ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 91 by nigel, Sat Feb 24 21:41:34 2007 UTC revision 1319 by ph10, Wed May 1 16:03:57 2013 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.33 28-April-2013
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. (*SKIP) is now supported by the JIT compiler.
109    29. (*THEN) is now supported by the JIT compiler.
111    30. Update RunTest with additional test selector options.
113    31. The way PCRE handles backtracking verbs has been changed in two 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 for positive assertions, except for (*ACCEPT). Again,
122        this sometimes improves Perl compatibility, and sometimes does not.
124    32. A number of tests that were in test 2 because Perl did things differently
125        have been moved to test 1, because either Perl or PCRE has changed, and
126        these tests are now compatible.
128    32. Control verbs are handled in the same way in JIT and interpreter.
130    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
131        contained a forward subroutine reference caused a compile error.
133    34. Auto-detect and optimize limited repetitions in JIT.
135    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
136        blocking (*UTF) etc.
138    36. In the interpreter, maximizing pattern repetitions for characters and
139        character types now use tail recursion, which reduces stack usage.
141    37. The value of the max lookbehind was not correctly preserved if a compiled
142        and saved regex was reloaded on a host of different endianness.
144    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
145        of the compiled pattern block, expand the flags field from 16 to 32 bits
146        because it was almost full.
149    Version 8.32 30-November-2012
150    -----------------------------
152    1.  Improved JIT compiler optimizations for first character search and single
153        character iterators.
155    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
156        Patch by Daniel Richard G.
158    3.  Single character iterator optimizations in the JIT compiler.
160    4.  Improved JIT compiler optimizations for character ranges.
162    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
163        Reported by Giuseppe D'Angelo.
165    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
166        of a line, was being set incorrectly in cases where .* appeared inside
167        atomic brackets at the start of a pattern, or where there was a subsequent
168        *PRUNE or *SKIP.
170    7.  Improved instruction cache flush for POWER/PowerPC.
171        Patch by Daniel Richard G.
173    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
174        grep:
176        (a) There is now no limit to the number of patterns to be matched.
178        (b) An error is given if a pattern is too long.
180        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
181            are now supported.
183        (d) --exclude-from and --include-from (multiple use) have been added.
185        (e) Exclusions and inclusions now apply to all files and directories, not
186            just to those obtained from scanning a directory recursively.
188        (f) Multiple uses of -f and --file-list are now supported.
190        (g) In a Windows environment, the default for -d has been changed from
191            "read" (the GNU grep default) to "skip", because otherwise the presence
192            of a directory in the file list provokes an error.
194        (h) The documentation has been revised and clarified in places.
196    9.  Improve the matching speed of capturing brackets.
198    10. Changed the meaning of \X so that it now matches a Unicode extended
199        grapheme cluster.
201    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
202        out POSIX threads when JIT support is configured.
204    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
206    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
207        structure could go wrong in environments where size_t is not the same size
208        as int.
210    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
211        set.
213    15. The EBCDIC support had decayed; later updates to the code had included
214        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
215        general tidy up of EBCDIC-related issues, and the documentation was also
216        not quite right. There is now a test that can be run on ASCII systems to
217        check some of the EBCDIC-related things (but is it not a full test).
219    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
220        in a small tidy to the code.
222    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
224    18. If the --only-matching (-o) option in pcregrep is specified multiple
225        times, each one causes appropriate output. For example, -o1 -o2 outputs the
226        substrings matched by the 1st and 2nd capturing parentheses. A separating
227        string can be specified by --om-separator (default empty).
229    19. Improving the first n character searches.
231    20. Turn case lists for horizontal and vertical white space into macros so that
232        they are defined only once.
234    21. This set of changes together give more compatible Unicode case-folding
235        behaviour for characters that have more than one other case when UCP
236        support is available.
238        (a) The Unicode property table now has offsets into a new table of sets of
239            three or more characters that are case-equivalent. The MultiStage2.py
240            script that generates these tables (the pcre_ucd.c file) now scans
241            CaseFolding.txt instead of UnicodeData.txt for character case
242            information.
244        (b) The code for adding characters or ranges of characters to a character
245            class has been abstracted into a generalized function that also handles
246            case-independence. In UTF-mode with UCP support, this uses the new data
247            to handle characters with more than one other case.
249        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
250            whose other case is greater than 256 are now correctly matched
251            caselessly. Previously, the high codepoint matched the low one, but not
252            vice versa.
254        (d) The processing of \h, \H, \v, and \ in character classes now makes use
255            of the new class addition function, using character lists defined as
256            macros alongside the case definitions of 20 above.
258        (e) Caseless back references now work with characters that have more than
259            one other case.
261        (f) General caseless matching of characters with more than one other case
262            is supported.
264    22. Unicode character properties were updated from Unicode 6.2.0
266    23. Improved CMake support under Windows. Patch by Daniel Richard G.
268    24. Add support for 32-bit character strings, and UTF-32
270    25. Major JIT compiler update (code refactoring and bugfixing).
271        Experimental Sparc 32 support is added.
273    26. Applied a modified version of Daniel Richard G's patch to create
274        pcre.h.generic and config.h.generic by "make" instead of in the
275        PrepareRelease script.
277    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
278        pcre_compile.c when checking for a zero character.
280    28. Introducing a native interface for JIT. Through this interface, the compiled
281        machine code can be directly executed. The purpose of this interface is to
282        provide fast pattern matching, so several sanity checks are not performed.
283        However, feature tests are still performed. The new interface provides
284        1.4x speedup compared to the old one.
286    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
287        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
288        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
290    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
291        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
292        the "old" RFC 2279). Instead, it ended up passing a negative length to
293        pcre_exec().
295    31. Add support for GCC's visibility feature to hide internal functions.
297    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
298        "unknown -C option" after outputting 0 or 1.
300    33. There is now support for generating a code coverage report for the test
301        suite in environments where gcc is the compiler and lcov is installed. This
302        is mainly for the benefit of the developers.
304    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
305        unaddressable using valgrind annotations, allowing valgrind to detect
306        invalid memory accesses. This is mainly for the benefit of the developers.
308    25. (*UTF) can now be used to start a pattern in any of the three libraries.
310    26. Give configure error if --enable-cpp but no C++ compiler found.
313    Version 8.31 06-July-2012
314    -------------------------
316    1.  Fixing a wrong JIT test case and some compiler warnings.
318    2.  Removed a bashism from the RunTest script.
320    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
321        to unsigned type, result still unsigned" that was given by an MS compiler
322        on encountering the code "-sizeof(xxx)".
324    4.  Partial matching support is added to the JIT compiler.
326    5.  Fixed several bugs concerned with partial matching of items that consist
327        of more than one character:
329        (a) /^(..)\1/ did not partially match "aba" because checking references was
330            done on an "all or nothing" basis. This also applied to repeated
331            references.
333        (b) \R did not give a hard partial match if \r was found at the end of the
334            subject.
336        (c) \X did not give a hard partial match after matching one or more
337            characters at the end of the subject.
339        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
340            a partial match for the string "\r".
342        (e) When newline was set to CRLF, the metacharacter "." did not recognize
343            a partial match for a CR character at the end of the subject string.
345    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
346        running pcretest, the text "(JIT)" added to the output whenever JIT is
347        actually used to run the match.
349    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
350        or /S+[+] with a digit between 1 and 7.
352    8.  OP_NOT now supports any UTF character not just single-byte ones.
354    9.  (*MARK) control verb is now supported by the JIT compiler.
356    10. The command "./RunTest list" lists the available tests without actually
357        running any of them. (Because I keep forgetting what they all are.)
361    12. Applied a (slightly modified) user-supplied patch that improves performance
362        when the heap is used for recursion (compiled with --disable-stack-for-
363        recursion). Instead of malloc and free for each heap frame each time a
364        logical recursion happens, frames are retained on a chain and re-used where
365        possible. This sometimes gives as much as 30% improvement.
367    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
368        call.
370    14. As documented, (*COMMIT) is now confined to within a positive assertion.
372    15. It is now possible to link pcretest with libedit as an alternative to
373        libreadline.
375    16. (*COMMIT) control verb is now supported by the JIT compiler.
377    17. The Unicode data tables have been updated to Unicode 6.1.0.
379    18. Added --file-list option to pcregrep.
381    19. Added binary file support to pcregrep, including the -a, --binary-files,
382        -I, and --text options.
384    20. The madvise function is renamed for posix_madvise for QNX compatibility
385        reasons. Fixed by Giuseppe D'Angelo.
387    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
389    22. Changed the option for creating symbolic links for 16-bit man pages from
390        -s to -sf so that re-installing does not cause issues.
392    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
394    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
395        matches in certain environments (the workspace was not being correctly
396        retained). Also added to pcre_dfa_exec() a simple plausibility check on
397        some of the workspace data at the beginning of a restart.
399    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
400        was not doing so when it should - probably a typo introduced by SVN 528
401        (change 8.10/14).
403    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
404        \w+ when the character tables indicated that \x{c4} was a word character.
405        There were several related cases, all because the tests for doing a table
406        lookup were testing for characters less than 127 instead of 255.
408    27. If a pattern contains capturing parentheses that are not used in a match,
409        their slots in the ovector are set to -1. For those that are higher than
410        any matched groups, this happens at the end of processing. In the case when
411        there were back references that the ovector was too small to contain
412        (causing temporary malloc'd memory to be used during matching), and the
413        highest capturing number was not used, memory off the end of the ovector
414        was incorrectly being set to -1. (It was using the size of the temporary
415        memory instead of the true size.)
417    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
418        ovector size, it uses memory at the end of the block that it has got.
420    29. Check for an overlong MARK name and give an error at compile time. The
421        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
423    30. JIT compiler update.
425    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
426        Rill for the patch.
428    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
430    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
432    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
433        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
434        was enabled.
436    35. Improve JIT code generation for greedy plus quantifier.
438    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
439        1 to "aa" instead of to an empty string. The bug affected repeated groups
440        that could potentially match an empty string.
442    37. Optimizing single character iterators in JIT.
444    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
445        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
446        codepoints that are too big for the mode are faulted, and in a UTF mode,
447        disallowed codepoints are also faulted.
449    39. If PCRE was compiled with UTF support, in three places in the DFA
450        matcher there was code that should only have been obeyed in UTF mode, but
451        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
452        processing when bytes with values greater than 127 were present. In 16-bit
453        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
454        both cases the values are those that cannot be the first data item in a UTF
455        character. The three items that might have provoked this were recursions,
456        possessively repeated groups, and atomic groups.
458    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
459        and pcregrep, because some OS require shared objects to be explicitly
460        passed to ld, causing the link step to fail if they are not.
462    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
463        mode, patterns that started with \h* or \R* might be incorrectly matched.
466    Version 8.30 04-February-2012
467    -----------------------------
469    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
470        name is defined in ctype.h.
472    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
473        only in quite long subpatterns.
475    3.  Removed the function pcre_info(), which has been obsolete and deprecated
476        since it was replaced by pcre_fullinfo() in February 2000.
478    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
479        match a (*MARK), and the match failed at the start of the subject, a
480        reference to memory before the start of the subject could occur. This bug
481        was introduced by fix 17 of release 8.21.
483    5.  A reference to an unset group with zero minimum repetition was giving
484        totally wrong answers (in non-JavaScript-compatibility mode). For example,
485        /(another)?(\1?)test/ matched against "hello world test". This bug was
486        introduced in release 8.13.
