intended. [But see next item.]
693    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
694        pattern fragments (but not containing any | characters). If A and B are
695        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
696        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
697        D. In other words, Perl considers parentheses that do not contain any |
698        characters to be part of a surrounding alternative, whereas PCRE was
699        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
700        differently. PCRE now behaves in the same way as Perl, except in the case
701        of subroutine/recursion calls such as (?1) which have in any case always
702        been different (but PCRE had them first :-).
704    16. Related to 15 above: Perl does not treat the | in a conditional group as
705        creating alternatives. Such a group is treated in the same way as an
706        ordinary group without any | characters when processing (*THEN). PCRE has
707        been changed to match Perl's behaviour.
709    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
710        RunGrepTest script failed.
712    18. Change 22 for version 13 caused atomic groups to use more stack. This is
713        inevitable for groups that contain captures, but it can lead to a lot of
714        stack use in large patterns. The old behaviour has been restored for atomic
715        groups that do not contain any capturing parentheses.
717    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
718        suppress the check for a minimum subject length at run time. (If it was
719        given to pcre_exec() or pcre_dfa_exec() it did work.)
721    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
722        fail to work when decoding hex characters in data strings in EBCDIC
723        environments.
725    21. It appears that in at least one Mac OS environment, the isxdigit() function
726        is implemented as a macro that evaluates to its argument more than once,
727        contravening the C 90 Standard (I haven't checked a later standard). There
728        was an instance in pcretest which caused it to go wrong when processing
729        \x{...} escapes in subject strings. The has been rewritten to avoid using
730        things like p++ in the argument of isxdigit().
733    Version 8.13 16-Aug-2011
734    ------------------------
736    1.  The Unicode data tables have been updated to Unicode 6.0.0.
738    2.  Two minor typos in pcre_internal.h have been fixed.
740    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
741        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
742        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
744    4.  There were a number of related bugs in the code for matching backrefences
745        caselessly in UTF-8 mode when codes for the characters concerned were
746        different numbers of bytes. For example, U+023A and U+2C65 are an upper
747        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
748        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
749        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
750        2-byte code at the end of the subject (it thought there wasn't enough data
751        left).
753    5.  Comprehensive information about what went wrong is now returned by
754        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
755        as the output vector has at least 2 elements. The offset of the start of
756        the failing character and a reason code are placed in the vector.
758    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
759        now returned is for the first byte of the failing character, instead of the
760        last byte inspected. This is an incompatible change, but I hope it is small
761        enough not to be a problem. It makes the returned offset consistent with
762        pcre_exec() and pcre_dfa_exec().
764    7.  pcretest now gives a text phrase as well as the error number when
765        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
766        failure, the offset and reason code are output.
768    8.  When \R was used with a maximizing quantifier it failed to skip backwards
769        over a \r\n pair if the subsequent match failed. Instead, it just skipped
770        back over a single character (\n). This seems wrong (because it treated the
771        two characters as a single entity when going forwards), conflicts with the
772        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
773        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
774        has been changed.
776    9.  Some internal refactoring has changed the processing so that the handling
777        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
778        time (the PCRE_DOTALL option was changed this way some time ago: version
779        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
780        which was always a bit of a fudge. It also means that there is one less
781        argument for the match() function, which reduces its stack requirements
782        slightly. This change also fixes an incompatibility with Perl: the pattern
783        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
785    10. More internal refactoring has drastically reduced the number of recursive
786        calls to match() for possessively repeated groups such as (abc)++ when
787        using pcre_exec().
789    11. While implementing 10, a number of bugs in the handling of groups were
790        discovered and fixed:
792        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
793        (a|)*(?1) gave a compile-time internal error.
794        ((a|)+)+  did not notice that the outer group could match an empty string.
795        (^a|^)+   was not marked as anchored.
796        (.*a|.*)+ was not marked as matching at start or after a newline.
798    12. Yet more internal refactoring has removed another argument from the match()
799        function. Special calls to this function are now indicated by setting a
800        value in a variable in the "match data" data block.
802    13. Be more explicit in pcre_study() instead of relying on "default" for
803        opcodes that mean there is no starting character; this means that when new
804        ones are added and accidentally left out of pcre_study(), testing should
805        pick them up.
807    14. The -s option of pcretest has been documented for ages as being an old
808        synonym of -m (show memory usage). I have changed it to mean "force study
809        for every regex", that is, assume /S for every regex. This is similar to -i
810        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
811        using it. It makes it easier to run collections of tests with and without
812        study enabled, and thereby test pcre_study() more easily. All the standard
813        tests are now run with and without -s (but some patterns can be marked as
814        "never study" - see 20 below).
816    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
817        restoration of the capturing data to the outer values was not happening
818        correctly.
820    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
821        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
822        pattern had matched an empty string, and so incorrectly returned a no
823        match.
825    17. There was optimizing code for the last branch of non-capturing parentheses,
826        and also for the obeyed branch of a conditional subexpression, which used
827        tail recursion to cut down on stack usage. Unfortunately, now that there is
828        the possibility of (*THEN) occurring in these branches, tail recursion is
829        no longer possible because the return has to be checked for (*THEN). These
830        two optimizations have therefore been removed. [But see 8.20/11 above.]
832    18. If a pattern containing \R was studied, it was assumed that \R always
833        matched two bytes, thus causing the minimum subject length to be
834        incorrectly computed because \R can also match just one byte.
836    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
837        was incorrectly computed.
839    20. If /S is present twice on a test pattern in pcretest input, it now
840        *disables* studying, thereby overriding the use of -s on the command line
841        (see 14 above). This is necessary for one or two tests to keep the output
842        identical in both cases.
844    21. When (*ACCEPT) was used in an assertion that matched an empty string and
845        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
847    22. When an atomic group that contained a capturing parenthesis was
848        successfully matched, but the branch in which it appeared failed, the
849        capturing was not being forgotten if a higher numbered group was later
850        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
851        group 1 to "a", when in fact it should be unset. This applied to multi-
852        branched capturing and non-capturing groups, repeated or not, and also to
853        positive assertions (capturing in negative assertions does not happen
854        in PCRE) and also to nested atomic groups.
856    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
857        subject after a captured substring, to make it easier to tell which of a
858        number of identical substrings has been captured.
860    24. The way atomic groups are processed by pcre_exec() has been changed so that
861        if they are repeated, backtracking one repetition now resets captured
862        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
863        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
864        "aaa". Previously, it would have been "a". As part of this code
865        refactoring, the way recursive calls are handled has also been changed.
867    25. If an assertion condition captured any substrings, they were not passed
868        back unless some other capturing happened later. For example, if
869        (?(?=(a))a) was matched against "a", no capturing was returned.
871    26. When studying a pattern that contained subroutine calls or assertions,
872        the code for finding the minimum length of a possible match was handling
873        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
874        group 1 called group 2 while simultaneously a separate group 2 called group
875        1). A stack overflow occurred in this case. I have fixed this by limiting
876        the recursion depth to 10.
878    27. Updated RunTest.bat in the distribution to the version supplied by Tom
879        Fortmann. This supports explicit test numbers on the command line, and has
880        argument validation and error reporting.
