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 1328 by ph10, Sun May 12 16:28:22 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.
148    39. Try madvise first before posix_madvise.
150    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
151        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
152        original change locked out all matches of empty strings. This has been
153        changed so that one match of an empty string per line is recognized.
154        Subsequent searches on the same line (for colouring or for --only-matching,
155        for example) do not recognize empty strings.
157    41. Applied a user patch to fix a number of spelling mistakes in comments.
159    42. Data lines longer than 65536 caused pcretest to crash.
161    44. Clarified the data type for length and startoffset arguments for pcre_exec
162        and pcre_dfa_exec in the function-specific man pages, where they were
163        explicitly stated to be in bytes, never having been updated. I also added
164        some clarification to the pcreapi man page.
167    Version 8.32 30-November-2012
168    -----------------------------
170    1.  Improved JIT compiler optimizations for first character search and single
171        character iterators.
173    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
174        Patch by Daniel Richard G.
176    3.  Single character iterator optimizations in the JIT compiler.
178    4.  Improved JIT compiler optimizations for character ranges.
180    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
181        Reported by Giuseppe D'Angelo.
183    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
184        of a line, was being set incorrectly in cases where .* appeared inside
185        atomic brackets at the start of a pattern, or where there was a subsequent
186        *PRUNE or *SKIP.
188    7.  Improved instruction cache flush for POWER/PowerPC.
189        Patch by Daniel Richard G.
191    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
192        grep:
194        (a) There is now no limit to the number of patterns to be matched.
196        (b) An error is given if a pattern is too long.
198        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
199            are now supported.
201        (d) --exclude-from and --include-from (multiple use) have been added.
203        (e) Exclusions and inclusions now apply to all files and directories, not
204            just to those obtained from scanning a directory recursively.
206        (f) Multiple uses of -f and --file-list are now supported.
208        (g) In a Windows environment, the default for -d has been changed from
209            "read" (the GNU grep default) to "skip", because otherwise the presence
210            of a directory in the file list provokes an error.
212        (h) The documentation has been revised and clarified in places.
214    9.  Improve the matching speed of capturing brackets.
216    10. Changed the meaning of \X so that it now matches a Unicode extended
217        grapheme cluster.
219    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
220        out POSIX threads when JIT support is configured.
222    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
224    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
225        structure could go wrong in environments where size_t is not the same size
226        as int.
228    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
229        set.
231    15. The EBCDIC support had decayed; later updates to the code had included
232        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
233        general tidy up of EBCDIC-related issues, and the documentation was also
234        not quite right. There is now a test that can be run on ASCII systems to
235        check some of the EBCDIC-related things (but is it not a full test).
237    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
238        in a small tidy to the code.
240    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
242    18. If the --only-matching (-o) option in pcregrep is specified multiple
243        times, each one causes appropriate output. For example, -o1 -o2 outputs the
244        substrings matched by the 1st and 2nd capturing parentheses. A separating
245        string can be specified by --om-separator (default empty).
247    19. Improving the first n character searches.
249    20. Turn case lists for horizontal and vertical white space into macros so that
250        they are defined only once.
252    21. This set of changes together give more compatible Unicode case-folding
253        behaviour for characters that have more than one other case when UCP
254        support is available.
256        (a) The Unicode property table now has offsets into a new table of sets of
257            three or more characters that are case-equivalent. The MultiStage2.py
258            script that generates these tables (the pcre_ucd.c file) now scans
259            CaseFolding.txt instead of UnicodeData.txt for character case
260            information.
262        (b) The code for adding characters or ranges of characters to a character
263            class has been abstracted into a generalized function that also handles
264            case-independence. In UTF-mode with UCP support, this uses the new data
265            to handle characters with more than one other case.
267        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
268            whose other case is greater than 256 are now correctly matched
269            caselessly. Previously, the high codepoint matched the low one, but not
270            vice versa.
272        (d) The processing of \h, \H, \v, and \ in character classes now makes use
273            of the new class addition function, using character lists defined as
274            macros alongside the case definitions of 20 above.
276        (e) Caseless back references now work with characters that have more than
277            one other case.
279        (f) General caseless matching of characters with more than one other case
280            is supported.
282    22. Unicode character properties were updated from Unicode 6.2.0
284    23. Improved CMake support under Windows. Patch by Daniel Richard G.
286    24. Add support for 32-bit character strings, and UTF-32
288    25. Major JIT compiler update (code refactoring and bugfixing).
289        Experimental Sparc 32 support is added.
291    26. Applied a modified version of Daniel Richard G's patch to create
292        pcre.h.generic and config.h.generic by "make" instead of in the
293        PrepareRelease script.
295    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
296        pcre_compile.c when checking for a zero character.
298    28. Introducing a native interface for JIT. Through this interface, the compiled
299        machine code can be directly executed. The purpose of this interface is to
300        provide fast pattern matching, so several sanity checks are not performed.
301        However, feature tests are still performed. The new interface provides
302        1.4x speedup compared to the old one.
304    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
305        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
306        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
308    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
309        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
310        the "old" RFC 2279). Instead, it ended up passing a negative length to
311        pcre_exec().
313    31. Add support for GCC's visibility feature to hide internal functions.
315    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
316        "unknown -C option" after outputting 0 or 1.
318    33. There is now support for generating a code coverage report for the test
319        suite in environments where gcc is the compiler and lcov is installed. This
320        is mainly for the benefit of the developers.
322    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
323        unaddressable using valgrind annotations, allowing valgrind to detect
324        invalid memory accesses. This is mainly for the benefit of the developers.
326    25. (*UTF) can now be used to start a pattern in any of the three libraries.
328    26. Give configure error if --enable-cpp but no C++ compiler found.
331    Version 8.31 06-July-2012
332    -------------------------
334    1.  Fixing a wrong JIT test case and some compiler warnings.
336    2.  Removed a bashism from the RunTest script.
338    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
339        to unsigned type, result still unsigned" that was given by an MS compiler
340        on encountering the code "-sizeof(xxx)".
342    4.  Partial matching support is added to the JIT compiler.
344    5.  Fixed several bugs concerned with partial matching of items that consist
345        of more than one character:
347        (a) /^(..)\1/ did not partially match "aba" because checking references was
348            done on an "all or nothing" basis. This also applied to repeated
349            references.
351        (b) \R did not give a hard partial match if \r was found at the end of the
352            subject.
354        (c) \X did not give a hard partial match after matching one or more
355            characters at the end of the subject.
357        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
358            a partial match for the string "\r".
360        (e) When newline was set to CRLF, the metacharacter "." did not recognize
361            a partial match for a CR character at the end of the subject string.
363    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
364        running pcretest, the text "(JIT)" added to the output whenever JIT is
365        actually used to run the match.
367    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
368        or /S+[+] with a digit between 1 and 7.
370    8.  OP_NOT now supports any UTF character not just single-byte ones.
372    9.  (*MARK) control verb is now supported by the JIT compiler.
374    10. The command "./RunTest list" lists the available tests without actually
375        running any of them. (Because I keep forgetting what they all are.)
379    12. Applied a (slightly modified) user-supplied patch that improves performance
380        when the heap is used for recursion (compiled with --disable-stack-for-
381        recursion). Instead of malloc and free for each heap frame each time a
382        logical recursion happens, frames are retained on a chain and re-used where
383        possible. This sometimes gives as much as 30% improvement.
385    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
386        call.
388    14. As documented, (*COMMIT) is now confined to within a positive assertion.
390    15. It is now possible to link pcretest with libedit as an alternative to
391        libreadline.
393    16. (*COMMIT) control verb is now supported by the JIT compiler.
395    17. The Unicode data tables have been updated to Unicode 6.1.0.
397    18. Added --file-list option to pcregrep.
399    19. Added binary file support to pcregrep, including the -a, --binary-files,
400        -I, and --text options.
402    20. The madvise function is renamed for posix_madvise for QNX compatibility
403        reasons. Fixed by Giuseppe D'Angelo.
405    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
407    22. Changed the option for creating symbolic links for 16-bit man pages from
408        -s to -sf so that re-installing does not cause issues.