488    6.  Add support for 16-bit character strings (a large amount of work involving
489        many changes and refactorings).
491    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
492        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
493        from a file.
495    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
496        rounding is not applied in this particular case).
498    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
499        if they appear, or are escaped, in patterns.
501    10. Get rid of a number of -Wunused-but-set-variable warnings.
503    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
504        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
505        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
506        also returns the mark "x". This bug applied to capturing parentheses,
507        non-capturing parentheses, and atomic parentheses. It also applied to some
508        assertions.
510    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
511        information out of configure.ac instead of relying on pcre.h.generic, which
512        is not stored in the repository.
514    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
515        -lreadline.
517    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
519    15. Applied Graycode's patch to put the top-level frame on the stack rather
520        than the heap when not using the stack for recursion. This gives a
521        performance improvement in many cases when recursion is not deep.
523    16. Experimental code added to "pcretest -C" to output the stack frame size.
526    Version 8.21 12-Dec-2011
527    ------------------------
529    1.  Updating the JIT compiler.
531    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
532        are added as well.
534    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
535        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
536        calling _pcre_jit_exec. Some extra comments are added.
538    4.  (*MARK) settings inside atomic groups that do not contain any capturing
539        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
540        was introduced by change 18 for 8.20.
542    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
543        ECMA-262 standard.
545    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
546        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
547        This bug was probably introduced by change 9 of 8.13.
549    7.  While fixing 6 above, I noticed that a number of other items were being
550        incorrectly rejected as "not fixed length". This arose partly because newer
551        opcodes had not been added to the fixed-length checking code. I have (a)
552        corrected the bug and added tests for these items, and (b) arranged for an
553        error to occur if an unknown opcode is encountered while checking for fixed
554        length instead of just assuming "not fixed length". The items that were
555        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
556        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
557        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
559    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
560        being incorrectly compiled and would have given unpredicatble results.
562    9.  A possessively repeated subpattern with minimum repeat count greater than
563        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
564        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
565        the first (A) could occur when it should not.
567    10. Add a cast and remove a redundant test from the code.
569    11. JIT should use pcre_malloc/pcre_free for allocation.
571    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
572        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
573        is anything other than /usr, -L is still shown).
575    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
577    14. Perl does not support \N without a following name in a [] class; PCRE now
578        also gives an error.
580    15. If a forward reference was repeated with an upper limit of around 2000,
581        it caused the error "internal error: overran compiling workspace". The
582        maximum number of forward references (including repeats) was limited by the
583        internal workspace, and dependent on the LINK_SIZE. The code has been
584        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
585        the default depends on LINK_SIZE. There is a new upper limit (for safety)
586        of around 200,000 forward references. While doing this, I also speeded up
587        the filling in of repeated forward references.
589    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
590        incorrectly expecting the subject to contain another "a" after the start.
592    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
593        in the match, the SKIP should be ignored. This was not happening; instead
594        the SKIP was being treated as NOMATCH. For patterns such as
595        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
596        tested.
598    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
599        now much more compatible with Perl, in particular in cases where the result
600        is a non-match for a non-anchored pattern. For example, if
601        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
602        "m", where previously it did not return a name. A side effect of this
603        change is that for partial matches, the last encountered mark name is
604        returned, as for non matches. A number of tests that were previously not
605        Perl-compatible have been moved into the Perl-compatible test files. The
606        refactoring has had the pleasing side effect of removing one argument from
607        the match() function, thus reducing its stack requirements.
609    19. If the /S+ option was used in pcretest to study a pattern using JIT,
610        subsequent uses of /S (without +) incorrectly behaved like /S+.
612    21. Retrieve executable code size support for the JIT compiler and fixing
613        some warnings.
615    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
616        not work when the shorter character appeared right at the end of the
617        subject string.
619    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
620        systems.
622    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
623        output it when the /M option is used in pcretest.
625    25. The CheckMan script was not being included in the distribution. Also, added
626        an explicit "perl" to run Perl scripts from the PrepareRelease script
627        because this is reportedly needed in Windows.
629    26. If study data was being save in a file and studying had not found a set of
630        "starts with" bytes for the pattern, the data written to the file (though
631        never used) was taken from uninitialized memory and so caused valgrind to
632        complain.
634    27. Updated RunTest.bat as provided by Sheri Pierce.
636    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
638    29. Computation of memory usage for the table of capturing group names was
639        giving an unnecessarily large value.
642    Version 8.20 21-Oct-2011
643    ------------------------
645    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
646        a POSIX class. After further experiments with Perl, which convinced me that
647        Perl has bugs and confusions, a closing square bracket is no longer allowed
648        in a POSIX name. This bug also affected patterns with classes that started
649        with full stops.
651    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
652        captured substring, but while checking the failing first alternative,
653        substring 1 is temporarily captured. If the output vector supplied to
654        pcre_exec() was not big enough for this capture, the yield of the function
655        was still zero ("insufficient space for captured substrings"). This cannot
656        be totally fixed without adding another stack variable, which seems a lot
657        of expense for a edge case. However, I have improved the situation in cases
658        such as /(a)(b)x|abc/ matched against "abc", where the return code
659        indicates that fewer than the maximum number of slots in the ovector have
660        been set.
662    3.  Related to (2) above: when there are more back references in a pattern than
663        slots in the output vector, pcre_exec() uses temporary memory during
664        matching, and copies in the captures as far as possible afterwards. It was
665        using the entire output vector, but this conflicts with the specification
666        that only 2/3 is used for passing back captured substrings. Now it uses
667        only the first 2/3, for compatibility. This is, of course, another edge
668        case.
670    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
671        main code base, and can be used by building with --enable-jit. When this is
672        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
673        runtime --no-jit option is given.
675    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
676        ovector, the return from the function was zero, implying that there were
677        other matches that did not fit. The correct "exactly full" value is now
678        returned.
680    6.  If a subpattern that was called recursively or as a subroutine contained
681        (*PRUNE) or any other control that caused it to give a non-standard return,
682        invalid errors such as "Error -26 (nested recursion at the same subject
683        position)" or even infinite loops could occur.
685    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
686        computing the minimum length on reaching *ACCEPT, and so ended up with the
687        wrong value of 1 rather than 0. Further investigation indicates that
688        computing a minimum subject length in the presence of *ACCEPT is difficult
689        (think back references, subroutine calls), and so I have changed the code
690        so that no minimum is registered for a pattern that contains *ACCEPT.
692    8.  If (*THEN) was present in the first (true) branch of a conditional group,
693        it was not handled as intended. [But see 16 below.]
695    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
696        Sheri Pierce.
698    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
699        the first byte in a match must be "a".
701    11. Change 17 for 8.13 increased the recursion depth for patterns like
702        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
703        pattern contains any instances of (*THEN). If it does not, the old
704        optimizations are restored. It would be nice to do this on a per-group
705        basis, but at the moment that is not feasible.
707    12. In some environments, the output of pcretest -C is CRLF terminated. This
708        broke RunTest's code that checks for the link size. A single white space
709        character after the value is now allowed for.
711    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
712        For "fr", it uses the Windows-specific input and output files.
714    14. If (*THEN) appeared in a group that was called recursively or as a
715        subroutine, it did not work as intended. [But see next item.]
717    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
718        pattern fragments (but not containing any | characters). If A and B are
719        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
720        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
721        D. In other words, Perl considers parentheses that do not contain any |
722        characters to be part of a surrounding alternative, whereas PCRE was
723        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
724        differently. PCRE now behaves in the same way as Perl, except in the case
725        of subroutine/recursion calls such as (?1) which have in any case always
726        been different (but PCRE had them first :-).
728    16. Related to 15 above: Perl does not treat the | in a conditional group as
729        creating alternatives. Such a group is treated in the same way as an
730        ordinary group without any | characters when processing (*THEN). PCRE has
731        been changed to match Perl's behaviour.
733    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
734        RunGrepTest script failed.
736    18. Change 22 for version 13 caused atomic groups to use more stack. This is
737        inevitable for groups that contain captures, but it can lead to a lot of
738        stack use in large patterns. The old behaviour has been restored for atomic
739        groups that do not contain any capturing parentheses.
741    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
742        suppress the check for a minimum subject length at run time. (If it was
743        given to pcre_exec() or pcre_dfa_exec() it did work.)
745    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
746        fail to work when decoding hex characters in data strings in EBCDIC
747        environments.
749    21. It appears that in at least one Mac OS environment, the isxdigit() function
750        is implemented as a macro that evaluates to its argument more than once,
751        contravening the C 90 Standard (I haven't checked a later standard). There
752        was an instance in pcretest which caused it to go wrong when processing
753        \x{...} escapes in subject strings. The has been rewritten to avoid using
754        things like p++ in the argument of isxdigit().
757    Version 8.13 16-Aug-2011
758    ------------------------
760    1.  The Unicode data tables have been updated to Unicode 6.0.0.
762    2.  Two minor typos in pcre_internal.h have been fixed.
764    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
765        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
766        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
768    4.  There were a number of related bugs in the code for matching backrefences
769        caselessly in UTF-8 mode when codes for the characters concerned were
770        different numbers of bytes. For example, U+023A and U+2C65 are an upper
771        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
772        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
773        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
774        2-byte code at the end of the subject (it thought there wasn't enough data
775        left).
777    5.  Comprehensive information about what went wrong is now returned by
778        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
779        as the output vector has at least 2 elements. The offset of the start of
780        the failing character and a reason code are placed in the vector.
782    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
783        now returned is for the first byte of the failing character, instead of the
784        last byte inspected. This is an incompatible change, but I hope it is small
785        enough not to be a problem. It makes the returned offset consistent with
786        pcre_exec() and pcre_dfa_exec().
788    7.  pcretest now gives a text phrase as well as the error number when
789        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
790        failure, the offset and reason code are output.
792    8.  When \R was used with a maximizing quantifier it failed to skip backwards
793        over a \r\n pair if the subsequent match failed. Instead, it just skipped
794        back over a single character (\n). This seems wrong (because it treated the
795        two characters as a single entity when going forwards), conflicts with the
796        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
797        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
798        has been changed.
800    9.  Some internal refactoring has changed the processing so that the handling
801        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
802        time (the PCRE_DOTALL option was changed this way some time ago: version
803        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
804        which was always a bit of a fudge. It also means that there is one less
805        argument for the match() function, which reduces its stack requirements
806        slightly. This change also fixes an incompatibility with Perl: the pattern
807        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
809    10. More internal refactoring has drastically reduced the number of recursive
810        calls to match() for possessively repeated groups such as (abc)++ when
811        using pcre_exec().
813    11. While implementing 10, a number of bugs in the handling of groups were
814        discovered and fixed:
816        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
817        (a|)*(?1) gave a compile-time internal error.
818        ((a|)+)+  did not notice that the outer group could match an empty string.
819        (^a|^)+   was not marked as anchored.
820        (.*a|.*)+ was not marked as matching at start or after a newline.
822    12. Yet more internal refactoring has removed another argument from the match()
823        function. Special calls to this function are now indicated by setting a
824        value in a variable in the "match data" data block.
826    13. Be more explicit in pcre_study() instead of relying on "default" for
827        opcodes that mean there is no starting character; this means that when new
828        ones are added and accidentally left out of pcre_study(), testing should
829        pick them up.