882    28. An instance of \X with an unlimited repeat could fail if at any point the
883        first character it looked at was a mark character.
885    29. Some minor code refactoring concerning Unicode properties and scripts
886        should reduce the stack requirement of match() slightly.
888    30. Added the '=' option to pcretest to check the setting of unused capturing
889        slots at the end of the pattern, which are documented as being -1, but are
890        not included in the return count.
892    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
893        compiled something random. Now it gives a compile-time error (as does
894        Perl).
896    32. A *MARK encountered during the processing of a positive assertion is now
897        recorded and passed back (compatible with Perl).
899    33. If --only-matching or --colour was set on a pcregrep call whose pattern
900        had alternative anchored branches, the search for a second match in a line
901        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
902        matched the line "0102" twice. The same bug affected patterns that started
903        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
904        twice.
906    34. Previously, PCRE did not allow quantification of assertions. However, Perl
907        does, and because of capturing effects, quantifying parenthesized
908        assertions may at times be useful. Quantifiers are now allowed for
909        parenthesized assertions.
911    35. A minor code tidy in pcre_compile() when checking options for \R usage.
913    36. \g was being checked for fancy things in a character class, when it should
914        just be a literal "g".
916    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
917        appearance of a nested POSIX class supersedes an apparent external class.
918        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
919        unescaped square brackets may also appear as part of class names. For
920        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
921        more like Perl. (But see 8.20/1 above.)
923    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
924        was because it thought it was \N{name}, which is not supported).
926    39. Add minix to OS list not supporting the -S option in pcretest.
928    40. PCRE tries to detect cases of infinite recursion at compile time, but it
929        cannot analyze patterns in sufficient detail to catch mutual recursions
930        such as ((?1))((?2)). There is now a runtime test that gives an error if a
931        subgroup is called recursively as a subpattern for a second time at the
932        same position in the subject string. In previous releases this might have
933        been caught by the recursion limit, or it might have run out of stack.
935    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
936        happen only once. PCRE was, however incorrectly giving a compile time error
937        "recursive call could loop indefinitely" because it cannot analyze the
938        pattern in sufficient detail. The compile time test no longer happens when
939        PCRE is compiling a conditional subpattern, but actual runaway loops are
940        now caught at runtime (see 40 above).
942    42. It seems that Perl allows any characters other than a closing parenthesis
943        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
944        has been changed to be the same.
946    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
947        as not to get warnings when autogen.sh is called. Also changed
948        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
950    44. To help people who use pcregrep to scan files containing exceedingly long
951        lines, the following changes have been made:
953        (a) The default value of the buffer size parameter has been increased from
954            8K to 20K. (The actual buffer used is three times this size.)
956        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
957            PCRE is built.
959        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
960            to be set at run time.
962        (d) Numerical values in pcregrep options can be followed by K or M, for
963            example --buffer-size=50K.
965        (e) If a line being scanned overflows pcregrep's buffer, an error is now
966            given and the return code is set to 2.
968    45. Add a pointer to the latest mark to the callout data block.
970    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
971        partial match of an empty string instead of no match. This was specific to
972        the use of ".".
974    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
975        complete match instead of a partial match. This bug was dependent on both
976        the PCRE_UTF8 and PCRE_DOTALL options being set.
978    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
979        starting byte set, because \b was not being ignored.
982    Version 8.12 15-Jan-2011
983    ------------------------
985    1.  Fixed some typos in the markup of the man pages, and wrote a script that
986        checks for such things as part of the documentation building process.
988    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
989        --match-limit and --recursion-limit options (added for 8.11). In
990        particular, this made one of the standard tests fail. (The integer value
991        went into the wrong half of a long int.)
993    3.  If the --colour option was given to pcregrep with -v (invert match), it
994        did strange things, either producing crazy output, or crashing. It should,
995        of course, ignore a request for colour when reporting lines that do not
996        match.
998    4.  Another pcregrep bug caused similar problems if --colour was specified with
999        -M (multiline) and the pattern match finished with a line ending.
1001    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1002        matched in multiline mode, the following line was shown as part of the
1003        match. This seems wrong, so I have changed it.
1005    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1006        the check for further matches in the same line (so they could be coloured)
1007        to overrun the end of the current line. If another match was found, it was
1008        incorrectly shown (and then shown again when found in the next line).
1010    7.  If pcregrep was compiled under Windows, there was a reference to the
1011        function pcregrep_exit() before it was defined. I am assuming this was
1012        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1013        reported by a user. I've moved the definition above the reference.
1016    Version 8.11 10-Dec-2010
1017    ------------------------
1019    1.  (*THEN) was not working properly if there were untried alternatives prior
1020        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1021        backtracked to try for "b" instead of moving to the next alternative branch
1022        at the same level (in this case, to look for "c"). The Perl documentation
1023        is clear that when (*THEN) is backtracked onto, it goes to the "next
1024        alternative in the innermost enclosing group".
1026    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1027        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1028        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1029        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1030        (*THEN).
1032    3.  If \s appeared in a character class, it removed the VT character from
1033        the class, even if it had been included by some previous item, for example
1034        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1035        of \s, but is part of the POSIX "space" class.)
1037    4.  A partial match never returns an empty string (because you can always
1038        match an empty string at the end of the subject); however the checking for
1039        an empty string was starting at the "start of match" point. This has been
1040        changed to the "earliest inspected character" point, because the returned
1041        data for a partial match starts at this character. This means that, for
1042        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1043        (previously it gave "no match").
1045    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1046        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1047        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1048        has an implication that the given string is incomplete (because a partial
1049        match is preferred over a full match). For this reason, these items now
1050        give a partial match in this situation. [Aside: previously, the one case
1051        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1052        match rather than a full match, which was wrong by the old rules, but is
1053        now correct.]
1055    6.  There was a bug in the handling of #-introduced comments, recognized when
1056        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1057        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1058        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1059        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1060        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1061        places in pcre_compile().
1063    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1064        comments when looking ahead for named forward references to subpatterns,
1065        the only newline sequence it recognized was NL. It now handles newlines
1066        according to the set newline convention.
1068    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1069        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1070        cater for a lack of strerror(). These oversights have been fixed.
1072    9.  Added --match-limit and --recursion-limit to pcregrep.
1074    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1076    11. When the -o option was used, pcregrep was setting a return code of 1, even
1077        when matches were found, and --line-buffered was not being honoured.
1079    12. Added an optional parentheses number to the -o and --only-matching options
1080        of pcregrep.
1082    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1083        can match an empty string. The code to do it in pcretest and pcredemo
1084        needed fixing:
1086        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1087            only one byte after an empty string match just before CRLF (this case
1088            just got forgotten; "any" and "anycrlf" were OK).
1090        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1091            mode when an empty string match preceded an ASCII character followed by
1092            a non-ASCII character. (The code for advancing by one character rather
1093            than one byte was nonsense.)
1095        (c) The pcredemo.c sample program did not have any code at all to handle
1096            the cases when CRLF is a valid newline sequence.
1098    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1099        as a starting offset was within the subject string. There is now a new
1100        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1101        negative or greater than the length of the string. In order to test this,
1102        pcretest is extended to allow the setting of negative starting offsets.