410    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
412    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
413        matches in certain environments (the workspace was not being correctly
414        retained). Also added to pcre_dfa_exec() a simple plausibility check on
415        some of the workspace data at the beginning of a restart.
417    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
418        was not doing so when it should - probably a typo introduced by SVN 528
419        (change 8.10/14).
421    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
422        \w+ when the character tables indicated that \x{c4} was a word character.
423        There were several related cases, all because the tests for doing a table
424        lookup were testing for characters less than 127 instead of 255.
426    27. If a pattern contains capturing parentheses that are not used in a match,
427        their slots in the ovector are set to -1. For those that are higher than
428        any matched groups, this happens at the end of processing. In the case when
429        there were back references that the ovector was too small to contain
430        (causing temporary malloc'd memory to be used during matching), and the
431        highest capturing number was not used, memory off the end of the ovector
432        was incorrectly being set to -1. (It was using the size of the temporary
433        memory instead of the true size.)
435    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
436        ovector size, it uses memory at the end of the block that it has got.
438    29. Check for an overlong MARK name and give an error at compile time. The
439        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
441    30. JIT compiler update.
443    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
444        Rill for the patch.
446    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
448    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
450    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
451        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
452        was enabled.
454    35. Improve JIT code generation for greedy plus quantifier.
456    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
457        1 to "aa" instead of to an empty string. The bug affected repeated groups
458        that could potentially match an empty string.
460    37. Optimizing single character iterators in JIT.
462    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
463        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
464        codepoints that are too big for the mode are faulted, and in a UTF mode,
465        disallowed codepoints are also faulted.
467    39. If PCRE was compiled with UTF support, in three places in the DFA
468        matcher there was code that should only have been obeyed in UTF mode, but
469        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
470        processing when bytes with values greater than 127 were present. In 16-bit
471        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
472        both cases the values are those that cannot be the first data item in a UTF
473        character. The three items that might have provoked this were recursions,
474        possessively repeated groups, and atomic groups.
476    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
477        and pcregrep, because some OS require shared objects to be explicitly
478        passed to ld, causing the link step to fail if they are not.
480    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
481        mode, patterns that started with \h* or \R* might be incorrectly matched.
484    Version 8.30 04-February-2012
485    -----------------------------
487    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
488        name is defined in ctype.h.
490    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
491        only in quite long subpatterns.
493    3.  Removed the function pcre_info(), which has been obsolete and deprecated
494        since it was replaced by pcre_fullinfo() in February 2000.
496    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
497        match a (*MARK), and the match failed at the start of the subject, a
498        reference to memory before the start of the subject could occur. This bug
499        was introduced by fix 17 of release 8.21.
501    5.  A reference to an unset group with zero minimum repetition was giving
502        totally wrong answers (in non-JavaScript-compatibility mode). For example,
503        /(another)?(\1?)test/ matched against "hello world test". This bug was
504        introduced in release 8.13.
506    6.  Add support for 16-bit character strings (a large amount of work involving
507        many changes and refactorings).
509    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
510        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
511        from a file.
513    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
514        rounding is not applied in this particular case).
516    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
517        if they appear, or are escaped, in patterns.
519    10. Get rid of a number of -Wunused-but-set-variable warnings.
521    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
522        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
523        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
524        also returns the mark "x". This bug applied to capturing parentheses,
525        non-capturing parentheses, and atomic parentheses. It also applied to some
526        assertions.
528    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
529        information out of configure.ac instead of relying on pcre.h.generic, which
530        is not stored in the repository.
532    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
533        -lreadline.
535    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
537    15. Applied Graycode's patch to put the top-level frame on the stack rather
538        than the heap when not using the stack for recursion. This gives a
539        performance improvement in many cases when recursion is not deep.
541    16. Experimental code added to "pcretest -C" to output the stack frame size.
544    Version 8.21 12-Dec-2011
545    ------------------------
547    1.  Updating the JIT compiler.
549    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
550        are added as well.
552    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
553        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
554        calling _pcre_jit_exec. Some extra comments are added.
556    4.  (*MARK) settings inside atomic groups that do not contain any capturing
557        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
558        was introduced by change 18 for 8.20.
560    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
561        ECMA-262 standard.
563    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
564        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
565        This bug was probably introduced by change 9 of 8.13.
567    7.  While fixing 6 above, I noticed that a number of other items were being
568        incorrectly rejected as "not fixed length". This arose partly because newer
569        opcodes had not been added to the fixed-length checking code. I have (a)
570        corrected the bug and added tests for these items, and (b) arranged for an
571        error to occur if an unknown opcode is encountered while checking for fixed
572        length instead of just assuming "not fixed length". The items that were
573        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
574        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
575        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
577    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
578        being incorrectly compiled and would have given unpredicatble results.
580    9.  A possessively repeated subpattern with minimum repeat count greater than
581        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
582        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
583        the first (A) could occur when it should not.
585    10. Add a cast and remove a redundant test from the code.
587    11. JIT should use pcre_malloc/pcre_free for allocation.
589    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
590        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
591        is anything other than /usr, -L is still shown).
593    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
595    14. Perl does not support \N without a following name in a [] class; PCRE now
596        also gives an error.
598    15. If a forward reference was repeated with an upper limit of around 2000,
599        it caused the error "internal error: overran compiling workspace". The
600        maximum number of forward references (including repeats) was limited by the
601        internal workspace, and dependent on the LINK_SIZE. The code has been
602        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
603        the default depends on LINK_SIZE. There is a new upper limit (for safety)
604        of around 200,000 forward references. While doing this, I also speeded up
605        the filling in of repeated forward references.
607    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
608        incorrectly expecting the subject to contain another "a" after the start.
610    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
611        in the match, the SKIP should be ignored. This was not happening; instead
612        the SKIP was being treated as NOMATCH. For patterns such as
613        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
614        tested.
616    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
617        now much more compatible with Perl, in particular in cases where the result
618        is a non-match for a non-anchored pattern. For example, if
619        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
620        "m", where previously it did not return a name. A side effect of this
621        change is that for partial matches, the last encountered mark name is
622        returned, as for non matches. A number of tests that were previously not
623        Perl-compatible have been moved into the Perl-compatible test files. The
624        refactoring has had the pleasing side effect of removing one argument from
625        the match() function, thus reducing its stack requirements.
627    19. If the /S+ option was used in pcretest to study a pattern using JIT,
628        subsequent uses of /S (without +) incorrectly behaved like /S+.
630    21. Retrieve executable code size support for the JIT compiler and fixing
631        some warnings.
633    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
634        not work when the shorter character appeared right at the end of the
635        subject string.
637    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
638        systems.
640    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
641        output it when the /M option is used in pcretest.
643    25. The CheckMan script was not being included in the distribution. Also, added
644        an explicit "perl" to run Perl scripts from the PrepareRelease script
645        because this is reportedly needed in Windows.
647    26. If study data was being save in a file and studying had not found a set of
648        "starts with" bytes for the pattern, the data written to the file (though
649        never used) was taken from uninitialized memory and so caused valgrind to
650        complain.
652    27. Updated RunTest.bat as provided by Sheri Pierce.
654    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
656    29. Computation of memory usage for the table of capturing group names was
657        giving an unnecessarily large value.
660    Version 8.20 21-Oct-2011
661    ------------------------
663    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
664        a POSIX class. After further experiments with Perl, which convinced me that
665        Perl has bugs and confusions, a closing square bracket is no longer allowed
666        in a POSIX name. This bug also affected patterns with classes that started
667        with full stops.
669    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
670        captured substring, but while checking the failing first alternative,
671        substring 1 is temporarily captured. If the output vector supplied to
672        pcre_exec() was not big enough for this capture, the yield of the function
673        was still zero ("insufficient space for captured substrings"). This cannot
674        be totally fixed without adding another stack variable, which seems a lot
675        of expense for a edge case. However, I have improved the situation in cases
676        such as /(a)(b)x|abc/ matched against "abc", where the return code
677        indicates that fewer than the maximum number of slots in the ovector have
678        been set.