831    14. The -s option of pcretest has been documented for ages as being an old
832        synonym of -m (show memory usage). I have changed it to mean "force study
833        for every regex", that is, assume /S for every regex. This is similar to -i
834        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
835        using it. It makes it easier to run collections of tests with and without
836        study enabled, and thereby test pcre_study() more easily. All the standard
837        tests are now run with and without -s (but some patterns can be marked as
838        "never study" - see 20 below).
840    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
841        restoration of the capturing data to the outer values was not happening
842        correctly.
844    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
845        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
846        pattern had matched an empty string, and so incorrectly returned a no
847        match.
849    17. There was optimizing code for the last branch of non-capturing parentheses,
850        and also for the obeyed branch of a conditional subexpression, which used
851        tail recursion to cut down on stack usage. Unfortunately, now that there is
852        the possibility of (*THEN) occurring in these branches, tail recursion is
853        no longer possible because the return has to be checked for (*THEN). These
854        two optimizations have therefore been removed. [But see 8.20/11 above.]
856    18. If a pattern containing \R was studied, it was assumed that \R always
857        matched two bytes, thus causing the minimum subject length to be
858        incorrectly computed because \R can also match just one byte.
860    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
861        was incorrectly computed.
863    20. If /S is present twice on a test pattern in pcretest input, it now
864        *disables* studying, thereby overriding the use of -s on the command line
865        (see 14 above). This is necessary for one or two tests to keep the output
866        identical in both cases.
868    21. When (*ACCEPT) was used in an assertion that matched an empty string and
869        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
871    22. When an atomic group that contained a capturing parenthesis was
872        successfully matched, but the branch in which it appeared failed, the
873        capturing was not being forgotten if a higher numbered group was later
874        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
875        group 1 to "a", when in fact it should be unset. This applied to multi-
876        branched capturing and non-capturing groups, repeated or not, and also to
877        positive assertions (capturing in negative assertions does not happen
878        in PCRE) and also to nested atomic groups.
880    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
881        subject after a captured substring, to make it easier to tell which of a
882        number of identical substrings has been captured.
884    24. The way atomic groups are processed by pcre_exec() has been changed so that
885        if they are repeated, backtracking one repetition now resets captured
886        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
887        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
888        "aaa". Previously, it would have been "a". As part of this code
889        refactoring, the way recursive calls are handled has also been changed.
891    25. If an assertion condition captured any substrings, they were not passed
892        back unless some other capturing happened later. For example, if
893        (?(?=(a))a) was matched against "a", no capturing was returned.
895    26. When studying a pattern that contained subroutine calls or assertions,
896        the code for finding the minimum length of a possible match was handling
897        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
898        group 1 called group 2 while simultaneously a separate group 2 called group
899        1). A stack overflow occurred in this case. I have fixed this by limiting
900        the recursion depth to 10.
902    27. Updated RunTest.bat in the distribution to the version supplied by Tom
903        Fortmann. This supports explicit test numbers on the command line, and has
904        argument validation and error reporting.
906    28. An instance of \X with an unlimited repeat could fail if at any point the
907        first character it looked at was a mark character.
909    29. Some minor code refactoring concerning Unicode properties and scripts
910        should reduce the stack requirement of match() slightly.
912    30. Added the '=' option to pcretest to check the setting of unused capturing
913        slots at the end of the pattern, which are documented as being -1, but are
914        not included in the return count.
916    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
917        compiled something random. Now it gives a compile-time error (as does
918        Perl).
920    32. A *MARK encountered during the processing of a positive assertion is now
921        recorded and passed back (compatible with Perl).
923    33. If --only-matching or --colour was set on a pcregrep call whose pattern
924        had alternative anchored branches, the search for a second match in a line
925        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
926        matched the line "0102" twice. The same bug affected patterns that started
927        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
928        twice.
930    34. Previously, PCRE did not allow quantification of assertions. However, Perl
931        does, and because of capturing effects, quantifying parenthesized
932        assertions may at times be useful. Quantifiers are now allowed for
933        parenthesized assertions.
935    35. A minor code tidy in pcre_compile() when checking options for \R usage.
937    36. \g was being checked for fancy things in a character class, when it should
938        just be a literal "g".
940    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
941        appearance of a nested POSIX class supersedes an apparent external class.
942        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
943        unescaped square brackets may also appear as part of class names. For
944        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
945        more like Perl. (But see 8.20/1 above.)
947    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
948        was because it thought it was \N{name}, which is not supported).
950    39. Add minix to OS list not supporting the -S option in pcretest.
952    40. PCRE tries to detect cases of infinite recursion at compile time, but it
953        cannot analyze patterns in sufficient detail to catch mutual recursions
954        such as ((?1))((?2)). There is now a runtime test that gives an error if a
955        subgroup is called recursively as a subpattern for a second time at the
956        same position in the subject string. In previous releases this might have
957        been caught by the recursion limit, or it might have run out of stack.
959    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
960        happen only once. PCRE was, however incorrectly giving a compile time error
961        "recursive call could loop indefinitely" because it cannot analyze the
962        pattern in sufficient detail. The compile time test no longer happens when
963        PCRE is compiling a conditional subpattern, but actual runaway loops are
964        now caught at runtime (see 40 above).
966    42. It seems that Perl allows any characters other than a closing parenthesis
967        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
968        has been changed to be the same.
970    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
971        as not to get warnings when autogen.sh is called. Also changed
972        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
974    44. To help people who use pcregrep to scan files containing exceedingly long
975        lines, the following changes have been made:
977        (a) The default value of the buffer size parameter has been increased from
978            8K to 20K. (The actual buffer used is three times this size.)
980        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
981            PCRE is built.
983        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
984            to be set at run time.
986        (d) Numerical values in pcregrep options can be followed by K or M, for
987            example --buffer-size=50K.
989        (e) If a line being scanned overflows pcregrep's buffer, an error is now
990            given and the return code is set to 2.
992    45. Add a pointer to the latest mark to the callout data block.
994    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
995        partial match of an empty string instead of no match. This was specific to
996        the use of ".".
998    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
999        complete match instead of a partial match. This bug was dependent on both
1000        the PCRE_UTF8 and PCRE_DOTALL options being set.
1002    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1003        starting byte set, because \b was not being ignored.
1006    Version 8.12 15-Jan-2011
1007    ------------------------
1009    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1010        checks for such things as part of the documentation building process.
1012    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1013        --match-limit and --recursion-limit options (added for 8.11). In
1014        particular, this made one of the standard tests fail. (The integer value
1015        went into the wrong half of a long int.)
1017    3.  If the --colour option was given to pcregrep with -v (invert match), it
1018        did strange things, either producing crazy output, or crashing. It should,
1019        of course, ignore a request for colour when reporting lines that do not
1020        match.
1022    4.  Another pcregrep bug caused similar problems if --colour was specified with
1023        -M (multiline) and the pattern match finished with a line ending.
1025    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1026        matched in multiline mode, the following line was shown as part of the
1027        match. This seems wrong, so I have changed it.
1029    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1030        the check for further matches in the same line (so they could be coloured)
1031        to overrun the end of the current line. If another match was found, it was
1032        incorrectly shown (and then shown again when found in the next line).
1034    7.  If pcregrep was compiled under Windows, there was a reference to the
1035        function pcregrep_exit() before it was defined. I am assuming this was
1036        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1037        reported by a user. I've moved the definition above the reference.
1040    Version 8.11 10-Dec-2010
1041    ------------------------
1043    1.  (*THEN) was not working properly if there were untried alternatives prior
1044        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1045        backtracked to try for "b" instead of moving to the next alternative branch
1046        at the same level (in this case, to look for "c"). The Perl documentation
1047        is clear that when (*THEN) is backtracked onto, it goes to the "next
1048        alternative in the innermost enclosing group".
1050    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1051        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1052        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1053        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1054        (*THEN).
1056    3.  If \s appeared in a character class, it removed the VT character from
1057        the class, even if it had been included by some previous item, for example
1058        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1059        of \s, but is part of the POSIX "space" class.)
1061    4.  A partial match never returns an empty string (because you can always
1062        match an empty string at the end of the subject); however the checking for
1063        an empty string was starting at the "start of match" point. This has been
1064        changed to the "earliest inspected character" point, because the returned
1065        data for a partial match starts at this character. This means that, for
1066        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1067        (previously it gave "no match").
1069    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1070        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1071        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1072        has an implication that the given string is incomplete (because a partial
1073        match is preferred over a full match). For this reason, these items now
1074        give a partial match in this situation. [Aside: previously, the one case
1075        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1076        match rather than a full match, which was wrong by the old rules, but is
1077        now correct.]
1079    6.  There was a bug in the handling of #-introduced comments, recognized when
1080        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1081        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1082        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1083        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1084        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1085        places in pcre_compile().
1087    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1088        comments when looking ahead for named forward references to subpatterns,
1089        the only newline sequence it recognized was NL. It now handles newlines
1090        according to the set newline convention.
1092    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1093        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1094        cater for a lack of strerror(). These oversights have been fixed.
1096    9.  Added --match-limit and --recursion-limit to pcregrep.
1098    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1100    11. When the -o option was used, pcregrep was setting a return code of 1, even
1101        when matches were found, and --line-buffered was not being honoured.
1103    12. Added an optional parentheses number to the -o and --only-matching options
1104        of pcregrep.
1106    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1107        can match an empty string. The code to do it in pcretest and pcredemo
1108        needed fixing:
1110        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1111            only one byte after an empty string match just before CRLF (this case
1112            just got forgotten; "any" and "anycrlf" were OK).
1114        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1115            mode when an empty string match preceded an ASCII character followed by
1116            a non-ASCII character. (The code for advancing by one character rather
1117            than one byte was nonsense.)
1119        (c) The pcredemo.c sample program did not have any code at all to handle
1120            the cases when CRLF is a valid newline sequence.
1122    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1123        as a starting offset was within the subject string. There is now a new
1124        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1125        negative or greater than the length of the string. In order to test this,
1126        pcretest is extended to allow the setting of negative starting offsets.
1128    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1129        starting offset points to the beginning of a UTF-8 character was
1130        unnecessarily clumsy. I tidied it up.
1132    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1133        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1135    17. Nobody had reported that the --include_dir option, which was added in
1136        release 7.7 should have been called --include-dir (hyphen, not underscore)
1137        for compatibility with GNU grep. I have changed it to --include-dir, but
1138        left --include_dir as an undocumented synonym, and the same for
1139        --exclude-dir, though that is not available in GNU grep, at least as of
1140        release 2.5.4.
1142    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1143        characters from a string of bytes) have been redefined so as not to use
1144        loops, in order to improve performance in some environments. At the same
1145        time, I abstracted some of the common code into auxiliary macros to save
1146        repetition (this should not affect the compiled code).
1148    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1149        compile-time error is now given if \c is not followed by an ASCII
1150        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1151        different, and any byte value is allowed.)
1153    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1154        START_OPTIMIZE option, which is now allowed at compile time - but just
1155        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1156        to pcregrep and other applications that have no direct access to PCRE
1157        options. The new /Y option in pcretest sets this option when calling
1158        pcre_compile().
1160    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1161        back references. Groups containing recursive back references were forced to
1162        be atomic by that change, but in the case of named groups, the amount of
1163        memory required was incorrectly computed, leading to "Failed: internal
1164        error: code overflow". This has been fixed.
1166    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1167        pcretest.c, to avoid build problems in some Borland environments.