1104    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1105        starting offset points to the beginning of a UTF-8 character was
1106        unnecessarily clumsy. I tidied it up.
1108    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1109        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1111    17. Nobody had reported that the --include_dir option, which was added in
1112        release 7.7 should have been called --include-dir (hyphen, not underscore)
1113        for compatibility with GNU grep. I have changed it to --include-dir, but
1114        left --include_dir as an undocumented synonym, and the same for
1115        --exclude-dir, though that is not available in GNU grep, at least as of
1116        release 2.5.4.
1118    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1119        characters from a string of bytes) have been redefined so as not to use
1120        loops, in order to improve performance in some environments. At the same
1121        time, I abstracted some of the common code into auxiliary macros to save
1122        repetition (this should not affect the compiled code).
1124    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1125        compile-time error is now given if \c is not followed by an ASCII
1126        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1127        different, and any byte value is allowed.)
1129    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1130        START_OPTIMIZE option, which is now allowed at compile time - but just
1131        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1132        to pcregrep and other applications that have no direct access to PCRE
1133        options. The new /Y option in pcretest sets this option when calling
1134        pcre_compile().
1136    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1137        back references. Groups containing recursive back references were forced to
1138        be atomic by that change, but in the case of named groups, the amount of
1139        memory required was incorrectly computed, leading to "Failed: internal
1140        error: code overflow". This has been fixed.
1142    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1143        pcretest.c, to avoid build problems in some Borland environments.
1146    Version 8.10 25-Jun-2010
1147    ------------------------
1149    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1150        THEN.
1152    2.  (*ACCEPT) was not working when inside an atomic group.
1154    3.  Inside a character class, \B is treated as a literal by default, but
1155        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1156        causes the error). The code is unchanged, but I tidied the documentation.
1158    4.  Inside a character class, PCRE always treated \R and \X as literals,
1159        whereas Perl faults them if its -w option is set. I have changed PCRE so
1160        that it faults them when PCRE_EXTRA is set.
1162    5.  Added support for \N, which always matches any character other than
1163        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1165    6.  When compiling pcregrep with newer versions of gcc which may have
1166        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1167        declared with attribute warn_unused_result" were given. Just casting the
1168        result to (void) does not stop the warnings; a more elaborate fudge is
1169        needed. I've used a macro to implement this.
1171    7.  Minor change to pcretest.c to avoid a compiler warning.
1173    8.  Added four artifical Unicode properties to help with an option to make
1174        \s etc use properties (see next item). The new properties are: Xan
1175        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1177    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1178        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1179        this option. Modified pcretest to add /W to test this facility. Added
1180        REG_UCP to make it available via the POSIX interface.
1182    10. Added --line-buffered to pcregrep.
1184    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1185        studied, and the match started with a letter with a code point greater than
1186        127 whose first byte was different to the first byte of the other case of
1187        the letter, the other case of this starting letter was not recognized
1188        (#976).
1190    12. If a pattern that was studied started with a repeated Unicode property
1191        test, for example, \p{Nd}+, there was the theoretical possibility of
1192        setting up an incorrect bitmap of starting bytes, but fortunately it could
1193        not have actually happened in practice until change 8 above was made (it
1194        added property types that matched character-matching opcodes).
1196    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1197        possible starting bytes for non-anchored patterns.
1199    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1200        \R, and also a number of cases that involve Unicode properties, both
1201        explicit and implicit when PCRE_UCP is set.
1203    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1204        input, it could crash or give wrong results if characters with values
1205        greater than 0xc0 were present in the subject string. (Detail: it assumed
1206        UTF-8 input when processing these items.)
1208    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1209        size_t is 64-bit (#991).
1211    17. Added a check for running out of memory when PCRE is compiled with
1212        --disable-stack-for-recursion (#990).
1214    18. If the last data line in a file for pcretest does not have a newline on
1215        the end, a newline was missing in the output.
1217    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1218        less than 128) in its various bitmaps. However, there is a facility for
1219        generating tables according to the current locale when PCRE is compiled. It
1220        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1221        characters, are recognized by isspace() and therefore were getting set in
1222        these tables, and indeed these tables seem to approximate to ISO 8859. This
1223        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1224        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1225        which of course cannot start UTF-8 characters. I have changed the code so
1226        that only real ASCII characters (less than 128) and the correct starting
1227        bytes for UTF-8 encodings are set for characters greater than 127 when in
1228        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1229        altogether.)
1231    20. Added the /T option to pcretest so as to be able to run tests with non-
1232        standard character tables, thus making it possible to include the tests
1233        used for 19 above in the standard set of tests.
1235    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1236        reference to a subpattern the other side of a comment that contains an
1237        opening parenthesis caused either an internal compiling error, or a
1238        reference to the wrong subpattern.
1241    Version 8.02 19-Mar-2010
1242    ------------------------
1244    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1246    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1247        configured.
1249    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1250        original author of that file, following a query about its status.
1252    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1253        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1255    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1256        quantifier applied to a forward-referencing subroutine call, could compile
1257        incorrect code or give the error "internal error: previously-checked
1258        referenced subpattern not found".
1260    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1261        variables to point to external functions. For these systems, therefore,
1262        pcre_malloc etc. are now initialized to local functions that call the
1263        relevant global functions.
1265    7.  There were two entries missing in the vectors called coptable and poptable
1266        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1267        I've fixed the data, and added a kludgy way of testing at compile time that
1268        the lengths are correct (equal to the number of opcodes).
1270    8.  Following on from 7, I added a similar kludge to check the length of the
1271        eint vector in pcreposix.c.
1273    9.  Error texts for pcre_compile() are held as one long string to avoid too
1274        much relocation at load time. To find a text, the string is searched,
1275        counting zeros. There was no check for running off the end of the string,
1276        which could happen if a new error number was added without updating the
1277        string.
1279    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1281    11. \K was not working if it appeared in an atomic group or in a group that
1282        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1283        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1284        the assertion is positive, but not if it is negative.
1286    12. Change 11 fortuitously reduced the size of the stack frame used in the
1287        "match()" function of pcre_exec.c by one pointer. Forthcoming
1288        implementation of support for (*MARK) will need an extra pointer on the
1289        stack; I have reserved it now, so that the stack frame size does not
1290        decrease.
1292    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1293        item in branch that calls a recursion is a subroutine call - as in the
1294        second branch in the above example - was incorrectly given the compile-
1295        time error "recursive call could loop indefinitely" because pcre_compile()
1296        was not correctly checking the subroutine for matching a non-empty string.
1298    14. The checks for overrunning compiling workspace could trigger after an
1299        overrun had occurred. This is a "should never occur" error, but it can be
1300        triggered by pathological patterns such as hundreds of nested parentheses.
1301        The checks now trigger 100 bytes before the end of the workspace.
1303    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1306    Version 8.01 19-Jan-2010
1307    ------------------------
1309    1.  If a pattern contained a conditional subpattern with only one branch (in
1310        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1311        computed the wrong minimum data length (which is of course zero for such
1312        subpatterns). This could cause incorrect "no match" results.