680    3.  Related to (2) above: when there are more back references in a pattern than
681        slots in the output vector, pcre_exec() uses temporary memory during
682        matching, and copies in the captures as far as possible afterwards. It was
683        using the entire output vector, but this conflicts with the specification
684        that only 2/3 is used for passing back captured substrings. Now it uses
685        only the first 2/3, for compatibility. This is, of course, another edge
686        case.
688    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
689        main code base, and can be used by building with --enable-jit. When this is
690        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
691        runtime --no-jit option is given.
693    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
694        ovector, the return from the function was zero, implying that there were
695        other matches that did not fit. The correct "exactly full" value is now
696        returned.
698    6.  If a subpattern that was called recursively or as a subroutine contained
699        (*PRUNE) or any other control that caused it to give a non-standard return,
700        invalid errors such as "Error -26 (nested recursion at the same subject
701        position)" or even infinite loops could occur.
703    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
704        computing the minimum length on reaching *ACCEPT, and so ended up with the
705        wrong value of 1 rather than 0. Further investigation indicates that
706        computing a minimum subject length in the presence of *ACCEPT is difficult
707        (think back references, subroutine calls), and so I have changed the code
708        so that no minimum is registered for a pattern that contains *ACCEPT.
710    8.  If (*THEN) was present in the first (true) branch of a conditional group,
711        it was not handled as intended. [But see 16 below.]
713    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
714        Sheri Pierce.
716    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
717        the first byte in a match must be "a".
719    11. Change 17 for 8.13 increased the recursion depth for patterns like
720        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
721        pattern contains any instances of (*THEN). If it does not, the old
722        optimizations are restored. It would be nice to do this on a per-group
723        basis, but at the moment that is not feasible.
725    12. In some environments, the output of pcretest -C is CRLF terminated. This
726        broke RunTest's code that checks for the link size. A single white space
727        character after the value is now allowed for.
729    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
730        For "fr", it uses the Windows-specific input and output files.
732    14. If (*THEN) appeared in a group that was called recursively or as a
733        subroutine, it did not work as intended. [But see next item.]
735    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
736        pattern fragments (but not containing any | characters). If A and B are
737        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
738        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
739        D. In other words, Perl considers parentheses that do not contain any |
740        characters to be part of a surrounding alternative, whereas PCRE was
741        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
742        differently. PCRE now behaves in the same way as Perl, except in the case
743        of subroutine/recursion calls such as (?1) which have in any case always
744        been different (but PCRE had them first :-).
746    16. Related to 15 above: Perl does not treat the | in a conditional group as
747        creating alternatives. Such a group is treated in the same way as an
748        ordinary group without any | characters when processing (*THEN). PCRE has
749        been changed to match Perl's behaviour.
751    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
752        RunGrepTest script failed.
754    18. Change 22 for version 13 caused atomic groups to use more stack. This is
755        inevitable for groups that contain captures, but it can lead to a lot of
756        stack use in large patterns. The old behaviour has been restored for atomic
757        groups that do not contain any capturing parentheses.
759    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
760        suppress the check for a minimum subject length at run time. (If it was
761        given to pcre_exec() or pcre_dfa_exec() it did work.)
763    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
764        fail to work when decoding hex characters in data strings in EBCDIC
765        environments.
767    21. It appears that in at least one Mac OS environment, the isxdigit() function
768        is implemented as a macro that evaluates to its argument more than once,
769        contravening the C 90 Standard (I haven't checked a later standard). There
770        was an instance in pcretest which caused it to go wrong when processing
771        \x{...} escapes in subject strings. The has been rewritten to avoid using
772        things like p++ in the argument of isxdigit().
775    Version 8.13 16-Aug-2011
776    ------------------------
778    1.  The Unicode data tables have been updated to Unicode 6.0.0.
780    2.  Two minor typos in pcre_internal.h have been fixed.
782    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
783        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
784        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
786    4.  There were a number of related bugs in the code for matching backrefences
787        caselessly in UTF-8 mode when codes for the characters concerned were
788        different numbers of bytes. For example, U+023A and U+2C65 are an upper
789        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
790        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
791        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
792        2-byte code at the end of the subject (it thought there wasn't enough data
793        left).
795    5.  Comprehensive information about what went wrong is now returned by
796        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
797        as the output vector has at least 2 elements. The offset of the start of
798        the failing character and a reason code are placed in the vector.
800    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
801        now returned is for the first byte of the failing character, instead of the
802        last byte inspected. This is an incompatible change, but I hope it is small
803        enough not to be a problem. It makes the returned offset consistent with
804        pcre_exec() and pcre_dfa_exec().
806    7.  pcretest now gives a text phrase as well as the error number when
807        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
808        failure, the offset and reason code are output.
810    8.  When \R was used with a maximizing quantifier it failed to skip backwards
811        over a \r\n pair if the subsequent match failed. Instead, it just skipped
812        back over a single character (\n). This seems wrong (because it treated the
813        two characters as a single entity when going forwards), conflicts with the
814        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
815        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
816        has been changed.
818    9.  Some internal refactoring has changed the processing so that the handling
819        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
820        time (the PCRE_DOTALL option was changed this way some time ago: version
821        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
822        which was always a bit of a fudge. It also means that there is one less
823        argument for the match() function, which reduces its stack requirements
824        slightly. This change also fixes an incompatibility with Perl: the pattern
825        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
827    10. More internal refactoring has drastically reduced the number of recursive
828        calls to match() for possessively repeated groups such as (abc)++ when
829        using pcre_exec().
831    11. While implementing 10, a number of bugs in the handling of groups were
832        discovered and fixed:
834        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
835        (a|)*(?1) gave a compile-time internal error.
836        ((a|)+)+  did not notice that the outer group could match an empty string.
837        (^a|^)+   was not marked as anchored.
838        (.*a|.*)+ was not marked as matching at start or after a newline.
840    12. Yet more internal refactoring has removed another argument from the match()
841        function. Special calls to this function are now indicated by setting a
842        value in a variable in the "match data" data block.
844    13. Be more explicit in pcre_study() instead of relying on "default" for
845        opcodes that mean there is no starting character; this means that when new
846        ones are added and accidentally left out of pcre_study(), testing should
847        pick them up.
849    14. The -s option of pcretest has been documented for ages as being an old
850        synonym of -m (show memory usage). I have changed it to mean "force study
851        for every regex", that is, assume /S for every regex. This is similar to -i
852        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
853        using it. It makes it easier to run collections of tests with and without
854        study enabled, and thereby test pcre_study() more easily. All the standard
855        tests are now run with and without -s (but some patterns can be marked as
856        "never study" - see 20 below).
858    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
859        restoration of the capturing data to the outer values was not happening
860        correctly.
862    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
863        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
864        pattern had matched an empty string, and so incorrectly returned a no
865        match.
867    17. There was optimizing code for the last branch of non-capturing parentheses,
868        and also for the obeyed branch of a conditional subexpression, which used
869        tail recursion to cut down on stack usage. Unfortunately, now that there is
870        the possibility of (*THEN) occurring in these branches, tail recursion is
871        no longer possible because the return has to be checked for (*THEN). These
872        two optimizations have therefore been removed. [But see 8.20/11 above.]
874    18. If a pattern containing \R was studied, it was assumed that \R always
875        matched two bytes, thus causing the minimum subject length to be
876        incorrectly computed because \R can also match just one byte.
878    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
879        was incorrectly computed.
881    20. If /S is present twice on a test pattern in pcretest input, it now
882        *disables* studying, thereby overriding the use of -s on the command line
883        (see 14 above). This is necessary for one or two tests to keep the output
884        identical in both cases.
886    21. When (*ACCEPT) was used in an assertion that matched an empty string and
887        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
889    22. When an atomic group that contained a capturing parenthesis was
890        successfully matched, but the branch in which it appeared failed, the
891        capturing was not being forgotten if a higher numbered group was later
892        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
893        group 1 to "a", when in fact it should be unset. This applied to multi-
894        branched capturing and non-capturing groups, repeated or not, and also to
895        positive assertions (capturing in negative assertions does not happen
896        in PCRE) and also to nested atomic groups.
898    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
899        subject after a captured substring, to make it easier to tell which of a
900        number of identical substrings has been captured.