1170    Version 8.10 25-Jun-2010
1171    ------------------------
1173    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1174        THEN.
1176    2.  (*ACCEPT) was not working when inside an atomic group.
1178    3.  Inside a character class, \B is treated as a literal by default, but
1179        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1180        causes the error). The code is unchanged, but I tidied the documentation.
1182    4.  Inside a character class, PCRE always treated \R and \X as literals,
1183        whereas Perl faults them if its -w option is set. I have changed PCRE so
1184        that it faults them when PCRE_EXTRA is set.
1186    5.  Added support for \N, which always matches any character other than
1187        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1189    6.  When compiling pcregrep with newer versions of gcc which may have
1190        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1191        declared with attribute warn_unused_result" were given. Just casting the
1192        result to (void) does not stop the warnings; a more elaborate fudge is
1193        needed. I've used a macro to implement this.
1195    7.  Minor change to pcretest.c to avoid a compiler warning.
1197    8.  Added four artifical Unicode properties to help with an option to make
1198        \s etc use properties (see next item). The new properties are: Xan
1199        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1201    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1202        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1203        this option. Modified pcretest to add /W to test this facility. Added
1204        REG_UCP to make it available via the POSIX interface.
1206    10. Added --line-buffered to pcregrep.
1208    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1209        studied, and the match started with a letter with a code point greater than
1210        127 whose first byte was different to the first byte of the other case of
1211        the letter, the other case of this starting letter was not recognized
1212        (#976).
1214    12. If a pattern that was studied started with a repeated Unicode property
1215        test, for example, \p{Nd}+, there was the theoretical possibility of
1216        setting up an incorrect bitmap of starting bytes, but fortunately it could
1217        not have actually happened in practice until change 8 above was made (it
1218        added property types that matched character-matching opcodes).
1220    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1221        possible starting bytes for non-anchored patterns.
1223    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1224        \R, and also a number of cases that involve Unicode properties, both
1225        explicit and implicit when PCRE_UCP is set.
1227    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1228        input, it could crash or give wrong results if characters with values
1229        greater than 0xc0 were present in the subject string. (Detail: it assumed
1230        UTF-8 input when processing these items.)
1232    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1233        size_t is 64-bit (#991).
1235    17. Added a check for running out of memory when PCRE is compiled with
1236        --disable-stack-for-recursion (#990).
1238    18. If the last data line in a file for pcretest does not have a newline on
1239        the end, a newline was missing in the output.
1241    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1242        less than 128) in its various bitmaps. However, there is a facility for
1243        generating tables according to the current locale when PCRE is compiled. It
1244        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1245        characters, are recognized by isspace() and therefore were getting set in
1246        these tables, and indeed these tables seem to approximate to ISO 8859. This
1247        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1248        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1249        which of course cannot start UTF-8 characters. I have changed the code so
1250        that only real ASCII characters (less than 128) and the correct starting
1251        bytes for UTF-8 encodings are set for characters greater than 127 when in
1252        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1253        altogether.)
1255    20. Added the /T option to pcretest so as to be able to run tests with non-
1256        standard character tables, thus making it possible to include the tests
1257        used for 19 above in the standard set of tests.
1259    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1260        reference to a subpattern the other side of a comment that contains an
1261        opening parenthesis caused either an internal compiling error, or a
1262        reference to the wrong subpattern.
1265    Version 8.02 19-Mar-2010
1266    ------------------------
1268    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1270    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1271        configured.
1273    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1274        original author of that file, following a query about its status.
1276    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1277        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1279    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1280        quantifier applied to a forward-referencing subroutine call, could compile
1281        incorrect code or give the error "internal error: previously-checked
1282        referenced subpattern not found".
1284    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1285        variables to point to external functions. For these systems, therefore,
1286        pcre_malloc etc. are now initialized to local functions that call the
1287        relevant global functions.
1289    7.  There were two entries missing in the vectors called coptable and poptable
1290        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1291        I've fixed the data, and added a kludgy way of testing at compile time that
1292        the lengths are correct (equal to the number of opcodes).
1294    8.  Following on from 7, I added a similar kludge to check the length of the
1295        eint vector in pcreposix.c.
1297    9.  Error texts for pcre_compile() are held as one long string to avoid too
1298        much relocation at load time. To find a text, the string is searched,
1299        counting zeros. There was no check for running off the end of the string,
1300        which could happen if a new error number was added without updating the
1301        string.
1303    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1305    11. \K was not working if it appeared in an atomic group or in a group that
1306        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1307        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1308        the assertion is positive, but not if it is negative.
1310    12. Change 11 fortuitously reduced the size of the stack frame used in the
1311        "match()" function of pcre_exec.c by one pointer. Forthcoming
1312        implementation of support for (*MARK) will need an extra pointer on the
1313        stack; I have reserved it now, so that the stack frame size does not
1314        decrease.
1316    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1317        item in branch that calls a recursion is a subroutine call - as in the
1318        second branch in the above example - was incorrectly given the compile-
1319        time error "recursive call could loop indefinitely" because pcre_compile()
1320        was not correctly checking the subroutine for matching a non-empty string.
1322    14. The checks for overrunning compiling workspace could trigger after an
1323        overrun had occurred. This is a "should never occur" error, but it can be
1324        triggered by pathological patterns such as hundreds of nested parentheses.
1325        The checks now trigger 100 bytes before the end of the workspace.
1327    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1330    Version 8.01 19-Jan-2010
1331    ------------------------
1333    1.  If a pattern contained a conditional subpattern with only one branch (in
1334        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1335        computed the wrong minimum data length (which is of course zero for such
1336        subpatterns). This could cause incorrect "no match" results.
1338    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1339        the pattern is reset in the first branch, pcre_compile() failed with
1340        "internal error: code overflow at offset...". This happened only when
1341        the reset was to the original external option setting. (An optimization
1342        abstracts leading options settings into an external setting, which was the
1343        cause of this.)
1345    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1346        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1347        assertion pattern did not match (meaning that the assertion was true), it
1348        was incorrectly treated as false if the SKIP had been reached during the
1349        matching. This also applied to assertions used as conditions.
1351    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1352        assertion subpattern, including such a pattern used as a condition,
1353        unpredictable results occurred, instead of the error return
1356    5.  The C++ GlobalReplace function was not working like Perl for the special
1357        situation when an empty string is matched. It now does the fancy magic
1358        stuff that is necessary.
1360    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1361        removed. (These were left over from very, very early versions of PCRE.)
1363    7.  Some cosmetic changes to the code to make life easier when compiling it
1364        as part of something else:
1366        (a) Change DEBUG to PCRE_DEBUG.
1368        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1369            called "current" as "current_branch", to prevent a collision with the
1370            Linux macro when compiled as a kernel module.
1372        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1373            prevent a collision with the Linux macro when compiled as a kernel
1374            module.
1376    8.  In pcre_compile() there are some checks for integer overflows that used to
1377        cast potentially large values to (double). This has been changed to that
1378        when building, a check for int64_t is made, and if it is found, it is used
1379        instead, thus avoiding the use of floating point arithmetic. (There is no
1380        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1381        double.
1383    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1384        2005 (difference between two addresses compared to an unsigned value).
1386    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1387        custom one, because of the following reported problem in Windows:
1389          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1390              under Win32.
1391          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1392              therefore missing the function definition.
1393          - The compiler thus generates a "C" signature for the test function.
1394          - The linker fails to find the "C" function.
1395          - PCRE fails to configure if asked to do so against libbz2.
1397    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1398        messages were output:
1400          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1401          rerunning libtoolize, to keep the correct libtool macros in-tree.
1402          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1404        I have done both of these things.
1406    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1407        most of the time, it *can* run out if it is given a pattern that contains a
1408        runaway infinite recursion. I updated the discussion in the pcrestack man
1409        page.
1411    13. Now that we have gone to the x.xx style of version numbers, the minor
1412        version may start with zero. Using 08 or 09 is a bad idea because users
1413        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1414        interpreted as invalid octal numbers. I've updated the previous comment in
1415        configure.ac, and also added a check that gives an error if 08 or 09 are
1416        used.
1418    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1419        causing partial matching to fail when the end of the subject matched \W
1420        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1422    15. There were some discrepancies between the declarations in pcre_internal.h
1423        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1424        their definitions. The declarations used "const uschar *" and the
1425        definitions used USPTR. Even though USPTR is normally defined as "const
1426        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1427        reported that: "This difference in casting confuses some C++ compilers, for
1428        example, SunCC recognizes above declarations as different functions and
1429        generates broken code for hbpcre." I have changed the declarations to use
1430        USPTR.
1432    16. GNU libtool is named differently on some systems. The autogen.sh script now
1433        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1434        (FreeBSD).
1436    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1437        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1438        comment: "Figure out how to create a longlong from a string: strtoll and
1439        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1440        instance, but it only takes 2 args instead of 3!"
1442    18. A subtle bug concerned with back references has been fixed by a change of
1443        specification, with a corresponding code fix. A pattern such as
1444        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1445        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1446        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1447        same bug. Such groups have to be quantified to be useful, or contained
1448        inside another quantified group. (If there's no repetition, the reference
1449        can never match.) The problem arises because, having left the group and
1450        moved on to the rest of the pattern, a later failure that backtracks into
1451        the group uses the captured value from the final iteration of the group
1452        rather than the correct earlier one. I have fixed this in PCRE by forcing
1453        any group that contains a reference to itself to be an atomic group; that
1454        is, there cannot be any backtracking into it once it has completed. This is
1455        similar to recursive and subroutine calls.
1458    Version 8.00 19-Oct-09
1459    ----------------------
1461    1.  The table for translating pcre_compile() error codes into POSIX error codes
1462        was out-of-date, and there was no check on the pcre_compile() error code
1463        being within the table. This could lead to an OK return being given in
1464        error.
1466    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1467        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1468        in a Windows environment.
1470    3.  The pcregrep --count option prints the count for each file even when it is
1471        zero, as does GNU grep. However, pcregrep was also printing all files when
1472        --files-with-matches was added. Now, when both options are given, it prints
1473        counts only for those files that have at least one match. (GNU grep just
1474        prints the file name in this circumstance, but including the count seems
1475        more useful - otherwise, why use --count?) Also ensured that the
1476        combination -clh just lists non-zero counts, with no names.
1478    4.  The long form of the pcregrep -F option was incorrectly implemented as
1479        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1480        but it seems right to fix it, and I didn't think it was worth preserving
1481        the old behaviour.
1483    5.  The command line items --regex=pattern and --regexp=pattern were not
1484        recognized by pcregrep, which required --regex pattern or --regexp pattern
1485        (with a space rather than an '='). The man page documented the '=' forms,
1486        which are compatible with GNU grep; these now work.
1488    6.  No libpcreposix.pc file was created for pkg-config; there was just
1489        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1491    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1492        when UCP support is not needed, by modifying the Python script that
1493        generates it from Unicode data files. This should not matter if the module
1494        is correctly used as a library, but I received one complaint about 50K of
1495        unwanted data. My guess is that the person linked everything into his
1496        program rather than using a library. Anyway, it does no harm.
1498    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1499        was a minimum greater than 1 for a wide character in a possessive
1500        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1501        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1502        character. Chaos in the form of incorrect output or a compiling loop could
1503        result.