1314    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1315        the pattern is reset in the first branch, pcre_compile() failed with
1316        "internal error: code overflow at offset...". This happened only when
1317        the reset was to the original external option setting. (An optimization
1318        abstracts leading options settings into an external setting, which was the
1319        cause of this.)
1321    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1322        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1323        assertion pattern did not match (meaning that the assertion was true), it
1324        was incorrectly treated as false if the SKIP had been reached during the
1325        matching. This also applied to assertions used as conditions.
1327    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1328        assertion subpattern, including such a pattern used as a condition,
1329        unpredictable results occurred, instead of the error return
1332    5.  The C++ GlobalReplace function was not working like Perl for the special
1333        situation when an empty string is matched. It now does the fancy magic
1334        stuff that is necessary.
1336    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1337        removed. (These were left over from very, very early versions of PCRE.)
1339    7.  Some cosmetic changes to the code to make life easier when compiling it
1340        as part of something else:
1342        (a) Change DEBUG to PCRE_DEBUG.
1344        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1345            called "current" as "current_branch", to prevent a collision with the
1346            Linux macro when compiled as a kernel module.
1348        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1349            prevent a collision with the Linux macro when compiled as a kernel
1350            module.
1352    8.  In pcre_compile() there are some checks for integer overflows that used to
1353        cast potentially large values to (double). This has been changed to that
1354        when building, a check for int64_t is made, and if it is found, it is used
1355        instead, thus avoiding the use of floating point arithmetic. (There is no
1356        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1357        double.
1359    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1360        2005 (difference between two addresses compared to an unsigned value).
1362    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1363        custom one, because of the following reported problem in Windows:
1365          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1366              under Win32.
1367          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1368              therefore missing the function definition.
1369          - The compiler thus generates a "C" signature for the test function.
1370          - The linker fails to find the "C" function.
1371          - PCRE fails to configure if asked to do so against libbz2.
1373    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1374        messages were output:
1376          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1377          rerunning libtoolize, to keep the correct libtool macros in-tree.
1378          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1380        I have done both of these things.
1382    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1383        most of the time, it *can* run out if it is given a pattern that contains a
1384        runaway infinite recursion. I updated the discussion in the pcrestack man
1385        page.
1387    13. Now that we have gone to the x.xx style of version numbers, the minor
1388        version may start with zero. Using 08 or 09 is a bad idea because users
1389        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1390        interpreted as invalid octal numbers. I've updated the previous comment in
1391        configure.ac, and also added a check that gives an error if 08 or 09 are
1392        used.
1394    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1395        causing partial matching to fail when the end of the subject matched \W
1396        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1398    15. There were some discrepancies between the declarations in pcre_internal.h
1399        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1400        their definitions. The declarations used "const uschar *" and the
1401        definitions used USPTR. Even though USPTR is normally defined as "const
1402        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1403        reported that: "This difference in casting confuses some C++ compilers, for
1404        example, SunCC recognizes above declarations as different functions and
1405        generates broken code for hbpcre." I have changed the declarations to use
1406        USPTR.
1408    16. GNU libtool is named differently on some systems. The autogen.sh script now
1409        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1410        (FreeBSD).
1412    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1413        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1414        comment: "Figure out how to create a longlong from a string: strtoll and
1415        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1416        instance, but it only takes 2 args instead of 3!"
1418    18. A subtle bug concerned with back references has been fixed by a change of
1419        specification, with a corresponding code fix. A pattern such as
1420        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1421        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1422        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1423        same bug. Such groups have to be quantified to be useful, or contained
1424        inside another quantified group. (If there's no repetition, the reference
1425        can never match.) The problem arises because, having left the group and
1426        moved on to the rest of the pattern, a later failure that backtracks into
1427        the group uses the captured value from the final iteration of the group
1428        rather than the correct earlier one. I have fixed this in PCRE by forcing
1429        any group that contains a reference to itself to be an atomic group; that
1430        is, there cannot be any backtracking into it once it has completed. This is
1431        similar to recursive and subroutine calls.
1434    Version 8.00 19-Oct-09
1435    ----------------------
1437    1.  The table for translating pcre_compile() error codes into POSIX error codes
1438        was out-of-date, and there was no check on the pcre_compile() error code
1439        being within the table. This could lead to an OK return being given in
1440        error.
1442    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1443        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1444        in a Windows environment.
1446    3.  The pcregrep --count option prints the count for each file even when it is
1447        zero, as does GNU grep. However, pcregrep was also printing all files when
1448        --files-with-matches was added. Now, when both options are given, it prints
1449        counts only for those files that have at least one match. (GNU grep just
1450        prints the file name in this circumstance, but including the count seems
1451        more useful - otherwise, why use --count?) Also ensured that the
1452        combination -clh just lists non-zero counts, with no names.
1454    4.  The long form of the pcregrep -F option was incorrectly implemented as
1455        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1456        but it seems right to fix it, and I didn't think it was worth preserving
1457        the old behaviour.
1459    5.  The command line items --regex=pattern and --regexp=pattern were not
1460        recognized by pcregrep, which required --regex pattern or --regexp pattern
1461        (with a space rather than an '='). The man page documented the '=' forms,
1462        which are compatible with GNU grep; these now work.
1464    6.  No libpcreposix.pc file was created for pkg-config; there was just
1465        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1467    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1468        when UCP support is not needed, by modifying the Python script that
1469        generates it from Unicode data files. This should not matter if the module
1470        is correctly used as a library, but I received one complaint about 50K of
1471        unwanted data. My guess is that the person linked everything into his
1472        program rather than using a library. Anyway, it does no harm.
1474    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1475        was a minimum greater than 1 for a wide character in a possessive
1476        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1477        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1478        character. Chaos in the form of incorrect output or a compiling loop could
1479        result.
1481    9.  The restrictions on what a pattern can contain when partial matching is
1482        requested for pcre_exec() have been removed. All patterns can now be
1483        partially matched by this function. In addition, if there are at least two
1484        slots in the offset vector, the offset of the earliest inspected character
1485        for the match and the offset of the end of the subject are set in them when
1486        PCRE_ERROR_PARTIAL is returned.
1488    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1489        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1490        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1491        and may be more useful for multi-segment matching.
1493    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1494        used to be given if ever the end of the subject was reached; now it is
1495        given only if matching could not proceed because another character was
1496        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1497        string "Z", which now yields "no match" instead of "partial match". In the
1498        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1499        final character ended with (*FAIL).
1501    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1502        if the pattern had a "must contain" character that was already found in the
1503        earlier partial match, unless partial matching was again requested. For
1504        example, with the pattern /dog.(body)?/, the "must contain" character is
1505        "g". If the first part-match was for the string "dog", restarting with
1506        "sbody" failed. This bug has been fixed.
1508    13. The string returned by pcre_dfa_exec() after a partial match has been
1509        changed so that it starts at the first inspected character rather than the
1510        first character of the match. This makes a difference only if the pattern
1511        starts with a lookbehind assertion or \b or \B (\K is not supported by
1512        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1513        matching functions compatible, and I think it's the right thing to do.
1515    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1516        so that the demonstration program is easily available in environments where
1517        PCRE has not been installed from source.
1519    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1520        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1521        library.
1523    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1524        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1525        is not the first non-POSIX option to be added. Clearly some people find
1526        these options useful.