902    24. The way atomic groups are processed by pcre_exec() has been changed so that
903        if they are repeated, backtracking one repetition now resets captured
904        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
905        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
906        "aaa". Previously, it would have been "a". As part of this code
907        refactoring, the way recursive calls are handled has also been changed.
909    25. If an assertion condition captured any substrings, they were not passed
910        back unless some other capturing happened later. For example, if
911        (?(?=(a))a) was matched against "a", no capturing was returned.
913    26. When studying a pattern that contained subroutine calls or assertions,
914        the code for finding the minimum length of a possible match was handling
915        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
916        group 1 called group 2 while simultaneously a separate group 2 called group
917        1). A stack overflow occurred in this case. I have fixed this by limiting
918        the recursion depth to 10.
920    27. Updated RunTest.bat in the distribution to the version supplied by Tom
921        Fortmann. This supports explicit test numbers on the command line, and has
922        argument validation and error reporting.
924    28. An instance of \X with an unlimited repeat could fail if at any point the
925        first character it looked at was a mark character.
927    29. Some minor code refactoring concerning Unicode properties and scripts
928        should reduce the stack requirement of match() slightly.
930    30. Added the '=' option to pcretest to check the setting of unused capturing
931        slots at the end of the pattern, which are documented as being -1, but are
932        not included in the return count.
934    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
935        compiled something random. Now it gives a compile-time error (as does
936        Perl).
938    32. A *MARK encountered during the processing of a positive assertion is now
939        recorded and passed back (compatible with Perl).
941    33. If --only-matching or --colour was set on a pcregrep call whose pattern
942        had alternative anchored branches, the search for a second match in a line
943        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
944        matched the line "0102" twice. The same bug affected patterns that started
945        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
946        twice.
948    34. Previously, PCRE did not allow quantification of assertions. However, Perl
949        does, and because of capturing effects, quantifying parenthesized
950        assertions may at times be useful. Quantifiers are now allowed for
951        parenthesized assertions.
953    35. A minor code tidy in pcre_compile() when checking options for \R usage.
955    36. \g was being checked for fancy things in a character class, when it should
956        just be a literal "g".
958    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
959        appearance of a nested POSIX class supersedes an apparent external class.
960        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
961        unescaped square brackets may also appear as part of class names. For
962        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
963        more like Perl. (But see 8.20/1 above.)
965    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
966        was because it thought it was \N{name}, which is not supported).
968    39. Add minix to OS list not supporting the -S option in pcretest.
970    40. PCRE tries to detect cases of infinite recursion at compile time, but it
971        cannot analyze patterns in sufficient detail to catch mutual recursions
972        such as ((?1))((?2)). There is now a runtime test that gives an error if a
973        subgroup is called recursively as a subpattern for a second time at the
974        same position in the subject string. In previous releases this might have
975        been caught by the recursion limit, or it might have run out of stack.
977    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
978        happen only once. PCRE was, however incorrectly giving a compile time error
979        "recursive call could loop indefinitely" because it cannot analyze the
980        pattern in sufficient detail. The compile time test no longer happens when
981        PCRE is compiling a conditional subpattern, but actual runaway loops are
982        now caught at runtime (see 40 above).
984    42. It seems that Perl allows any characters other than a closing parenthesis
985        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
986        has been changed to be the same.
988    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
989        as not to get warnings when autogen.sh is called. Also changed
990        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
992    44. To help people who use pcregrep to scan files containing exceedingly long
993        lines, the following changes have been made:
995        (a) The default value of the buffer size parameter has been increased from
996            8K to 20K. (The actual buffer used is three times this size.)
998        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
999            PCRE is built.
1001        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
1002            to be set at run time.
1004        (d) Numerical values in pcregrep options can be followed by K or M, for
1005            example --buffer-size=50K.
1007        (e) If a line being scanned overflows pcregrep's buffer, an error is now
1008            given and the return code is set to 2.
1010    45. Add a pointer to the latest mark to the callout data block.
1012    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
1013        partial match of an empty string instead of no match. This was specific to
1014        the use of ".".
1016    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
1017        complete match instead of a partial match. This bug was dependent on both
1018        the PCRE_UTF8 and PCRE_DOTALL options being set.
1020    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1021        starting byte set, because \b was not being ignored.
1024    Version 8.12 15-Jan-2011
1025    ------------------------
1027    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1028        checks for such things as part of the documentation building process.
1030    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1031        --match-limit and --recursion-limit options (added for 8.11). In
1032        particular, this made one of the standard tests fail. (The integer value
1033        went into the wrong half of a long int.)
1035    3.  If the --colour option was given to pcregrep with -v (invert match), it
1036        did strange things, either producing crazy output, or crashing. It should,
1037        of course, ignore a request for colour when reporting lines that do not
1038        match.
1040    4.  Another pcregrep bug caused similar problems if --colour was specified with
1041        -M (multiline) and the pattern match finished with a line ending.
1043    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1044        matched in multiline mode, the following line was shown as part of the
1045        match. This seems wrong, so I have changed it.
1047    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1048        the check for further matches in the same line (so they could be coloured)
1049        to overrun the end of the current line. If another match was found, it was
1050        incorrectly shown (and then shown again when found in the next line).
1052    7.  If pcregrep was compiled under Windows, there was a reference to the
1053        function pcregrep_exit() before it was defined. I am assuming this was
1054        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1055        reported by a user. I've moved the definition above the reference.
1058    Version 8.11 10-Dec-2010
1059    ------------------------
1061    1.  (*THEN) was not working properly if there were untried alternatives prior
1062        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1063        backtracked to try for "b" instead of moving to the next alternative branch
1064        at the same level (in this case, to look for "c"). The Perl documentation
1065        is clear that when (*THEN) is backtracked onto, it goes to the "next
1066        alternative in the innermost enclosing group".
1068    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1069        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1070        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1071        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1072        (*THEN).
1074    3.  If \s appeared in a character class, it removed the VT character from
1075        the class, even if it had been included by some previous item, for example
1076        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1077        of \s, but is part of the POSIX "space" class.)
1079    4.  A partial match never returns an empty string (because you can always
1080        match an empty string at the end of the subject); however the checking for
1081        an empty string was starting at the "start of match" point. This has been
1082        changed to the "earliest inspected character" point, because the returned
1083        data for a partial match starts at this character. This means that, for
1084        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1085        (previously it gave "no match").
1087    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1088        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1089        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1090        has an implication that the given string is incomplete (because a partial
1091        match is preferred over a full match). For this reason, these items now
1092        give a partial match in this situation. [Aside: previously, the one case
1093        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1094        match rather than a full match, which was wrong by the old rules, but is
1095        now correct.]
1097    6.  There was a bug in the handling of #-introduced comments, recognized when
1098        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1099        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1100        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1101        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1102        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1103        places in pcre_compile().
1105    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1106        comments when looking ahead for named forward references to subpatterns,
1107        the only newline sequence it recognized was NL. It now handles newlines
1108        according to the set newline convention.
1110    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1111        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1112        cater for a lack of strerror(). These oversights have been fixed.
1114    9.  Added --match-limit and --recursion-limit to pcregrep.
1116    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1118    11. When the -o option was used, pcregrep was setting a return code of 1, even
1119        when matches were found, and --line-buffered was not being honoured.
1121    12. Added an optional parentheses number to the -o and --only-matching options
1122        of pcregrep.
1124    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1125        can match an empty string. The code to do it in pcretest and pcredemo
1126        needed fixing:
1128        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1129            only one byte after an empty string match just before CRLF (this case
1130            just got forgotten; "any" and "anycrlf" were OK).
1132        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1133            mode when an empty string match preceded an ASCII character followed by
1134            a non-ASCII character. (The code for advancing by one character rather
1135            than one byte was nonsense.)
1137        (c) The pcredemo.c sample program did not have any code at all to handle
1138            the cases when CRLF is a valid newline sequence.
1140    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1141        as a starting offset was within the subject string. There is now a new
1142        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1143        negative or greater than the length of the string. In order to test this,
1144        pcretest is extended to allow the setting of negative starting offsets.
1146    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1147        starting offset points to the beginning of a UTF-8 character was
1148        unnecessarily clumsy. I tidied it up.