1505    9.  The restrictions on what a pattern can contain when partial matching is
1506        requested for pcre_exec() have been removed. All patterns can now be
1507        partially matched by this function. In addition, if there are at least two
1508        slots in the offset vector, the offset of the earliest inspected character
1509        for the match and the offset of the end of the subject are set in them when
1510        PCRE_ERROR_PARTIAL is returned.
1512    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1513        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1514        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1515        and may be more useful for multi-segment matching.
1517    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1518        used to be given if ever the end of the subject was reached; now it is
1519        given only if matching could not proceed because another character was
1520        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1521        string "Z", which now yields "no match" instead of "partial match". In the
1522        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1523        final character ended with (*FAIL).
1525    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1526        if the pattern had a "must contain" character that was already found in the
1527        earlier partial match, unless partial matching was again requested. For
1528        example, with the pattern /dog.(body)?/, the "must contain" character is
1529        "g". If the first part-match was for the string "dog", restarting with
1530        "sbody" failed. This bug has been fixed.
1532    13. The string returned by pcre_dfa_exec() after a partial match has been
1533        changed so that it starts at the first inspected character rather than the
1534        first character of the match. This makes a difference only if the pattern
1535        starts with a lookbehind assertion or \b or \B (\K is not supported by
1536        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1537        matching functions compatible, and I think it's the right thing to do.
1539    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1540        so that the demonstration program is easily available in environments where
1541        PCRE has not been installed from source.
1543    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1544        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1545        library.
1547    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1548        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1549        is not the first non-POSIX option to be added. Clearly some people find
1550        these options useful.
1552    17. If a caller to the POSIX matching function regexec() passes a non-zero
1553        value for nmatch with a NULL value for pmatch, the value of
1554        nmatch is forced to zero.
1556    18. RunGrepTest did not have a test for the availability of the -u option of
1557        the diff command, as RunTest does. It now checks in the same way as
1558        RunTest, and also checks for the -b option.
1560    19. If an odd number of negated classes containing just a single character
1561        interposed, within parentheses, between a forward reference to a named
1562        subpattern and the definition of the subpattern, compilation crashed with
1563        an internal error, complaining that it could not find the referenced
1564        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1565        [The bug was that it was starting one character too far in when skipping
1566        over the character class, thus treating the ] as data rather than
1567        terminating the class. This meant it could skip too much.]
1569    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1570        /g option in pcretest when the pattern contains \K, which makes it possible
1571        to have an empty string match not at the start, even when the pattern is
1572        anchored. Updated pcretest and pcredemo to use this option.
1574    21. If the maximum number of capturing subpatterns in a recursion was greater
1575        than the maximum at the outer level, the higher number was returned, but
1576        with unset values at the outer level. The correct (outer level) value is
1577        now given.
1579    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1580        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1581        make it do so. The string so far is captured, making this feature
1582        compatible with Perl.
1584    23. The tests have been re-organized, adding tests 11 and 12, to make it
1585        possible to check the Perl 5.10 features against Perl 5.10.
1587    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1588        pattern matches a fixed length string. PCRE did not allow this; now it
1589        does. Neither allows recursion.
1591    25. I finally figured out how to implement a request to provide the minimum
1592        length of subject string that was needed in order to match a given pattern.
1593        (It was back references and recursion that I had previously got hung up
1594        on.) This code has now been added to pcre_study(); it finds a lower bound
1595        to the length of subject needed. It is not necessarily the greatest lower
1596        bound, but using it to avoid searching strings that are too short does give
1597        some useful speed-ups. The value is available to calling programs via
1598        pcre_fullinfo().
1600    26. While implementing 25, I discovered to my embarrassment that pcretest had
1601        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1602        study optimizations had never been tested with that matching function.
1603        Oops. What is worse, even when it was passed study data, there was a bug in
1604        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1605        were also very few tests of studied patterns with pcre_dfa_exec().
1607    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1608        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1609        on the other side of the coin, they are no longer allowed to have different
1610        names, because these cannot be distinguished in PCRE, and this has caused
1611        confusion. (This is a difference from Perl.)
1613    28. When duplicate subpattern names are present (necessarily with different
1614        numbers, as required by 27 above), and a test is made by name in a
1615        conditional pattern, either for a subpattern having been matched, or for
1616        recursion in such a pattern, all the associated numbered subpatterns are
1617        tested, and the overall condition is true if the condition is true for any
1618        one of them. This is the way Perl works, and is also more like the way
1619        testing by number works.
1622    Version 7.9 11-Apr-09
1623    ---------------------
1625    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1626        (pcretest), all targets were linked against these libraries. This included
1627        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1628        libraries. This caused unwanted dependencies to be created. This problem
1629        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1630        pcretest is linked with readline.
1632    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1633        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1634        moved outside it again, because FALSE and TRUE are already defined in AIX,
1635        but BOOL is not.
1637    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1638        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1640    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1641        hyphens) following file names and line numbers when outputting matching
1642        lines. This is not true; no spaces are inserted. I have also clarified the
1643        wording for the --colour (or --color) option.
1645    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1646        was not coloured; this is different to GNU grep, so I have changed it to be
1647        the same.
1649    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1650        each matching line was coloured. Now it goes on to look for further matches
1651        of any of the test patterns, which is the same behaviour as GNU grep.
1653    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1654        doesn't make sense to accept an empty string match in pcregrep, so I have
1655        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1656        seems to be how GNU grep behaves.
1658    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1659        start or after a newline", because the conditional assertion was not being
1660        correctly handled. The rule now is that both the assertion and what follows
1661        in the first alternative must satisfy the test.
1663    9.  If auto-callout was enabled in a pattern with a conditional group whose
1664        condition was an assertion, PCRE could crash during matching, both with
1665        pcre_exec() and pcre_dfa_exec().
1667    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1668        used for matching.
1670    11. Unicode property support in character classes was not working for
1671        characters (bytes) greater than 127 when not in UTF-8 mode.
1673    12. Added the -M command line option to pcretest.
1675    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1677    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1679    16. Added comments and documentation about mis-use of no_arg in the C++
1680        wrapper.
1682    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1683        from Martin Jerabek that uses macro names for all relevant character and
1684        string constants.
1686    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1687        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1688        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1689        these, but not everybody uses configure.
1691    19. A conditional group that had only one branch was not being correctly
1692        recognized as an item that could match an empty string. This meant that an
1693        enclosing group might also not be so recognized, causing infinite looping
1694        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1695        with the subject "ab", where knowledge that the repeated group can match
1696        nothing is needed in order to break the loop.
1698    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1699        exec(), but without supplying a callout function, matching went wrong.
1701    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1702        leak if the size of the offset vector was greater than 30. When the vector
1703        is smaller, the saved offsets during recursion go onto a local stack
1704        vector, but for larger vectors malloc() is used. It was failing to free
1705        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1706        error, in fact).
1708    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1709        heapframe that is used only when UTF-8 support is enabled. This caused no
1710        problem, but was untidy.
1712    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1713        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1714        included within another project.
1716    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1717        slightly modified by me:
1719          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1720              not building pcregrep.
1722          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1723              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1725    25. Forward references, both numeric and by name, in patterns that made use of
1726        duplicate group numbers, could behave incorrectly or give incorrect errors,
1727        because when scanning forward to find the reference group, PCRE was not
1728        taking into account the duplicate group numbers. A pattern such as
1729        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1731    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1732        the feature of a custom pointer more persuasive (as requested by a user).
1734    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1735        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1736        pre-defined.
1738    28. Added support for (*UTF8) at the start of a pattern.
1740    29. Arrange for flags added by the "release type" setting in CMake to be shown
1741        in the configuration summary.
1744    Version 7.8 05-Sep-08
1745    ---------------------
1747    1.  Replaced UCP searching code with optimized version as implemented for Ad
1748        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1749        stage table and inline lookup instead of a function, giving speed ups of 2
1750        to 5 times on some simple patterns that I tested. Permission was given to
1751        distribute the MultiStage2.py script that generates the tables (it's not in
1752        the tarball, but is in the Subversion repository).
1754    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1755        scripts.
1757    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1758        a group with a zero qualifier. The result of the study could be incorrect,
1759        or the function might crash, depending on the pattern.
1761    4.  Caseless matching was not working for non-ASCII characters in back
1762        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1763        It now works when Unicode Property Support is available.
1765    5.  In pcretest, an escape such as \x{de} in the data was always generating
1766        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1767        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1768        truncation.
1770    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1772    7.  Added two (int) casts to pcregrep when printing the difference of two
1773        pointers, in case they are 64-bit values.
1775    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1776        test 2 if it fails.
1778    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1779        and a #define of that name to empty if it is not externally set. This is to
1780        allow users of MSVC to set it if necessary.
1782    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1783        the convenience functions in the pcre_get.c source file.
1785    11. An option change at the start of a pattern that had top-level alternatives
1786        could cause overwriting and/or a crash. This command provoked a crash in
1787        some environments:
1789          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1791        This potential security problem was recorded as CVE-2008-2371.
1793    12. For a pattern where the match had to start at the beginning or immediately
1794        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1795        pcre_dfa_exec() could read past the end of the passed subject if there was
1796        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1797        pcretest so that it places the subject at the end of its malloc-ed buffer.
1799    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1800        exec() might read past the end of the data buffer in UTF-8 mode.
1802    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1803        the data contained the byte 0x85 as part of a UTF-8 character within its
1804        first line. This applied both to normal and DFA matching.
1806    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1807        /^[^d]*?$/8 failed to match "abc".
1809    16. Added a missing copyright notice to pcrecpp_internal.h.
1811    17. Make it more clear in the documentation that values returned from
1812        pcre_exec() in ovector are byte offsets, not character counts.
1814    18. Tidied a few places to stop certain compilers from issuing warnings.
1816    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1817        supplied by Stefan Weber. I made a further small update for 7.8 because
1818        there is a change of source arrangements: the pcre_searchfuncs.c module is
1819        replaced by pcre_ucd.c.
1822    Version 7.7 07-May-08
1823    ---------------------
1825    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1826        a string to a long long, pretend we don't even have a long long." This is
1827        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1829    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1830        pre-7.6 versions, which defined a global no_arg variable instead of putting
1831        it in the RE class. (See also #8 below.)
1833    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1834        Lopes.
1836    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1838        (1) The include/exclude patterns were being applied to the whole pathnames
1839            of files, instead of just to the final components.
1841        (2) If there was more than one level of directory, the subdirectories were
1842            skipped unless they satisfied the include/exclude conditions. This is
1843            inconsistent with GNU grep (and could even be seen as contrary to the
1844            pcregrep specification - which I improved to make it absolutely clear).
1845            The action now is always to scan all levels of directory, and just
1846            apply the include/exclude patterns to regular files.
1848    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1849        --exclude_dir in the tests to avoid scanning .svn directories.
1851    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1852        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1853        doesn't support NULs in patterns.
1855    7.  Added some missing "const"s to declarations of static tables in
1856        pcre_compile.c and pcre_dfa_exec.c.
1858    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1859        caused by fix #2  above. (Subsequently also a second patch to fix the
1860        first patch. And a third patch - this was a messy problem.)
1862    9.  Applied Craig's patch to remove the use of push_back().
1864    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1865        matching function regexec().
1867    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1868        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1869        references. PCRE supports relative numbers with this syntax (I don't think
1870        Oniguruma does).
1872    12. Previously, a group with a zero repeat such as (...){0} was completely
1873        omitted from the compiled regex. However, this means that if the group
1874        was called as a subroutine from elsewhere in the pattern, things went wrong
1875        (an internal error was given). Such groups are now left in the compiled
1876        pattern, with a new opcode that causes them to be skipped at execution
1877        time.