1528    17. If a caller to the POSIX matching function regexec() passes a non-zero
1529        value for nmatch with a NULL value for pmatch, the value of
1530        nmatch is forced to zero.
1532    18. RunGrepTest did not have a test for the availability of the -u option of
1533        the diff command, as RunTest does. It now checks in the same way as
1534        RunTest, and also checks for the -b option.
1536    19. If an odd number of negated classes containing just a single character
1537        interposed, within parentheses, between a forward reference to a named
1538        subpattern and the definition of the subpattern, compilation crashed with
1539        an internal error, complaining that it could not find the referenced
1540        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1541        [The bug was that it was starting one character too far in when skipping
1542        over the character class, thus treating the ] as data rather than
1543        terminating the class. This meant it could skip too much.]
1545    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1546        /g option in pcretest when the pattern contains \K, which makes it possible
1547        to have an empty string match not at the start, even when the pattern is
1548        anchored. Updated pcretest and pcredemo to use this option.
1550    21. If the maximum number of capturing subpatterns in a recursion was greater
1551        than the maximum at the outer level, the higher number was returned, but
1552        with unset values at the outer level. The correct (outer level) value is
1553        now given.
1555    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1556        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1557        make it do so. The string so far is captured, making this feature
1558        compatible with Perl.
1560    23. The tests have been re-organized, adding tests 11 and 12, to make it
1561        possible to check the Perl 5.10 features against Perl 5.10.
1563    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1564        pattern matches a fixed length string. PCRE did not allow this; now it
1565        does. Neither allows recursion.
1567    25. I finally figured out how to implement a request to provide the minimum
1568        length of subject string that was needed in order to match a given pattern.
1569        (It was back references and recursion that I had previously got hung up
1570        on.) This code has now been added to pcre_study(); it finds a lower bound
1571        to the length of subject needed. It is not necessarily the greatest lower
1572        bound, but using it to avoid searching strings that are too short does give
1573        some useful speed-ups. The value is available to calling programs via
1574        pcre_fullinfo().
1576    26. While implementing 25, I discovered to my embarrassment that pcretest had
1577        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1578        study optimizations had never been tested with that matching function.
1579        Oops. What is worse, even when it was passed study data, there was a bug in
1580        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1581        were also very few tests of studied patterns with pcre_dfa_exec().
1583    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1584        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1585        on the other side of the coin, they are no longer allowed to have different
1586        names, because these cannot be distinguished in PCRE, and this has caused
1587        confusion. (This is a difference from Perl.)
1589    28. When duplicate subpattern names are present (necessarily with different
1590        numbers, as required by 27 above), and a test is made by name in a
1591        conditional pattern, either for a subpattern having been matched, or for
1592        recursion in such a pattern, all the associated numbered subpatterns are
1593        tested, and the overall condition is true if the condition is true for any
1594        one of them. This is the way Perl works, and is also more like the way
1595        testing by number works.
1598    Version 7.9 11-Apr-09
1599    ---------------------
1601    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1602        (pcretest), all targets were linked against these libraries. This included
1603        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1604        libraries. This caused unwanted dependencies to be created. This problem
1605        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1606        pcretest is linked with readline.
1608    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1609        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1610        moved outside it again, because FALSE and TRUE are already defined in AIX,
1611        but BOOL is not.
1613    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1614        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1616    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1617        hyphens) following file names and line numbers when outputting matching
1618        lines. This is not true; no spaces are inserted. I have also clarified the
1619        wording for the --colour (or --color) option.
1621    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1622        was not coloured; this is different to GNU grep, so I have changed it to be
1623        the same.
1625    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1626        each matching line was coloured. Now it goes on to look for further matches
1627        of any of the test patterns, which is the same behaviour as GNU grep.
1629    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1630        doesn't make sense to accept an empty string match in pcregrep, so I have
1631        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1632        seems to be how GNU grep behaves.
1634    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1635        start or after a newline", because the conditional assertion was not being
1636        correctly handled. The rule now is that both the assertion and what follows
1637        in the first alternative must satisfy the test.
1639    9.  If auto-callout was enabled in a pattern with a conditional group whose
1640        condition was an assertion, PCRE could crash during matching, both with
1641        pcre_exec() and pcre_dfa_exec().
1643    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1644        used for matching.
1646    11. Unicode property support in character classes was not working for
1647        characters (bytes) greater than 127 when not in UTF-8 mode.
1649    12. Added the -M command line option to pcretest.
1651    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1653    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1655    16. Added comments and documentation about mis-use of no_arg in the C++
1656        wrapper.
1658    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1659        from Martin Jerabek that uses macro names for all relevant character and
1660        string constants.
1662    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1663        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1664        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1665        these, but not everybody uses configure.
1667    19. A conditional group that had only one branch was not being correctly
1668        recognized as an item that could match an empty string. This meant that an
1669        enclosing group might also not be so recognized, causing infinite looping
1670        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1671        with the subject "ab", where knowledge that the repeated group can match
1672        nothing is needed in order to break the loop.
1674    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1675        exec(), but without supplying a callout function, matching went wrong.
1677    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1678        leak if the size of the offset vector was greater than 30. When the vector
1679        is smaller, the saved offsets during recursion go onto a local stack
1680        vector, but for larger vectors malloc() is used. It was failing to free
1681        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1682        error, in fact).
1684    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1685        heapframe that is used only when UTF-8 support is enabled. This caused no
1686        problem, but was untidy.
1688    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1689        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1690        included within another project.
1692    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1693        slightly modified by me:
1695          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1696              not building pcregrep.
1698          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1699              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1701    25. Forward references, both numeric and by name, in patterns that made use of
1702        duplicate group numbers, could behave incorrectly or give incorrect errors,
1703        because when scanning forward to find the reference group, PCRE was not
1704        taking into account the duplicate group numbers. A pattern such as
1705        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1707    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1708        the feature of a custom pointer more persuasive (as requested by a user).
1710    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1711        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1712        pre-defined.
1714    28. Added support for (*UTF8) at the start of a pattern.
1716    29. Arrange for flags added by the "release type" setting in CMake to be shown
1717        in the configuration summary.
1720    Version 7.8 05-Sep-08
1721    ---------------------
1723    1.  Replaced UCP searching code with optimized version as implemented for Ad
1724        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1725        stage table and inline lookup instead of a function, giving speed ups of 2
1726        to 5 times on some simple patterns that I tested. Permission was given to
1727        distribute the MultiStage2.py script that generates the tables (it's not in
1728        the tarball, but is in the Subversion repository).
1730    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1731        scripts.
1733    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1734        a group with a zero qualifier. The result of the study could be incorrect,
1735        or the function might crash, depending on the pattern.
1737    4.  Caseless matching was not working for non-ASCII characters in back
1738        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1739        It now works when Unicode Property Support is available.
1741    5.  In pcretest, an escape such as \x{de} in the data was always generating
1742        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1743        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1744        truncation.
1746    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1748    7.  Added two (int) casts to pcregrep when printing the difference of two
1749        pointers, in case they are 64-bit values.
1751    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1752        test 2 if it fails.