1150    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1151        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1153    17. Nobody had reported that the --include_dir option, which was added in
1154        release 7.7 should have been called --include-dir (hyphen, not underscore)
1155        for compatibility with GNU grep. I have changed it to --include-dir, but
1156        left --include_dir as an undocumented synonym, and the same for
1157        --exclude-dir, though that is not available in GNU grep, at least as of
1158        release 2.5.4.
1160    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1161        characters from a string of bytes) have been redefined so as not to use
1162        loops, in order to improve performance in some environments. At the same
1163        time, I abstracted some of the common code into auxiliary macros to save
1164        repetition (this should not affect the compiled code).
1166    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1167        compile-time error is now given if \c is not followed by an ASCII
1168        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1169        different, and any byte value is allowed.)
1171    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1172        START_OPTIMIZE option, which is now allowed at compile time - but just
1173        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1174        to pcregrep and other applications that have no direct access to PCRE
1175        options. The new /Y option in pcretest sets this option when calling
1176        pcre_compile().
1178    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1179        back references. Groups containing recursive back references were forced to
1180        be atomic by that change, but in the case of named groups, the amount of
1181        memory required was incorrectly computed, leading to "Failed: internal
1182        error: code overflow". This has been fixed.
1184    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1185        pcretest.c, to avoid build problems in some Borland environments.
1188    Version 8.10 25-Jun-2010
1189    ------------------------
1191    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1192        THEN.
1194    2.  (*ACCEPT) was not working when inside an atomic group.
1196    3.  Inside a character class, \B is treated as a literal by default, but
1197        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1198        causes the error). The code is unchanged, but I tidied the documentation.
1200    4.  Inside a character class, PCRE always treated \R and \X as literals,
1201        whereas Perl faults them if its -w option is set. I have changed PCRE so
1202        that it faults them when PCRE_EXTRA is set.
1204    5.  Added support for \N, which always matches any character other than
1205        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1207    6.  When compiling pcregrep with newer versions of gcc which may have
1208        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1209        declared with attribute warn_unused_result" were given. Just casting the
1210        result to (void) does not stop the warnings; a more elaborate fudge is
1211        needed. I've used a macro to implement this.
1213    7.  Minor change to pcretest.c to avoid a compiler warning.
1215    8.  Added four artifical Unicode properties to help with an option to make
1216        \s etc use properties (see next item). The new properties are: Xan
1217        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1219    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1220        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1221        this option. Modified pcretest to add /W to test this facility. Added
1222        REG_UCP to make it available via the POSIX interface.
1224    10. Added --line-buffered to pcregrep.
1226    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1227        studied, and the match started with a letter with a code point greater than
1228        127 whose first byte was different to the first byte of the other case of
1229        the letter, the other case of this starting letter was not recognized
1230        (#976).
1232    12. If a pattern that was studied started with a repeated Unicode property
1233        test, for example, \p{Nd}+, there was the theoretical possibility of
1234        setting up an incorrect bitmap of starting bytes, but fortunately it could
1235        not have actually happened in practice until change 8 above was made (it
1236        added property types that matched character-matching opcodes).
1238    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1239        possible starting bytes for non-anchored patterns.
1241    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1242        \R, and also a number of cases that involve Unicode properties, both
1243        explicit and implicit when PCRE_UCP is set.
1245    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1246        input, it could crash or give wrong results if characters with values
1247        greater than 0xc0 were present in the subject string. (Detail: it assumed
1248        UTF-8 input when processing these items.)
1250    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1251        size_t is 64-bit (#991).
1253    17. Added a check for running out of memory when PCRE is compiled with
1254        --disable-stack-for-recursion (#990).
1256    18. If the last data line in a file for pcretest does not have a newline on
1257        the end, a newline was missing in the output.
1259    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1260        less than 128) in its various bitmaps. However, there is a facility for
1261        generating tables according to the current locale when PCRE is compiled. It
1262        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1263        characters, are recognized by isspace() and therefore were getting set in
1264        these tables, and indeed these tables seem to approximate to ISO 8859. This
1265        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1266        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1267        which of course cannot start UTF-8 characters. I have changed the code so
1268        that only real ASCII characters (less than 128) and the correct starting
1269        bytes for UTF-8 encodings are set for characters greater than 127 when in
1270        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1271        altogether.)
1273    20. Added the /T option to pcretest so as to be able to run tests with non-
1274        standard character tables, thus making it possible to include the tests
1275        used for 19 above in the standard set of tests.
1277    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1278        reference to a subpattern the other side of a comment that contains an
1279        opening parenthesis caused either an internal compiling error, or a
1280        reference to the wrong subpattern.
1283    Version 8.02 19-Mar-2010
1284    ------------------------
1286    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1288    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1289        configured.
1291    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1292        original author of that file, following a query about its status.
1294    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1295        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1297    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1298        quantifier applied to a forward-referencing subroutine call, could compile
1299        incorrect code or give the error "internal error: previously-checked
1300        referenced subpattern not found".
1302    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1303        variables to point to external functions. For these systems, therefore,
1304        pcre_malloc etc. are now initialized to local functions that call the
1305        relevant global functions.
1307    7.  There were two entries missing in the vectors called coptable and poptable
1308        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1309        I've fixed the data, and added a kludgy way of testing at compile time that
1310        the lengths are correct (equal to the number of opcodes).
1312    8.  Following on from 7, I added a similar kludge to check the length of the
1313        eint vector in pcreposix.c.
1315    9.  Error texts for pcre_compile() are held as one long string to avoid too
1316        much relocation at load time. To find a text, the string is searched,
1317        counting zeros. There was no check for running off the end of the string,
1318        which could happen if a new error number was added without updating the
1319        string.
1321    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1323    11. \K was not working if it appeared in an atomic group or in a group that
1324        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1325        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1326        the assertion is positive, but not if it is negative.
1328    12. Change 11 fortuitously reduced the size of the stack frame used in the
1329        "match()" function of pcre_exec.c by one pointer. Forthcoming
1330        implementation of support for (*MARK) will need an extra pointer on the
1331        stack; I have reserved it now, so that the stack frame size does not
1332        decrease.
1334    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1335        item in branch that calls a recursion is a subroutine call - as in the
1336        second branch in the above example - was incorrectly given the compile-
1337        time error "recursive call could loop indefinitely" because pcre_compile()
1338        was not correctly checking the subroutine for matching a non-empty string.
1340    14. The checks for overrunning compiling workspace could trigger after an
1341        overrun had occurred. This is a "should never occur" error, but it can be
1342        triggered by pathological patterns such as hundreds of nested parentheses.
1343        The checks now trigger 100 bytes before the end of the workspace.
1345    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1348    Version 8.01 19-Jan-2010
1349    ------------------------
1351    1.  If a pattern contained a conditional subpattern with only one branch (in
1352        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1353        computed the wrong minimum data length (which is of course zero for such
1354        subpatterns). This could cause incorrect "no match" results.
1356    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1357        the pattern is reset in the first branch, pcre_compile() failed with
1358        "internal error: code overflow at offset...". This happened only when
1359        the reset was to the original external option setting. (An optimization
1360        abstracts leading options settings into an external setting, which was the
1361        cause of this.)
1363    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1364        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1365        assertion pattern did not match (meaning that the assertion was true), it
1366        was incorrectly treated as false if the SKIP had been reached during the
1367        matching. This also applied to assertions used as conditions.
1369    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1370        assertion subpattern, including such a pattern used as a condition,
1371        unpredictable results occurred, instead of the error return
1374    5.  The C++ GlobalReplace function was not working like Perl for the special
1375        situation when an empty string is matched. It now does the fancy magic
1376        stuff that is necessary.
1378    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1379        removed. (These were left over from very, very early versions of PCRE.)
1381    7.  Some cosmetic changes to the code to make life easier when compiling it
1382        as part of something else:
1384        (a) Change DEBUG to PCRE_DEBUG.
1386        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1387            called "current" as "current_branch", to prevent a collision with the
1388            Linux macro when compiled as a kernel module.
1390        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1391            prevent a collision with the Linux macro when compiled as a kernel
1392            module.