1879    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1880        to the way PCRE behaves:
1882        (a) A lone ] character is dis-allowed (Perl treats it as data).
1884        (b) A back reference to an unmatched subpattern matches an empty string
1885            (Perl fails the current match path).
1887        (c) A data ] in a character class must be notated as \] because if the
1888            first data character in a class is ], it defines an empty class. (In
1889            Perl it is not possible to have an empty class.) The empty class []
1890            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1891            The negative empty class [^] matches any one character, independently
1892            of the DOTALL setting.
1894    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1895        non-existent subpattern following a character class starting with ']' and
1896        containing () gave an internal compiling error instead of "reference to
1897        non-existent subpattern". Fortunately, when the pattern did exist, the
1898        compiled code was correct. (When scanning forwards to check for the
1899        existencd of the subpattern, it was treating the data ']' as terminating
1900        the class, so got the count wrong. When actually compiling, the reference
1901        was subsequently set up correctly.)
1903    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1904        it was being rejected as not supported by pcre_dfa_exec(), even though
1905        other assertions are supported. I have made pcre_dfa_exec() support
1906        (*FAIL).
1908    16. The implementation of 13c above involved the invention of a new opcode,
1909        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1910        cannot be changed at match time, I realized I could make a small
1911        improvement to matching performance by compiling OP_ALLANY instead of
1912        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1913        on the OP_ANY path.
1915    17. Compiling pcretest on Windows with readline support failed without the
1916        following two fixes: (1) Make the unistd.h include conditional on
1917        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1919    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1920        ncurses library to be included for pcretest when ReadLine support is
1921        requested, but also to allow for it to be overridden. This patch came from
1922        Daniel Bergström.
1924    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1925        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1926        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1927        spotting this.
1930    Version 7.6 28-Jan-08
1931    ---------------------
1933    1.  A character class containing a very large number of characters with
1934        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1935        overflow.
1937    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1938        HAVE_LONG_LONG is not defined.
1940    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1941        bring them up to date and include new features. This patch includes:
1943        - Fixed PH's badly added libz and libbz2 support.
1944        - Fixed a problem with static linking.
1945        - Added pcredemo. [But later removed - see 7 below.]
1946        - Fixed dftables problem and added an option.
1947        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1948            HAVE_LONG_LONG.
1949        - Added readline support for pcretest.
1950        - Added an listing of the option settings after cmake has run.
1952    4.  A user submitted a patch to Makefile that makes it easy to create
1953        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1954        Makefile.am that cause it to include this special target, without
1955        affecting anything else. Note that the same mingw target plus all
1956        the other distribution libraries and programs are now supported
1957        when configuring with CMake (see 6 below) instead of with
1958        Configure/Make.
1960    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1961        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1962        exported in the Windows port". It has not yet been confirmed that the patch
1963        solves the problem, but it does no harm.
1965    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1966        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1967        with CMake, and also correct the comment about stack recursion.
1969    7.  Remove the automatic building of pcredemo from the ./configure system and
1970        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1971        of a program that users should build themselves after PCRE is installed, so
1972        building it automatically is not really right. What is more, it gave
1973        trouble in some build environments.
1975    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1978    Version 7.5 10-Jan-08
1979    ---------------------
1981    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1982        values in parens when parsing an RE using the C++ wrapper."
1984    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1985        Characters greater than 255 were excluded from the class instead of being
1986        included.
1988    3.  The same bug as (2) above applied to negated POSIX classes such as
1989        [:^space:].
1991    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1992        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1993        I have changed it.
1995    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1996        first named subpattern) and a construct such as (?&a) would reference the
1997        first named subpattern whose name started with "a" (in other words, the
1998        length check was missing). Both these problems are fixed. "Subpattern name
1999        expected" is now given for (?&) (a zero-length name), and this patch also
2000        makes it give the same error for \k'' (previously it complained that that
2001        was a reference to a non-existent subpattern).
2003    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
2004        this is right because (?- can be followed by option settings as well as by
2005        digits. I have, however, made the messages clearer.
2007    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
2008        than the number used in the conditional) now cause a compile-time error.
2009        This is actually not compatible with Perl, which accepts such patterns, but
2010        treats the conditional as always being FALSE (as PCRE used to), but it
2011        seems to me that giving a diagnostic is better.
2013    8.  Change "alphameric" to the more common word "alphanumeric" in comments
2014        and messages.
2016    9.  Fix two occurrences of "backslash" in comments that should have been
2017        "backspace".
2019    10. Remove two redundant lines of code that can never be obeyed (their function
2020        was moved elsewhere).
2022    11. The program that makes PCRE's Unicode character property table had a bug
2023        which caused it to generate incorrect table entries for sequences of
2024        characters that have the same character type, but are in different scripts.
2025        It amalgamated them into a single range, with the script of the first of
2026        them. In other words, some characters were in the wrong script. There were
2027        thirteen such cases, affecting characters in the following ranges:
2029          U+002b0 - U+002c1
2030          U+0060c - U+0060d
2031          U+0061e - U+00612
2032          U+0064b - U+0065e
2033          U+0074d - U+0076d
2034          U+01800 - U+01805
2035          U+01d00 - U+01d77
2036          U+01d9b - U+01dbf
2037          U+0200b - U+0200f
2038          U+030fc - U+030fe
2039          U+03260 - U+0327f
2040          U+0fb46 - U+0fbb1
2041          U+10450 - U+1049d
2043    12. The -o option (show only the matching part of a line) for pcregrep was not
2044        compatible with GNU grep in that, if there was more than one match in a
2045        line, it showed only the first of them. It now behaves in the same way as
2046        GNU grep.
2048    13. If the -o and -v options were combined for pcregrep, it printed a blank
2049        line for every non-matching line. GNU grep prints nothing, and pcregrep now
2050        does the same. The return code can be used to tell if there were any
2051        non-matching lines.
2053    14. Added --file-offsets and --line-offsets to pcregrep.
2055    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
2056        infinitely looping recursion. The bug was that positive lookaheads were not
2057        being skipped when checking for a possible empty match (negative lookaheads
2058        and both kinds of lookbehind were skipped).
2060    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
2061        inclusion of <windows.h> to before rather than after the definition of
2062        INVALID_FILE_ATTRIBUTES (patch from David Byron).
2064    17. Specifying a possessive quantifier with a specific limit for a Unicode
2065        character property caused pcre_compile() to compile bad code, which led at
2066        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
2067        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
2068        caused the error; without that there was no problem.
2070    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
2072    19. Added --enable-pcretest-libreadline.
2074    20. In pcrecpp.cc, the variable 'count' was incremented twice in
2075        RE::GlobalReplace(). As a result, the number of replacements returned was
2076        double what it should be. I removed one of the increments, but Craig sent a
2077        later patch that removed the other one (the right fix) and added unit tests
2078        that check the return values (which was not done before).
2080    21. Several CMake things:
2082        (1) Arranged that, when cmake is used on Unix, the libraries end up with
2083            the names libpcre and libpcreposix, not just pcre and pcreposix.
2085        (2) The above change means that pcretest and pcregrep are now correctly
2086            linked with the newly-built libraries, not previously installed ones.
2090    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
2091        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
2092        UTF-8 newline character). The key issue is that the pattern starts .*;
2093        this means that the match must be either at the beginning, or after a
2094        newline. The bug was in the code for advancing after a failed match and
2095        checking that the new position followed a newline. It was not taking
2096        account of UTF-8 characters correctly.
2098    23. PCRE was behaving differently from Perl in the way it recognized POSIX
2099        character classes. PCRE was not treating the sequence [:...:] as a
2100        character class unless the ... were all letters. Perl, however, seems to
2101        allow any characters between [: and :], though of course it rejects as
2102        unknown any "names" that contain non-letters, because all the known class
2103        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
2104        for example, whereas PCRE did not - it did not recognize a POSIX character
2105        class. This seemed a bit dangerous, so the code has been changed to be
2106        closer to Perl. The behaviour is not identical to Perl, because PCRE will
2107        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
2108        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
2109        Perl does, and where it didn't before.
2111    24. Rewrite so as to remove the single use of %n from pcregrep because in some
2112        Windows environments %n is disabled by default.
2115    Version 7.4 21-Sep-07
2116    ---------------------
2118    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
2119        means that a class such as [\s] counted as "explicit reference to CR or
2120        LF". That isn't really right - the whole point of the change was to try to
2121        help when there was an actual mention of one of the two characters. So now
2122        the change happens only if \r or \n (or a literal CR or LF) character is
2123        encountered.
2125    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
2126        of both had grown to the point where there were only 3 bits left.
2127        Fortunately, there was spare space in the data structure, and so I have
2128        moved the internal flags into a new 16-bit field to free up more option
2129        bits.
2131    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
2132        but did not set the internal JCHANGED flag - either of these is enough to
2133        control the way the "get" function works - but the PCRE_INFO_JCHANGED
2134        facility is supposed to tell if (?J) was ever used, so now (?J) at the
2135        start sets both bits.
2137    4.  Added options (at build time, compile time, exec time) to change \R from
2138        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
2140    5.  doc/pcresyntax.html was missing from the distribution.
2142    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
2143        compatibility, even though it is no longer used.
2145    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
2146        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
2147        windows.h file is present (where different names are used). [This was
2148        reversed later after testing - see 16 below.]
2150    8.  Changed all #include <config.h> to #include "config.h". There were also
2151        some further <pcre.h> cases that I changed to "pcre.h".
2153    9.  When pcregrep was used with the --colour option, it missed the line ending
2154        sequence off the lines that it output.
2156    10. It was pointed out to me that arrays of string pointers cause lots of
2157        relocations when a shared library is dynamically loaded. A technique of
2158        using a single long string with a table of offsets can drastically reduce
2159        these. I have refactored PCRE in four places to do this. The result is
2160        dramatic:
2162          Originally:                          290
2163          After changing UCP table:            187
2164          After changing error message table:   43
2165          After changing table of "verbs"       36
2166          After changing table of Posix names   22
2168        Thanks to the folks working on Gregex for glib for this insight.
2170    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
2171        unicode-properties was also set.
2173    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
2175    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
2176        checked only for CRLF.
2178    14. Added casts to pcretest.c to avoid compiler warnings.
2180    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
2182    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
2183        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2184        entirely. This removes changes made in 7 above.
2186    17. The CMake files have been updated, and there is now more information about
2187        building with CMake in the NON-UNIX-USE document.
2190    Version 7.3 28-Aug-07
2191    ---------------------
2193     1. In the rejigging of the build system that eventually resulted in 7.1, the
2194        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
2195        brackets there is not right, since it causes compilers to look for an
2196        installed pcre.h, not the version that is in the source that is being
2197        compiled (which of course may be different). I have changed it back to:
2199          #include "pcre.h"
2201        I have a vague recollection that the change was concerned with compiling in
2202        different directories, but in the new build system, that is taken care of
2203        by the VPATH setting the Makefile.
2205     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
2206        when the subject happened to end in the byte 0x85 (e.g. if the last
2207        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
2208        characters but of course it shouldn't be taken as a newline when it is part
2209        of another character. The bug was that, for an unlimited repeat of . in
2210        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
2211        characters when looking for a newline.