1754    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1755        and a #define of that name to empty if it is not externally set. This is to
1756        allow users of MSVC to set it if necessary.
1758    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1759        the convenience functions in the pcre_get.c source file.
1761    11. An option change at the start of a pattern that had top-level alternatives
1762        could cause overwriting and/or a crash. This command provoked a crash in
1763        some environments:
1765          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1767        This potential security problem was recorded as CVE-2008-2371.
1769    12. For a pattern where the match had to start at the beginning or immediately
1770        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1771        pcre_dfa_exec() could read past the end of the passed subject if there was
1772        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1773        pcretest so that it places the subject at the end of its malloc-ed buffer.
1775    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1776        exec() might read past the end of the data buffer in UTF-8 mode.
1778    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1779        the data contained the byte 0x85 as part of a UTF-8 character within its
1780        first line. This applied both to normal and DFA matching.
1782    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1783        /^[^d]*?$/8 failed to match "abc".
1785    16. Added a missing copyright notice to pcrecpp_internal.h.
1787    17. Make it more clear in the documentation that values returned from
1788        pcre_exec() in ovector are byte offsets, not character counts.
1790    18. Tidied a few places to stop certain compilers from issuing warnings.
1792    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1793        supplied by Stefan Weber. I made a further small update for 7.8 because
1794        there is a change of source arrangements: the pcre_searchfuncs.c module is
1795        replaced by pcre_ucd.c.
1798    Version 7.7 07-May-08
1799    ---------------------
1801    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1802        a string to a long long, pretend we don't even have a long long." This is
1803        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1805    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1806        pre-7.6 versions, which defined a global no_arg variable instead of putting
1807        it in the RE class. (See also #8 below.)
1809    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1810        Lopes.
1812    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1814        (1) The include/exclude patterns were being applied to the whole pathnames
1815            of files, instead of just to the final components.
1817        (2) If there was more than one level of directory, the subdirectories were
1818            skipped unless they satisfied the include/exclude conditions. This is
1819            inconsistent with GNU grep (and could even be seen as contrary to the
1820            pcregrep specification - which I improved to make it absolutely clear).
1821            The action now is always to scan all levels of directory, and just
1822            apply the include/exclude patterns to regular files.
1824    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1825        --exclude_dir in the tests to avoid scanning .svn directories.
1827    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1828        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1829        doesn't support NULs in patterns.
1831    7.  Added some missing "const"s to declarations of static tables in
1832        pcre_compile.c and pcre_dfa_exec.c.
1834    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1835        caused by fix #2  above. (Subsequently also a second patch to fix the
1836        first patch. And a third patch - this was a messy problem.)
1838    9.  Applied Craig's patch to remove the use of push_back().
1840    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1841        matching function regexec().
1843    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1844        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1845        references. PCRE supports relative numbers with this syntax (I don't think
1846        Oniguruma does).
1848    12. Previously, a group with a zero repeat such as (...){0} was completely
1849        omitted from the compiled regex. However, this means that if the group
1850        was called as a subroutine from elsewhere in the pattern, things went wrong
1851        (an internal error was given). Such groups are now left in the compiled
1852        pattern, with a new opcode that causes them to be skipped at execution
1853        time.
1855    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1856        to the way PCRE behaves:
1858        (a) A lone ] character is dis-allowed (Perl treats it as data).
1860        (b) A back reference to an unmatched subpattern matches an empty string
1861            (Perl fails the current match path).
1863        (c) A data ] in a character class must be notated as \] because if the
1864            first data character in a class is ], it defines an empty class. (In
1865            Perl it is not possible to have an empty class.) The empty class []
1866            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1867            The negative empty class [^] matches any one character, independently
1868            of the DOTALL setting.
1870    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1871        non-existent subpattern following a character class starting with ']' and
1872        containing () gave an internal compiling error instead of "reference to
1873        non-existent subpattern". Fortunately, when the pattern did exist, the
1874        compiled code was correct. (When scanning forwards to check for the
1875        existencd of the subpattern, it was treating the data ']' as terminating
1876        the class, so got the count wrong. When actually compiling, the reference
1877        was subsequently set up correctly.)
1879    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1880        it was being rejected as not supported by pcre_dfa_exec(), even though
1881        other assertions are supported. I have made pcre_dfa_exec() support
1882        (*FAIL).
1884    16. The implementation of 13c above involved the invention of a new opcode,
1885        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1886        cannot be changed at match time, I realized I could make a small
1887        improvement to matching performance by compiling OP_ALLANY instead of
1888        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1889        on the OP_ANY path.
1891    17. Compiling pcretest on Windows with readline support failed without the
1892        following two fixes: (1) Make the unistd.h include conditional on
1893        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1895    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1896        ncurses library to be included for pcretest when ReadLine support is
1897        requested, but also to allow for it to be overridden. This patch came from
1898        Daniel Bergström.
1900    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1901        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1902        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1903        spotting this.
1906    Version 7.6 28-Jan-08
1907    ---------------------
1909    1.  A character class containing a very large number of characters with
1910        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1911        overflow.
1913    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1914        HAVE_LONG_LONG is not defined.
1916    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1917        bring them up to date and include new features. This patch includes:
1919        - Fixed PH's badly added libz and libbz2 support.
1920        - Fixed a problem with static linking.
1921        - Added pcredemo. [But later removed - see 7 below.]
1922        - Fixed dftables problem and added an option.
1923        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1924            HAVE_LONG_LONG.
1925        - Added readline support for pcretest.
1926        - Added an listing of the option settings after cmake has run.
1928    4.  A user submitted a patch to Makefile that makes it easy to create
1929        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1930        Makefile.am that cause it to include this special target, without
1931        affecting anything else. Note that the same mingw target plus all
1932        the other distribution libraries and programs are now supported
1933        when configuring with CMake (see 6 below) instead of with
1934        Configure/Make.
1936    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1937        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1938        exported in the Windows port". It has not yet been confirmed that the patch
1939        solves the problem, but it does no harm.
1941    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1942        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1943        with CMake, and also correct the comment about stack recursion.
1945    7.  Remove the automatic building of pcredemo from the ./configure system and
1946        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1947        of a program that users should build themselves after PCRE is installed, so
1948        building it automatically is not really right. What is more, it gave
1949        trouble in some build environments.
1951    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1954    Version 7.5 10-Jan-08
1955    ---------------------
1957    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1958        values in parens when parsing an RE using the C++ wrapper."
1960    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1961        Characters greater than 255 were excluded from the class instead of being
1962        included.
1964    3.  The same bug as (2) above applied to negated POSIX classes such as
1965        [:^space:].
1967    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1968        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1969        I have changed it.
1971    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1972        first named subpattern) and a construct such as (?&a) would reference the
1973        first named subpattern whose name started with "a" (in other words, the
1974        length check was missing). Both these problems are fixed. "Subpattern name
1975        expected" is now given for (?&) (a zero-length name), and this patch also
1976        makes it give the same error for \k'' (previously it complained that that
1977        was a reference to a non-existent subpattern).
1979    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1980        this is right because (?- can be followed by option settings as well as by
1981        digits. I have, however, made the messages clearer.
1983    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1984        than the number used in the conditional) now cause a compile-time error.