1394    8.  In pcre_compile() there are some checks for integer overflows that used to
1395        cast potentially large values to (double). This has been changed to that
1396        when building, a check for int64_t is made, and if it is found, it is used
1397        instead, thus avoiding the use of floating point arithmetic. (There is no
1398        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1399        double.
1401    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1402        2005 (difference between two addresses compared to an unsigned value).
1404    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1405        custom one, because of the following reported problem in Windows:
1407          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1408              under Win32.
1409          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1410              therefore missing the function definition.
1411          - The compiler thus generates a "C" signature for the test function.
1412          - The linker fails to find the "C" function.
1413          - PCRE fails to configure if asked to do so against libbz2.
1415    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1416        messages were output:
1418          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1419          rerunning libtoolize, to keep the correct libtool macros in-tree.
1420          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1422        I have done both of these things.
1424    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1425        most of the time, it *can* run out if it is given a pattern that contains a
1426        runaway infinite recursion. I updated the discussion in the pcrestack man
1427        page.
1429    13. Now that we have gone to the x.xx style of version numbers, the minor
1430        version may start with zero. Using 08 or 09 is a bad idea because users
1431        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1432        interpreted as invalid octal numbers. I've updated the previous comment in
1433        configure.ac, and also added a check that gives an error if 08 or 09 are
1434        used.
1436    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1437        causing partial matching to fail when the end of the subject matched \W
1438        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1440    15. There were some discrepancies between the declarations in pcre_internal.h
1441        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1442        their definitions. The declarations used "const uschar *" and the
1443        definitions used USPTR. Even though USPTR is normally defined as "const
1444        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1445        reported that: "This difference in casting confuses some C++ compilers, for
1446        example, SunCC recognizes above declarations as different functions and
1447        generates broken code for hbpcre." I have changed the declarations to use
1448        USPTR.
1450    16. GNU libtool is named differently on some systems. The autogen.sh script now
1451        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1452        (FreeBSD).
1454    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1455        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1456        comment: "Figure out how to create a longlong from a string: strtoll and
1457        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1458        instance, but it only takes 2 args instead of 3!"
1460    18. A subtle bug concerned with back references has been fixed by a change of
1461        specification, with a corresponding code fix. A pattern such as
1462        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1463        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1464        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1465        same bug. Such groups have to be quantified to be useful, or contained
1466        inside another quantified group. (If there's no repetition, the reference
1467        can never match.) The problem arises because, having left the group and
1468        moved on to the rest of the pattern, a later failure that backtracks into
1469        the group uses the captured value from the final iteration of the group
1470        rather than the correct earlier one. I have fixed this in PCRE by forcing
1471        any group that contains a reference to itself to be an atomic group; that
1472        is, there cannot be any backtracking into it once it has completed. This is
1473        similar to recursive and subroutine calls.
1476    Version 8.00 19-Oct-09
1477    ----------------------
1479    1.  The table for translating pcre_compile() error codes into POSIX error codes
1480        was out-of-date, and there was no check on the pcre_compile() error code
1481        being within the table. This could lead to an OK return being given in
1482        error.
1484    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1485        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1486        in a Windows environment.
1488    3.  The pcregrep --count option prints the count for each file even when it is
1489        zero, as does GNU grep. However, pcregrep was also printing all files when
1490        --files-with-matches was added. Now, when both options are given, it prints
1491        counts only for those files that have at least one match. (GNU grep just
1492        prints the file name in this circumstance, but including the count seems
1493        more useful - otherwise, why use --count?) Also ensured that the
1494        combination -clh just lists non-zero counts, with no names.
1496    4.  The long form of the pcregrep -F option was incorrectly implemented as
1497        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1498        but it seems right to fix it, and I didn't think it was worth preserving
1499        the old behaviour.
1501    5.  The command line items --regex=pattern and --regexp=pattern were not
1502        recognized by pcregrep, which required --regex pattern or --regexp pattern
1503        (with a space rather than an '='). The man page documented the '=' forms,
1504        which are compatible with GNU grep; these now work.
1506    6.  No libpcreposix.pc file was created for pkg-config; there was just
1507        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1509    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1510        when UCP support is not needed, by modifying the Python script that
1511        generates it from Unicode data files. This should not matter if the module
1512        is correctly used as a library, but I received one complaint about 50K of
1513        unwanted data. My guess is that the person linked everything into his
1514        program rather than using a library. Anyway, it does no harm.
1516    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1517        was a minimum greater than 1 for a wide character in a possessive
1518        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1519        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1520        character. Chaos in the form of incorrect output or a compiling loop could
1521        result.
1523    9.  The restrictions on what a pattern can contain when partial matching is
1524        requested for pcre_exec() have been removed. All patterns can now be
1525        partially matched by this function. In addition, if there are at least two
1526        slots in the offset vector, the offset of the earliest inspected character
1527        for the match and the offset of the end of the subject are set in them when
1528        PCRE_ERROR_PARTIAL is returned.
1530    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1531        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1532        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1533        and may be more useful for multi-segment matching.
1535    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1536        used to be given if ever the end of the subject was reached; now it is
1537        given only if matching could not proceed because another character was
1538        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1539        string "Z", which now yields "no match" instead of "partial match". In the
1540        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1541        final character ended with (*FAIL).
1543    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1544        if the pattern had a "must contain" character that was already found in the
1545        earlier partial match, unless partial matching was again requested. For
1546        example, with the pattern /dog.(body)?/, the "must contain" character is
1547        "g". If the first part-match was for the string "dog", restarting with
1548        "sbody" failed. This bug has been fixed.
1550    13. The string returned by pcre_dfa_exec() after a partial match has been
1551        changed so that it starts at the first inspected character rather than the
1552        first character of the match. This makes a difference only if the pattern
1553        starts with a lookbehind assertion or \b or \B (\K is not supported by
1554        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1555        matching functions compatible, and I think it's the right thing to do.
1557    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1558        so that the demonstration program is easily available in environments where
1559        PCRE has not been installed from source.
1561    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1562        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1563        library.
1565    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1566        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1567        is not the first non-POSIX option to be added. Clearly some people find
1568        these options useful.
1570    17. If a caller to the POSIX matching function regexec() passes a non-zero
1571        value for nmatch with a NULL value for pmatch, the value of
1572        nmatch is forced to zero.
1574    18. RunGrepTest did not have a test for the availability of the -u option of
1575        the diff command, as RunTest does. It now checks in the same way as
1576        RunTest, and also checks for the -b option.
1578    19. If an odd number of negated classes containing just a single character
1579        interposed, within parentheses, between a forward reference to a named
1580        subpattern and the definition of the subpattern, compilation crashed with
1581        an internal error, complaining that it could not find the referenced
1582        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1583        [The bug was that it was starting one character too far in when skipping
1584        over the character class, thus treating the ] as data rather than
1585        terminating the class. This meant it could skip too much.]
1587    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1588        /g option in pcretest when the pattern contains \K, which makes it possible
1589        to have an empty string match not at the start, even when the pattern is
1590        anchored. Updated pcretest and pcredemo to use this option.
1592    21. If the maximum number of capturing subpatterns in a recursion was greater
1593        than the maximum at the outer level, the higher number was returned, but
1594        with unset values at the outer level. The correct (outer level) value is
1595        now given.
1597    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1598        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1599        make it do so. The string so far is captured, making this feature
1600        compatible with Perl.
1602    23. The tests have been re-organized, adding tests 11 and 12, to make it
1603        possible to check the Perl 5.10 features against Perl 5.10.
1605    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1606        pattern matches a fixed length string. PCRE did not allow this; now it
1607        does. Neither allows recursion.
1609    25. I finally figured out how to implement a request to provide the minimum
1610        length of subject string that was needed in order to match a given pattern.
1611        (It was back references and recursion that I had previously got hung up
1612        on.) This code has now been added to pcre_study(); it finds a lower bound
1613        to the length of subject needed. It is not necessarily the greatest lower
1614        bound, but using it to avoid searching strings that are too short does give
1615        some useful speed-ups. The value is available to calling programs via
1616        pcre_fullinfo().