2213     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
2215     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
2216        in debug output.
2218     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
2219        long printing in the pcrecpp unittest when running under MinGW.
2221     6. ESC_K was left out of the EBCDIC table.
2223     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
2224        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
2225        limit also applies to "virtual nesting" when a pattern is recursive, and in
2226        this case 1000 isn't so big. I have been able to remove this limit at the
2227        expense of backing off one optimization in certain circumstances. Normally,
2228        when pcre_exec() would call its internal match() function recursively and
2229        immediately return the result unconditionally, it uses a "tail recursion"
2230        feature to save stack. However, when a subpattern that can match an empty
2231        string has an unlimited repetition quantifier, it no longer makes this
2232        optimization. That gives it a stack frame in which to save the data for
2233        checking that an empty string has been matched. Previously this was taken
2234        from the 1000-entry workspace that had been reserved. So now there is no
2235        explicit limit, but more stack is used.
2237     8. Applied Daniel's patches to solve problems with the import/export magic
2238        syntax that is required for Windows, and which was going wrong for the
2239        pcreposix and pcrecpp parts of the library. These were overlooked when this
2240        problem was solved for the main library.
2242     9. There were some crude static tests to avoid integer overflow when computing
2243        the size of patterns that contain repeated groups with explicit upper
2244        limits. As the maximum quantifier is 65535, the maximum group length was
2245        set at 30,000 so that the product of these two numbers did not overflow a
2246        32-bit integer. However, it turns out that people want to use groups that
2247        are longer than 30,000 bytes (though not repeat them that many times).
2248        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
2249        made it possible to implement the integer overflow checks in a much more
2250        dynamic way, which I have now done. The artificial limitation on group
2251        length has been removed - we now have only the limit on the total length of
2252        the compiled pattern, which depends on the LINK_SIZE setting.
2254    10. Fixed a bug in the documentation for get/copy named substring when
2255        duplicate names are permitted. If none of the named substrings are set, the
2256        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
2257        empty string.
2259    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
2260        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
2261        because the ] is interpreted as the first data character and the
2262        terminating ] is not found. PCRE has been made compatible with Perl in this
2263        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
2264        cause memory overwriting.
2266    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
2267        string has been matched (to stop an infinite loop). It was not recognizing
2268        a conditional subpattern that could match an empty string if that
2269        subpattern was within another subpattern. For example, it looped when
2270        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
2271        condition was not nested. This bug has been fixed.
2273    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
2274        past the start of the subject in the presence of bytes with the top bit
2275        set, for example "\x8aBCD".
2277    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
2278        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
2280    14. Optimized (?!) to (*FAIL).
2282    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
2283        This restricts code points to be within the range 0 to 0x10FFFF, excluding
2284        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
2285        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
2286        does: it's just the validity check that is more restrictive.
2288    16. Inserted checks for integer overflows during escape sequence (backslash)
2289        processing, and also fixed erroneous offset values for syntax errors during
2290        backslash processing.
2292    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
2293        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
2295    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
2296        caused an overrun.
2298    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
2299        something other than just ASCII characters) inside a group that had an
2300        unlimited repeat caused a loop at compile time (while checking to see
2301        whether the group could match an empty string).
2303    20. Debugging a pattern containing \p or \P could cause a crash. For example,
2304        [\P{Any}] did so. (Error in the code for printing property names.)
2306    21. An orphan \E inside a character class could cause a crash.
2308    22. A repeated capturing bracket such as (A)? could cause a wild memory
2309        reference during compilation.
2311    23. There are several functions in pcre_compile() that scan along a compiled
2312        expression for various reasons (e.g. to see if it's fixed length for look
2313        behind). There were bugs in these functions when a repeated \p or \P was
2314        present in the pattern. These operators have additional parameters compared
2315        with \d, etc, and these were not being taken into account when moving along
2316        the compiled data. Specifically:
2318        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2319            length.
2321        (b) An item such as \pL+ within a repeated group could cause crashes or
2322            loops.
2324        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2325            "reference to non-existent subpattern" error.
2327        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2329    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2330        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2332    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2334    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2335        character were causing crashes (broken optimization).
2337    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2338        \p or \P) caused a compile-time loop.
2340    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2341        break. For example, the unstudied pattern [\r\n]A does not match the string
2342        "\r\nA" because change 7.0/46 below moves the current point on by two
2343        characters after failing to match at the start. However, the pattern \nA
2344        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2345        the same is true. There doesn't seem any very clean way out of this, but
2346        what I have chosen to do makes the common cases work: PCRE now takes note
2347        of whether there can be an explicit match for \r or \n anywhere in the
2348        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2349        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2350        pattern has explicit CR or LF references.
2352    29. Added (*CR) etc for changing newline setting at start of pattern.
2355    Version 7.2 19-Jun-07
2356    ---------------------
2358     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
2359        which is apparently normally available under Windows.
2361     2. Re-jig the pcregrep tests with different newline settings in an attempt
2362        to make them independent of the local environment's newline setting.
2364     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
2366     4. Some of the "internals" tests were previously cut out when the link size
2367        was not 2, because the output contained actual offsets. The recent new
2368        "Z" feature of pcretest means that these can be cut out, making the tests
2369        usable with all link sizes.
2371     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
2372        stack recursion. This gives a massive performance boost under BSD, but just
2373        a small improvement under Linux. However, it saves one field in the frame
2374        in all cases.
2376     6. Added more features from the forthcoming Perl 5.10:
2378        (a) (?-n) (where n is a string of digits) is a relative subroutine or
2379            recursion call. It refers to the nth most recently opened parentheses.
2381        (b) (?+n) is also a relative subroutine call; it refers to the nth next
2382            to be opened parentheses.
2384        (c) Conditions that refer to capturing parentheses can be specified
2385            relatively, for example, (?(-2)... or (?(+3)...
2387        (d) \K resets the start of the current match so that everything before
2388            is not part of it.
2390        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2392        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2393            reference syntax.
2395        (g) (?| introduces a group in which the numbering of parentheses in each
2396            alternative starts with the same number.
2398        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2400     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2403     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2404        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2405        for detecting groups that can match an empty string.
2407     9. A pattern with a very large number of alternatives (more than several
2408        hundred) was running out of internal workspace during the pre-compile
2409        phase, where pcre_compile() figures out how much memory will be needed. A
2410        bit of new cunning has reduced the workspace needed for groups with
2411        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2412        workspace instead of running out of the 4096 that are available.
2414    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2416    11. Applied patch from Google to remove an optimization that didn't quite work.
2417        The report of the bug said:
2419          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2420          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2421          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2423    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2424        it matched the wrong number of bytes.
2427    Version 7.1 24-Apr-07
2428    ---------------------
2430     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
2431        that is more "standard", making use of automake and other Autotools. There
2432        is some re-arrangement of the files and adjustment of comments consequent
2433        on this.
2435     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
2436        for recursive directory scanning broke on some systems because the files
2437        are not scanned in any specific order and on different systems the order
2438        was different. A call to "sort" has been inserted into RunGrepTest for the
2439        approprate test as a short-term fix. In the longer term there may be an
2440        alternative.
2442     3. I had an email from Eric Raymond about problems translating some of PCRE's
2443        man pages to HTML (despite the fact that I distribute HTML pages, some
2444        people do their own conversions for various reasons). The problems
2445        concerned the use of low-level troff macros .br and .in. I have therefore
2446        removed all such uses from the man pages (some were redundant, some could
2447        be replaced by .nf/.fi pairs). The 132html script that I use to generate
2448        HTML has been updated to handle .nf/.fi and to complain if it encounters
2449        .br or .in.
2451     4. Updated comments in configure.ac that get placed in config.h.in and also
2452        arranged for config.h to be included in the distribution, with the name
2453        config.h.generic, for the benefit of those who have to compile without
2454        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
2456     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
2457        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
2458        makevp.bat for the current PCRE, using the additional files
2459        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
2461     6. A Windows user reported a minor discrepancy with test 2, which turned out
2462        to be caused by a trailing space on an input line that had got lost in his
2463        copy. The trailing space was an accident, so I've just removed it.
2465     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
2466        that is needed.
2468     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
2469        as "const" (a) because they are and (b) because it helps the PHP
2470        maintainers who have recently made a script to detect big data structures
2471        in the php code that should be moved to the .rodata section. I remembered
2472        to update Builducptable as well, so it won't revert if ucptable.h is ever
2473        re-created.
2475     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
2476        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
2477        order to be able to cut out the UTF-8 tables in the latter when UTF-8
2478        support is not required. This saves 1.5-2K of code, which is important in
2479        some applications.
2481        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
2482        so as not to refer to the tables, even though these functions will never be
2483        called when UTF-8 support is disabled. Otherwise there are problems with a
2484        shared library.
2486    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
2488        (a) It was defining its arguments as char * instead of void *.
2490        (b) It was assuming that all moves were upwards in memory; this was true
2491            a long time ago when I wrote it, but is no longer the case.
2493        The emulated memove() is provided for those environments that have neither
2494        memmove() nor bcopy(). I didn't think anyone used it these days, but that
2495        is clearly not the case, as these two bugs were recently reported.
2497    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
2498        and Detrail to create the HTML documentation, the .txt form of the man
2499        pages, and it removes trailing spaces from listed files. It also creates
2500        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
2501        case, it wraps all the #defines with #ifndefs. This script should be run
2502        before "make dist".
2504    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
2505        with Unicode property support.
2507        (a) For a maximizing quantifier, if the two different cases of the
2508            character were of different lengths in their UTF-8 codings (there are
2509            some cases like this - I found 11), and the matching function had to
2510            back up over a mixture of the two cases, it incorrectly assumed they
2511            were both the same length.
2513        (b) When PCRE was configured to use the heap rather than the stack for
2514            recursion during matching, it was not correctly preserving the data for
2515            the other case of a UTF-8 character when checking ahead for a match
2516            while processing a minimizing repeat. If the check also involved
2517            matching a wide character, but failed, corruption could cause an
2518            erroneous result when trying to check for a repeat of the original
2519            character.
2521    13. Some tidying changes to the testing mechanism:
2523        (a) The RunTest script now detects the internal link size and whether there
2524            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
2525            values substituted by "configure". (The RunGrepTest script already did
2526            this for UTF-8.) The configure.ac script no longer substitutes the
2527            relevant variables.
2529        (b) The debugging options /B and /D in pcretest show the compiled bytecode
2530            with length and offset values. This means that the output is different
2531            for different internal link sizes. Test 2 is skipped for link sizes
2532            other than 2 because of this, bypassing the problem. Unfortunately,
2533            there was also a test in test 3 (the locale tests) that used /B and
2534            failed for link sizes other than 2. Rather than cut the whole test out,
2535            I have added a new /Z option to pcretest that replaces the length and
2536            offset values with spaces. This is now used to make test 3 independent
2537            of link size. (Test 2 will be tidied up later.)
2539    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
2540        segmentation fault instead of returning the appropriate error message.
2542    15. In multiline mode when the newline sequence was set to "any", the pattern
2543        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
2544        This doesn't seem right; it now treats the CRLF combination as the line
2545        ending, and so does not match in that case. It's only a pattern such as ^$
2546        that would hit this one: something like ^ABC$ would have failed after \r
2547        and then tried again after \r\n.
2549    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
2550        in an attempt to make files that differ only in their line terminators
2551        compare equal. This works on Linux.