1985        This is actually not compatible with Perl, which accepts such patterns, but
1986        treats the conditional as always being FALSE (as PCRE used to), but it
1987        seems to me that giving a diagnostic is better.
1989    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1990        and messages.
1992    9.  Fix two occurrences of "backslash" in comments that should have been
1993        "backspace".
1995    10. Remove two redundant lines of code that can never be obeyed (their function
1996        was moved elsewhere).
1998    11. The program that makes PCRE's Unicode character property table had a bug
1999        which caused it to generate incorrect table entries for sequences of
2000        characters that have the same character type, but are in different scripts.
2001        It amalgamated them into a single range, with the script of the first of
2002        them. In other words, some characters were in the wrong script. There were
2003        thirteen such cases, affecting characters in the following ranges:
2005          U+002b0 - U+002c1
2006          U+0060c - U+0060d
2007          U+0061e - U+00612
2008          U+0064b - U+0065e
2009          U+0074d - U+0076d
2010          U+01800 - U+01805
2011          U+01d00 - U+01d77
2012          U+01d9b - U+01dbf
2013          U+0200b - U+0200f
2014          U+030fc - U+030fe
2015          U+03260 - U+0327f
2016          U+0fb46 - U+0fbb1
2017          U+10450 - U+1049d
2019    12. The -o option (show only the matching part of a line) for pcregrep was not
2020        compatible with GNU grep in that, if there was more than one match in a
2021        line, it showed only the first of them. It now behaves in the same way as
2022        GNU grep.
2024    13. If the -o and -v options were combined for pcregrep, it printed a blank
2025        line for every non-matching line. GNU grep prints nothing, and pcregrep now
2026        does the same. The return code can be used to tell if there were any
2027        non-matching lines.
2029    14. Added --file-offsets and --line-offsets to pcregrep.
2031    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
2032        infinitely looping recursion. The bug was that positive lookaheads were not
2033        being skipped when checking for a possible empty match (negative lookaheads
2034        and both kinds of lookbehind were skipped).
2036    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
2037        inclusion of <windows.h> to before rather than after the definition of
2038        INVALID_FILE_ATTRIBUTES (patch from David Byron).
2040    17. Specifying a possessive quantifier with a specific limit for a Unicode
2041        character property caused pcre_compile() to compile bad code, which led at
2042        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
2043        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
2044        caused the error; without that there was no problem.
2046    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
2048    19. Added --enable-pcretest-libreadline.
2050    20. In pcrecpp.cc, the variable 'count' was incremented twice in
2051        RE::GlobalReplace(). As a result, the number of replacements returned was
2052        double what it should be. I removed one of the increments, but Craig sent a
2053        later patch that removed the other one (the right fix) and added unit tests
2054        that check the return values (which was not done before).
2056    21. Several CMake things:
2058        (1) Arranged that, when cmake is used on Unix, the libraries end up with
2059            the names libpcre and libpcreposix, not just pcre and pcreposix.
2061        (2) The above change means that pcretest and pcregrep are now correctly
2062            linked with the newly-built libraries, not previously installed ones.
2066    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
2067        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
2068        UTF-8 newline character). The key issue is that the pattern starts .*;
2069        this means that the match must be either at the beginning, or after a
2070        newline. The bug was in the code for advancing after a failed match and
2071        checking that the new position followed a newline. It was not taking
2072        account of UTF-8 characters correctly.
2074    23. PCRE was behaving differently from Perl in the way it recognized POSIX
2075        character classes. PCRE was not treating the sequence [:...:] as a
2076        character class unless the ... were all letters. Perl, however, seems to
2077        allow any characters between [: and :], though of course it rejects as
2078        unknown any "names" that contain non-letters, because all the known class
2079        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
2080        for example, whereas PCRE did not - it did not recognize a POSIX character
2081        class. This seemed a bit dangerous, so the code has been changed to be
2082        closer to Perl. The behaviour is not identical to Perl, because PCRE will
2083        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
2084        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
2085        Perl does, and where it didn't before.
2087    24. Rewrite so as to remove the single use of %n from pcregrep because in some
2088        Windows environments %n is disabled by default.
2091    Version 7.4 21-Sep-07
2092    ---------------------
2094    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
2095        means that a class such as [\s] counted as "explicit reference to CR or
2096        LF". That isn't really right - the whole point of the change was to try to
2097        help when there was an actual mention of one of the two characters. So now
2098        the change happens only if \r or \n (or a literal CR or LF) character is
2099        encountered.
2101    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
2102        of both had grown to the point where there were only 3 bits left.
2103        Fortunately, there was spare space in the data structure, and so I have
2104        moved the internal flags into a new 16-bit field to free up more option
2105        bits.
2107    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
2108        but did not set the internal JCHANGED flag - either of these is enough to
2109        control the way the "get" function works - but the PCRE_INFO_JCHANGED
2110        facility is supposed to tell if (?J) was ever used, so now (?J) at the
2111        start sets both bits.
2113    4.  Added options (at build time, compile time, exec time) to change \R from
2114        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
2116    5.  doc/pcresyntax.html was missing from the distribution.
2118    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
2119        compatibility, even though it is no longer used.
2121    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
2122        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
2123        windows.h file is present (where different names are used). [This was
2124        reversed later after testing - see 16 below.]
2126    8.  Changed all #include <config.h> to #include "config.h". There were also
2127        some further <pcre.h> cases that I changed to "pcre.h".
2129    9.  When pcregrep was used with the --colour option, it missed the line ending
2130        sequence off the lines that it output.
2132    10. It was pointed out to me that arrays of string pointers cause lots of
2133        relocations when a shared library is dynamically loaded. A technique of
2134        using a single long string with a table of offsets can drastically reduce
2135        these. I have refactored PCRE in four places to do this. The result is
2136        dramatic:
2138          Originally:                          290
2139          After changing UCP table:            187
2140          After changing error message table:   43
2141          After changing table of "verbs"       36
2142          After changing table of Posix names   22
2144        Thanks to the folks working on Gregex for glib for this insight.
2146    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
2147        unicode-properties was also set.
2149    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
2151    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
2152        checked only for CRLF.
2154    14. Added casts to pcretest.c to avoid compiler warnings.
2156    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
2158    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
2159        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2160        entirely. This removes changes made in 7 above.
2162    17. The CMake files have been updated, and there is now more information about
2163        building with CMake in the NON-UNIX-USE document.
2166    Version 7.3 28-Aug-07
2167    ---------------------
2169     1. In the rejigging of the build system that eventually resulted in 7.1, the
2170        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
2171        brackets there is not right, since it causes compilers to look for an
2172        installed pcre.h, not the version that is in the source that is being
2173        compiled (which of course may be different). I have changed it back to:
2175          #include "pcre.h"
2177        I have a vague recollection that the change was concerned with compiling in
2178        different directories, but in the new build system, that is taken care of
2179        by the VPATH setting the Makefile.
2181     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
2182        when the subject happened to end in the byte 0x85 (e.g. if the last
2183        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
2184        characters but of course it shouldn't be taken as a newline when it is part
2185        of another character. The bug was that, for an unlimited repeat of . in
2186        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
2187        characters when looking for a newline.