1618    26. While implementing 25, I discovered to my embarrassment that pcretest had
1619        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1620        study optimizations had never been tested with that matching function.
1621        Oops. What is worse, even when it was passed study data, there was a bug in
1622        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1623        were also very few tests of studied patterns with pcre_dfa_exec().
1625    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1626        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1627        on the other side of the coin, they are no longer allowed to have different
1628        names, because these cannot be distinguished in PCRE, and this has caused
1629        confusion. (This is a difference from Perl.)
1631    28. When duplicate subpattern names are present (necessarily with different
1632        numbers, as required by 27 above), and a test is made by name in a
1633        conditional pattern, either for a subpattern having been matched, or for
1634        recursion in such a pattern, all the associated numbered subpatterns are
1635        tested, and the overall condition is true if the condition is true for any
1636        one of them. This is the way Perl works, and is also more like the way
1637        testing by number works.
1640    Version 7.9 11-Apr-09
1641    ---------------------
1643    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1644        (pcretest), all targets were linked against these libraries. This included
1645        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1646        libraries. This caused unwanted dependencies to be created. This problem
1647        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1648        pcretest is linked with readline.
1650    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1651        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1652        moved outside it again, because FALSE and TRUE are already defined in AIX,
1653        but BOOL is not.
1655    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1656        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1658    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1659        hyphens) following file names and line numbers when outputting matching
1660        lines. This is not true; no spaces are inserted. I have also clarified the
1661        wording for the --colour (or --color) option.
1663    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1664        was not coloured; this is different to GNU grep, so I have changed it to be
1665        the same.
1667    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1668        each matching line was coloured. Now it goes on to look for further matches
1669        of any of the test patterns, which is the same behaviour as GNU grep.
1671    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1672        doesn't make sense to accept an empty string match in pcregrep, so I have
1673        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1674        seems to be how GNU grep behaves. [But see later change 40 for release
1675        8.33.]
1677    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1678        start or after a newline", because the conditional assertion was not being
1679        correctly handled. The rule now is that both the assertion and what follows
1680        in the first alternative must satisfy the test.
1682    9.  If auto-callout was enabled in a pattern with a conditional group whose
1683        condition was an assertion, PCRE could crash during matching, both with
1684        pcre_exec() and pcre_dfa_exec().
1686    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1687        used for matching.
1689    11. Unicode property support in character classes was not working for
1690        characters (bytes) greater than 127 when not in UTF-8 mode.
1692    12. Added the -M command line option to pcretest.
1694    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1696    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1698    16. Added comments and documentation about mis-use of no_arg in the C++
1699        wrapper.
1701    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1702        from Martin Jerabek that uses macro names for all relevant character and
1703        string constants.
1705    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1706        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1707        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1708        these, but not everybody uses configure.
1710    19. A conditional group that had only one branch was not being correctly
1711        recognized as an item that could match an empty string. This meant that an
1712        enclosing group might also not be so recognized, causing infinite looping
1713        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1714        with the subject "ab", where knowledge that the repeated group can match
1715        nothing is needed in order to break the loop.
1717    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1718        exec(), but without supplying a callout function, matching went wrong.
1720    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1721        leak if the size of the offset vector was greater than 30. When the vector
1722        is smaller, the saved offsets during recursion go onto a local stack
1723        vector, but for larger vectors malloc() is used. It was failing to free
1724        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1725        error, in fact).
1727    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1728        heapframe that is used only when UTF-8 support is enabled. This caused no
1729        problem, but was untidy.
1731    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1732        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1733        included within another project.
1735    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1736        slightly modified by me:
1738          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1739              not building pcregrep.
1741          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1742              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1744    25. Forward references, both numeric and by name, in patterns that made use of
1745        duplicate group numbers, could behave incorrectly or give incorrect errors,
1746        because when scanning forward to find the reference group, PCRE was not
1747        taking into account the duplicate group numbers. A pattern such as
1748        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1750    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1751        the feature of a custom pointer more persuasive (as requested by a user).
1753    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1754        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1755        pre-defined.
1757    28. Added support for (*UTF8) at the start of a pattern.
1759    29. Arrange for flags added by the "release type" setting in CMake to be shown
1760        in the configuration summary.
1763    Version 7.8 05-Sep-08
1764    ---------------------
1766    1.  Replaced UCP searching code with optimized version as implemented for Ad
1767        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1768        stage table and inline lookup instead of a function, giving speed ups of 2
1769        to 5 times on some simple patterns that I tested. Permission was given to
1770        distribute the MultiStage2.py script that generates the tables (it's not in
1771        the tarball, but is in the Subversion repository).
1773    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1774        scripts.
1776    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1777        a group with a zero qualifier. The result of the study could be incorrect,
1778        or the function might crash, depending on the pattern.
1780    4.  Caseless matching was not working for non-ASCII characters in back
1781        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1782        It now works when Unicode Property Support is available.
1784    5.  In pcretest, an escape such as \x{de} in the data was always generating
1785        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1786        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1787        truncation.
1789    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1791    7.  Added two (int) casts to pcregrep when printing the difference of two
1792        pointers, in case they are 64-bit values.
1794    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1795        test 2 if it fails.
1797    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1798        and a #define of that name to empty if it is not externally set. This is to
1799        allow users of MSVC to set it if necessary.
1801    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1802        the convenience functions in the pcre_get.c source file.
1804    11. An option change at the start of a pattern that had top-level alternatives
1805        could cause overwriting and/or a crash. This command provoked a crash in
1806        some environments:
1808          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1810        This potential security problem was recorded as CVE-2008-2371.
1812    12. For a pattern where the match had to start at the beginning or immediately
1813        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1814        pcre_dfa_exec() could read past the end of the passed subject if there was
1815        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1816        pcretest so that it places the subject at the end of its malloc-ed buffer.
1818    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1819        exec() might read past the end of the data buffer in UTF-8 mode.
1821    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1822        the data contained the byte 0x85 as part of a UTF-8 character within its
1823        first line. This applied both to normal and DFA matching.
1825    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1826        /^[^d]*?$/8 failed to match "abc".
1828    16. Added a missing copyright notice to pcrecpp_internal.h.
1830    17. Make it more clear in the documentation that values returned from
1831        pcre_exec() in ovector are byte offsets, not character counts.
1833    18. Tidied a few places to stop certain compilers from issuing warnings.
1835    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1836        supplied by Stefan Weber. I made a further small update for 7.8 because
1837        there is a change of source arrangements: the pcre_searchfuncs.c module is
1838        replaced by pcre_ucd.c.
1841    Version 7.7 07-May-08
1842    ---------------------
1844    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1845        a string to a long long, pretend we don't even have a long long." This is
1846        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1848    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1849        pre-7.6 versions, which defined a global no_arg variable instead of putting
1850        it in the RE class. (See also #8 below.)
1852    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1853        Lopes.
1855    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1857        (1) The include/exclude patterns were being applied to the whole pathnames
1858            of files, instead of just to the final components.
1860        (2) If there was more than one level of directory, the subdirectories were
1861            skipped unless they satisfied the include/exclude conditions. This is
1862            inconsistent with GNU grep (and could even be seen as contrary to the
1863            pcregrep specification - which I improved to make it absolutely clear).
1864            The action now is always to scan all levels of directory, and just
1865            apply the include/exclude patterns to regular files.
1867    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1868        --exclude_dir in the tests to avoid scanning .svn directories.
1870    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1871        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1872        doesn't support NULs in patterns.
1874    7.  Added some missing "const"s to declarations of static tables in
1875        pcre_compile.c and pcre_dfa_exec.c.
1877    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1878        caused by fix #2  above. (Subsequently also a second patch to fix the
1879        first patch. And a third patch - this was a messy problem.)
1881    9.  Applied Craig's patch to remove the use of push_back().
1883    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1884        matching function regexec().
1886    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1887        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1888        references. PCRE supports relative numbers with this syntax (I don't think
1889        Oniguruma does).
1891    12. Previously, a group with a zero repeat such as (...){0} was completely
1892        omitted from the compiled regex. However, this means that if the group
1893        was called as a subroutine from elsewhere in the pattern, things went wrong
1894        (an internal error was given). Such groups are now left in the compiled
1895        pattern, with a new opcode that causes them to be skipped at execution
1896        time.