2553    17. Under certain error circumstances pcregrep might try to free random memory
2554        as it exited. This is now fixed, thanks to valgrind.
2556    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
2557        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
2558        was because its rules for how to advance for /g after matching an empty
2559        string at the end of a line did not allow for this case. They now check for
2560        it specially.
2562    20. pcretest is supposed to handle patterns and data of any length, by
2563        extending its buffers when necessary. It was getting this wrong when the
2564        buffer for a data line had to be extended.
2566    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
2567        CRLF as a newline sequence.
2569    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
2570        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
2571        I have nevertheless tidied it up.
2573    23. Added some casts to kill warnings from HP-UX ia64 compiler.
2575    24. Added a man page for pcre-config.
2578    Version 7.0 19-Dec-06
2579    ---------------------
2581     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
2582        moving to gcc 4.1.1.
2584     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
2585        sys/time.h, which is documented as needed for this function. It doesn't
2586        seem to matter on Linux, but it showed up on some releases of OS X.
2588     3. It seems that there are systems where bytes whose values are greater than
2589        127 match isprint() in the "C" locale. The "C" locale should be the
2590        default when a C program starts up. In most systems, only ASCII printing
2591        characters match isprint(). This difference caused the output from pcretest
2592        to vary, making some of the tests fail. I have changed pcretest so that:
2594        (a) When it is outputting text in the compiled version of a pattern, bytes
2595            other than 32-126 are always shown as hex escapes.
2597        (b) When it is outputting text that is a matched part of a subject string,
2598            it does the same, unless a different locale has been set for the match
2599            (using the /L modifier). In this case, it uses isprint() to decide.
2601     4. Fixed a major bug that caused incorrect computation of the amount of memory
2602        required for a compiled pattern when options that changed within the
2603        pattern affected the logic of the preliminary scan that determines the
2604        length. The relevant options are -x, and -i in UTF-8 mode. The result was
2605        that the computed length was too small. The symptoms of this bug were
2606        either the PCRE error "internal error: code overflow" from pcre_compile(),
2607        or a glibc crash with a message such as "pcretest: free(): invalid next
2608        size (fast)". Examples of patterns that provoked this bug (shown in
2609        pcretest format) are:
2611          /(?-x: )/x
2612          /(?x)(?-x: \s*#\s*)/
2613          /((?i)[\x{c0}])/8
2614          /(?i:[\x{c0}])/8
2616        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
2617        is now done differently.
2619     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
2620        wrapper classes; (b) implement a new function in the C++ scanner that is
2621        more efficient than the old way of doing things because it avoids levels of
2622        recursion in the regex matching; (c) add a paragraph to the documentation
2623        for the FullMatch() function.
2625     6. The escape sequence \n was being treated as whatever was defined as
2626        "newline". Not only was this contrary to the documentation, which states
2627        that \n is character 10 (hex 0A), but it also went horribly wrong when
2628        "newline" was defined as CRLF. This has been fixed.
2630     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
2631        was being set to -1 for the "end of line" case (supposedly a value that no
2632        character can have). Though this value is never used (the check for end of
2633        line is "zero bytes in current character"), it caused compiler complaints.
2634        I've changed it to 0xffffffff.
2636     8. In pcre_version.c, the version string was being built by a sequence of
2637        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
2638        string (as it is for production releases) called a macro with an empty
2639        argument. The C standard says the result of this is undefined. The gcc
2640        compiler treats it as an empty string (which was what was wanted) but it is
2641        reported that Visual C gives an error. The source has been hacked around to
2642        avoid this problem.
2644     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
2645        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
2646        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
2647        of them did).
2649    10. Originally, pcretest opened its input and output without "b"; then I was
2650        told that "b" was needed in some environments, so it was added for release
2651        5.0 to both the input and output. (It makes no difference on Unix-like
2652        systems.) Later I was told that it is wrong for the input on Windows. I've
2653        now abstracted the modes into two macros, to make it easier to fiddle with
2654        them, and removed "b" from the input mode under Windows.
2656    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
2658    12. Added -help and --help to pcretest as an official way of being reminded
2659        of the options.
2661    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
2662        and pcrecpp.cc because they annoy compilers at high warning levels.
2664    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
2666    15. Fixed an occurrence of == in configure.ac that should have been = (shell
2667        scripts are not C programs :-) and which was not noticed because it works
2668        on Linux.
2670    16. pcretest is supposed to handle any length of pattern and data line (as one
2671        line or as a continued sequence of lines) by extending its input buffer if
2672        necessary. This feature was broken for very long pattern lines, leading to
2673        a string of junk being passed to pcre_compile() if the pattern was longer
2674        than about 50K.
2676    17. I have done a major re-factoring of the way pcre_compile() computes the
2677        amount of memory needed for a compiled pattern. Previously, there was code
2678        that made a preliminary scan of the pattern in order to do this. That was
2679        OK when PCRE was new, but as the facilities have expanded, it has become
2680        harder and harder to keep it in step with the real compile phase, and there
2681        have been a number of bugs (see for example, 4 above). I have now found a
2682        cunning way of running the real compile function in a "fake" mode that
2683        enables it to compute how much memory it would need, while actually only
2684        ever using a few hundred bytes of working memory and without too many
2685        tests of the mode. This should make future maintenance and development
2686        easier. A side effect of this work is that the limit of 200 on the nesting
2687        depth of parentheses has been removed (though this was never a serious
2688        limitation, I suspect). However, there is a downside: pcre_compile() now
2689        runs more slowly than before (30% or more, depending on the pattern). I
2690        hope this isn't a big issue. There is no effect on runtime performance.
2692    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
2693        newline (only possible for the last line of a file) and it was a
2694        pattern that set a locale (followed by /Lsomething), pcretest crashed.
2696    19. Added additional timing features to pcretest. (1) The -tm option now times
2697        matching only, not compiling. (2) Both -t and -tm can be followed, as a
2698        separate command line item, by a number that specifies the number of
2699        repeats to use when timing. The default is 50000; this gives better
2700        precision, but takes uncomfortably long for very large patterns.
2702    20. Extended pcre_study() to be more clever in cases where a branch of a
2703        subpattern has no definite first character. For example, (a*|b*)[cd] would
2704        previously give no result from pcre_study(). Now it recognizes that the
2705        first character must be a, b, c, or d.
2707    21. There was an incorrect error "recursive call could loop indefinitely" if
2708        a subpattern (or the entire pattern) that was being tested for matching an
2709        empty string contained only one non-empty item after a nested subpattern.
2710        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
2711        incorrectly, because the \d was being skipped in the check.
2713    22. The pcretest program now has a new pattern option /B and a command line
2714        option -b, which is equivalent to adding /B to every pattern. This causes
2715        it to show the compiled bytecode, without the additional information that
2716        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
2717        is the same as /B/I).
2719    23. A new optimization is now able automatically to treat some sequences such
2720        as a*b as a*+b. More specifically, if something simple (such as a character
2721        or a simple class like \d) has an unlimited quantifier, and is followed by
2722        something that cannot possibly match the quantified thing, the quantifier
2723        is automatically "possessified".
2725    24. A recursive reference to a subpattern whose number was greater than 39
2726        went wrong under certain circumstances in UTF-8 mode. This bug could also
2727        have affected the operation of pcre_study().
2729    25. Realized that a little bit of performance could be had by replacing
2730        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
2732    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
2734    27. Possessive quantifiers such as a++ were previously implemented by turning
2735        them into atomic groups such as ($>a+). Now they have their own opcodes,
2736        which improves performance. This includes the automatically created ones
2737        from 23 above.
2739    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
2740        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
2741        the first matched character to be a colon. This applied both to named and
2742        numbered groups.
2744    29. The ucpinternal.h header file was missing its idempotency #ifdef.
2746    30. I was sent a "project" file called libpcre.a.dev which I understand makes
2747        building PCRE on Windows easier, so I have included it in the distribution.
2749    31. There is now a check in pcretest against a ridiculously large number being
2750        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
2751        loop, the loop is abandoned.
2753    32. Forward references to subpatterns in conditions such as (?(2)...) where
2754        subpattern 2 is defined later cause pcre_compile() to search forwards in
2755        the pattern for the relevant set of parentheses. This search went wrong
2756        when there were unescaped parentheses in a character class, parentheses
2757        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2759    33. "Subroutine" calls and backreferences were previously restricted to
2760        referencing subpatterns earlier in the regex. This restriction has now
2761        been removed.
2763    34. Added a number of extra features that are going to be in Perl 5.10. On the
2764        whole, these are just syntactic alternatives for features that PCRE had
2765        previously implemented using the Python syntax or my own invention. The
2766        other formats are all retained for compatibility.
2768        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2769            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2770            also .NET compatible.
2772        (b) A recursion or subroutine call to a named group can now be defined as
2773            (?&name) as well as (?P>name).
2775        (c) A backreference to a named group can now be defined as \k<name> or
2776            \k'name' as well as (?P=name). The new forms, as well as being in Perl
2777            5.10, are also .NET compatible.
2779        (d) A conditional reference to a named group can now use the syntax
2780            (?(<name>) or (?('name') as well as (?(name).
2782        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2783            groups (named and numbered) that are never evaluated inline, but can be
2784            called as "subroutines" from elsewhere. In effect, the DEFINE condition
2785            is always false. There may be only one alternative in such a group.
2787        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2788            as the simple (?(R). The condition is true only if the most recent
2789            recursion is that of the given number or name. It does not search out
2790            through the entire recursion stack.
2792        (g) The escape \gN or \g{N} has been added, where N is a positive or
2793            negative number, specifying an absolute or relative reference.
2795    35. Tidied to get rid of some further signed/unsigned compiler warnings and
2796        some "unreachable code" warnings.
2798    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2799        things, this adds five new scripts.
2801    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2802        There were also incompatibilities regarding the handling of \Q..\E inside
2803        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2804        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2806    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2807        matches an empty string, and forcibly breaks the loop. There were bugs in
2808        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2809        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2810        separate and independent bugs (that affected different cases) have been
2811        fixed.
2813    39. Refactored the code to abolish the use of different opcodes for small
2814        capturing bracket numbers. This is a tidy that I avoided doing when I
2815        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2816        The new approach is not only tidier, it makes it possible to reduce the
2817        memory needed to fix the previous bug (38).
2819    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2820        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2821        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2822        mode.
2824    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2825        report.
2827    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2828        copy construction and assignment in the C++ wrapper.
2830    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2831        couple of bugs that could have given wrong results in the "--newline=crlf"
2832        case.
2834    44. Added a number of casts and did some reorganization of signed/unsigned int
2835        variables following suggestions from Dair Grant. Also renamed the variable
2836        "this" as "item" because it is a C++ keyword.
2838    45. Arranged for dftables to add
2840          #include "pcre_internal.h"
2842        to pcre_chartables.c because without it, gcc 4.x may remove the array
2843        definition from the final binary if PCRE is built into a static library and
2844        dead code stripping is activated.
2846    46. For an unanchored pattern, if a match attempt fails at the start of a
2847        newline sequence, and the newline setting is CRLF or ANY, and the next two
2848        characters are CRLF, advance by two characters instead of one.
2851  Version 6.7 04-Jul-06  Version 6.7 04-Jul-06
2852  ---------------------  ---------------------

Removed from v.91  
changed lines
  Added in v.1319

  ViewVC Help
Powered by ViewVC 1.1.5