2189     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
2191     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
2192        in debug output.
2194     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
2195        long printing in the pcrecpp unittest when running under MinGW.
2197     6. ESC_K was left out of the EBCDIC table.
2199     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
2200        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
2201        limit also applies to "virtual nesting" when a pattern is recursive, and in
2202        this case 1000 isn't so big. I have been able to remove this limit at the
2203        expense of backing off one optimization in certain circumstances. Normally,
2204        when pcre_exec() would call its internal match() function recursively and
2205        immediately return the result unconditionally, it uses a "tail recursion"
2206        feature to save stack. However, when a subpattern that can match an empty
2207        string has an unlimited repetition quantifier, it no longer makes this
2208        optimization. That gives it a stack frame in which to save the data for
2209        checking that an empty string has been matched. Previously this was taken
2210        from the 1000-entry workspace that had been reserved. So now there is no
2211        explicit limit, but more stack is used.
2213     8. Applied Daniel's patches to solve problems with the import/export magic
2214        syntax that is required for Windows, and which was going wrong for the
2215        pcreposix and pcrecpp parts of the library. These were overlooked when this
2216        problem was solved for the main library.
2218     9. There were some crude static tests to avoid integer overflow when computing
2219        the size of patterns that contain repeated groups with explicit upper
2220        limits. As the maximum quantifier is 65535, the maximum group length was
2221        set at 30,000 so that the product of these two numbers did not overflow a
2222        32-bit integer. However, it turns out that people want to use groups that
2223        are longer than 30,000 bytes (though not repeat them that many times).
2224        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
2225        made it possible to implement the integer overflow checks in a much more
2226        dynamic way, which I have now done. The artificial limitation on group
2227        length has been removed - we now have only the limit on the total length of
2228        the compiled pattern, which depends on the LINK_SIZE setting.
2230    10. Fixed a bug in the documentation for get/copy named substring when
2231        duplicate names are permitted. If none of the named substrings are set, the
2232        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
2233        empty string.
2235    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
2236        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
2237        because the ] is interpreted as the first data character and the
2238        terminating ] is not found. PCRE has been made compatible with Perl in this
2239        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
2240        cause memory overwriting.
2242    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
2243        string has been matched (to stop an infinite loop). It was not recognizing
2244        a conditional subpattern that could match an empty string if that
2245        subpattern was within another subpattern. For example, it looped when
2246        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
2247        condition was not nested. This bug has been fixed.
2249    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
2250        past the start of the subject in the presence of bytes with the top bit
2251        set, for example "\x8aBCD".
2253    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
2254        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
2256    14. Optimized (?!) to (*FAIL).
2258    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
2259        This restricts code points to be within the range 0 to 0x10FFFF, excluding
2260        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
2261        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
2262        does: it's just the validity check that is more restrictive.
2264    16. Inserted checks for integer overflows during escape sequence (backslash)
2265        processing, and also fixed erroneous offset values for syntax errors during
2266        backslash processing.
2268    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
2269        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
2271    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
2272        caused an overrun.
2274    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
2275        something other than just ASCII characters) inside a group that had an
2276        unlimited repeat caused a loop at compile time (while checking to see
2277        whether the group could match an empty string).
2279    20. Debugging a pattern containing \p or \P could cause a crash. For example,
2280        [\P{Any}] did so. (Error in the code for printing property names.)
2282    21. An orphan \E inside a character class could cause a crash.
2284    22. A repeated capturing bracket such as (A)? could cause a wild memory
2285        reference during compilation.
2287    23. There are several functions in pcre_compile() that scan along a compiled
2288        expression for various reasons (e.g. to see if it's fixed length for look
2289        behind). There were bugs in these functions when a repeated \p or \P was
2290        present in the pattern. These operators have additional parameters compared
2291        with \d, etc, and these were not being taken into account when moving along
2292        the compiled data. Specifically:
2294        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2295            length.
2297        (b) An item such as \pL+ within a repeated group could cause crashes or
2298            loops.
2300        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2301            "reference to non-existent subpattern" error.
2303        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2305    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2306        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2308    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2310    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2311        character were causing crashes (broken optimization).
2313    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2314        \p or \P) caused a compile-time loop.
2316    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2317        break. For example, the unstudied pattern [\r\n]A does not match the string
2318        "\r\nA" because change 7.0/46 below moves the current point on by two
2319        characters after failing to match at the start. However, the pattern \nA
2320        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2321        the same is true. There doesn't seem any very clean way out of this, but
2322        what I have chosen to do makes the common cases work: PCRE now takes note
2323        of whether there can be an explicit match for \r or \n anywhere in the
2324        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2325        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2326        pattern has explicit CR or LF references.
2328    29. Added (*CR) etc for changing newline setting at start of pattern.
2331    Version 7.2 19-Jun-07
2332  ---------------------  ---------------------
2334   1. If the fr_FR locale cannot be found for test 3, try the "french" locale,   1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
# Line 21  Version 7.2 01-May-07 Line 2348  Version 7.2 01-May-07
2348      stack recursion. This gives a massive performance boost under BSD, but just      stack recursion. This gives a massive performance boost under BSD, but just
2349      a small improvement under Linux. However, it saves one field in the frame      a small improvement under Linux. However, it saves one field in the frame
2350      in all cases.      in all cases.
2352   6. Added more features from the forthcoming Perl 5.10:   6. Added more features from the forthcoming Perl 5.10:
2354      (a) (?-n) (where n is a string of digits) is a relative subroutine or      (a) (?-n) (where n is a string of digits) is a relative subroutine or
2355          recursion call. It refers to the nth most recently opened parentheses.          recursion call. It refers to the nth most recently opened parentheses.
2357      (b) (?+n) is also a relative subroutine call; it refers to the nth next      (b) (?+n) is also a relative subroutine call; it refers to the nth next
2358          to be opened parentheses.          to be opened parentheses.
2360      (c) Conditions that refer to capturing parentheses can be specified      (c) Conditions that refer to capturing parentheses can be specified
2361          relatively, for example, (?(-2)... or (?(+3)...          relatively, for example, (?(-2)... or (?(+3)...
2363        (d) \K resets the start of the current match so that everything before
2364            is not part of it.
2366        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2368        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2369            reference syntax.
2371        (g) (?| introduces a group in which the numbering of parentheses in each
2372            alternative starts with the same number.
2374        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2376     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2379     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2380        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2381        for detecting groups that can match an empty string.
2383     9. A pattern with a very large number of alternatives (more than several
2384        hundred) was running out of internal workspace during the pre-compile
2385        phase, where pcre_compile() figures out how much memory will be needed. A
2386        bit of new cunning has reduced the workspace needed for groups with
2387        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2388        workspace instead of running out of the 4096 that are available.
2390    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2392    11. Applied patch from Google to remove an optimization that didn't quite work.
2393        The report of the bug said:
2395          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2396          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2397          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2399    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2400        it matched the wrong number of bytes.
2403  Version 7.1 24-Apr-07  Version 7.1 24-Apr-07
2404  ---------------------  ---------------------

Removed from v.167  
changed lines
  Added in v.1298

  ViewVC Help
Powered by ViewVC 1.1.5