1898    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1899        to the way PCRE behaves:
1901        (a) A lone ] character is dis-allowed (Perl treats it as data).
1903        (b) A back reference to an unmatched subpattern matches an empty string
1904            (Perl fails the current match path).
1906        (c) A data ] in a character class must be notated as \] because if the
1907            first data character in a class is ], it defines an empty class. (In
1908            Perl it is not possible to have an empty class.) The empty class []
1909            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1910            The negative empty class [^] matches any one character, independently
1911            of the DOTALL setting.
1913    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1914        non-existent subpattern following a character class starting with ']' and
1915        containing () gave an internal compiling error instead of "reference to
1916        non-existent subpattern". Fortunately, when the pattern did exist, the
1917        compiled code was correct. (When scanning forwards to check for the
1918        existence of the subpattern, it was treating the data ']' as terminating
1919        the class, so got the count wrong. When actually compiling, the reference
1920        was subsequently set up correctly.)
1922    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1923        it was being rejected as not supported by pcre_dfa_exec(), even though
1924        other assertions are supported. I have made pcre_dfa_exec() support
1925        (*FAIL).
1927    16. The implementation of 13c above involved the invention of a new opcode,
1928        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1929        cannot be changed at match time, I realized I could make a small
1930        improvement to matching performance by compiling OP_ALLANY instead of
1931        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1932        on the OP_ANY path.
1934    17. Compiling pcretest on Windows with readline support failed without the
1935        following two fixes: (1) Make the unistd.h include conditional on
1936        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1938    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1939        ncurses library to be included for pcretest when ReadLine support is
1940        requested, but also to allow for it to be overridden. This patch came from
1941        Daniel Bergström.
1943    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1944        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1945        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1946        spotting this.
1949    Version 7.6 28-Jan-08
1950    ---------------------
1952    1.  A character class containing a very large number of characters with
1953        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1954        overflow.
1956    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1957        HAVE_LONG_LONG is not defined.
1959    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1960        bring them up to date and include new features. This patch includes:
1962        - Fixed PH's badly added libz and libbz2 support.
1963        - Fixed a problem with static linking.
1964        - Added pcredemo. [But later removed - see 7 below.]
1965        - Fixed dftables problem and added an option.
1966        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1967            HAVE_LONG_LONG.
1968        - Added readline support for pcretest.
1969        - Added an listing of the option settings after cmake has run.
1971    4.  A user submitted a patch to Makefile that makes it easy to create
1972        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1973        Makefile.am that cause it to include this special target, without
1974        affecting anything else. Note that the same mingw target plus all
1975        the other distribution libraries and programs are now supported
1976        when configuring with CMake (see 6 below) instead of with
1977        Configure/Make.
1979    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1980        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1981        exported in the Windows port". It has not yet been confirmed that the patch
1982        solves the problem, but it does no harm.
1984    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1985        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1986        with CMake, and also correct the comment about stack recursion.
1988    7.  Remove the automatic building of pcredemo from the ./configure system and
1989        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1990        of a program that users should build themselves after PCRE is installed, so
1991        building it automatically is not really right. What is more, it gave
1992        trouble in some build environments.
1994    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1997    Version 7.5 10-Jan-08
1998    ---------------------
2000    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
2001        values in parens when parsing an RE using the C++ wrapper."
2003    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
2004        Characters greater than 255 were excluded from the class instead of being
2005        included.
2007    3.  The same bug as (2) above applied to negated POSIX classes such as
2008        [:^space:].
2010    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
2011        defined or documented. It seems to have been a typo for PCRE_STATIC, so
2012        I have changed it.
2014    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
2015        first named subpattern) and a construct such as (?&a) would reference the
2016        first named subpattern whose name started with "a" (in other words, the
2017        length check was missing). Both these problems are fixed. "Subpattern name
2018        expected" is now given for (?&) (a zero-length name), and this patch also
2019        makes it give the same error for \k'' (previously it complained that that
2020        was a reference to a non-existent subpattern).
2022    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
2023        this is right because (?- can be followed by option settings as well as by
2024        digits. I have, however, made the messages clearer.
2026    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
2027        than the number used in the conditional) now cause a compile-time error.
2028        This is actually not compatible with Perl, which accepts such patterns, but
2029        treats the conditional as always being FALSE (as PCRE used to), but it
2030        seems to me that giving a diagnostic is better.
2032    8.  Change "alphameric" to the more common word "alphanumeric" in comments
2033        and messages.
2035    9.  Fix two occurrences of "backslash" in comments that should have been
2036        "backspace".
2038    10. Remove two redundant lines of code that can never be obeyed (their function
2039        was moved elsewhere).
2041    11. The program that makes PCRE's Unicode character property table had a bug
2042        which caused it to generate incorrect table entries for sequences of
2043        characters that have the same character type, but are in different scripts.
2044        It amalgamated them into a single range, with the script of the first of
2045        them. In other words, some characters were in the wrong script. There were
2046        thirteen such cases, affecting characters in the following ranges:
2048          U+002b0 - U+002c1
2049          U+0060c - U+0060d
2050          U+0061e - U+00612
2051          U+0064b - U+0065e
2052          U+0074d - U+0076d
2053          U+01800 - U+01805
2054          U+01d00 - U+01d77
2055          U+01d9b - U+01dbf
2056          U+0200b - U+0200f
2057          U+030fc - U+030fe
2058          U+03260 - U+0327f
2059          U+0fb46 - U+0fbb1
2060          U+10450 - U+1049d
2062    12. The -o option (show only the matching part of a line) for pcregrep was not
2063        compatible with GNU grep in that, if there was more than one match in a
2064        line, it showed only the first of them. It now behaves in the same way as
2065        GNU grep.
2067    13. If the -o and -v options were combined for pcregrep, it printed a blank
2068        line for every non-matching line. GNU grep prints nothing, and pcregrep now
2069        does the same. The return code can be used to tell if there were any
2070        non-matching lines.
2072    14. Added --file-offsets and --line-offsets to pcregrep.
2074    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
2075        infinitely looping recursion. The bug was that positive lookaheads were not
2076        being skipped when checking for a possible empty match (negative lookaheads
2077        and both kinds of lookbehind were skipped).
2079    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
2080        inclusion of <windows.h> to before rather than after the definition of
2081        INVALID_FILE_ATTRIBUTES (patch from David Byron).
2083    17. Specifying a possessive quantifier with a specific limit for a Unicode
2084        character property caused pcre_compile() to compile bad code, which led at
2085        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
2086        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
2087        caused the error; without that there was no problem.
2089    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
2091    19. Added --enable-pcretest-libreadline.
2093    20. In pcrecpp.cc, the variable 'count' was incremented twice in
2094        RE::GlobalReplace(). As a result, the number of replacements returned was
2095        double what it should be. I removed one of the increments, but Craig sent a
2096        later patch that removed the other one (the right fix) and added unit tests
2097        that check the return values (which was not done before).
2099    21. Several CMake things:
2101        (1) Arranged that, when cmake is used on Unix, the libraries end up with
2102            the names libpcre and libpcreposix, not just pcre and pcreposix.
2104        (2) The above change means that pcretest and pcregrep are now correctly
2105            linked with the newly-built libraries, not previously installed ones.
2109    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
2110        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
2111        UTF-8 newline character). The key issue is that the pattern starts .*;
2112        this means that the match must be either at the beginning, or after a
2113        newline. The bug was in the code for advancing after a failed match and
2114        checking that the new position followed a newline. It was not taking
2115        account of UTF-8 characters correctly.
2117    23. PCRE was behaving differently from Perl in the way it recognized POSIX
2118        character classes. PCRE was not treating the sequence [:...:] as a
2119        character class unless the ... were all letters. Perl, however, seems to
2120        allow any characters between [: and :], though of course it rejects as
2121        unknown any "names" that contain non-letters, because all the known class
2122        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
2123        for example, whereas PCRE did not - it did not recognize a POSIX character
2124        class. This seemed a bit dangerous, so the code has been changed to be
2125        closer to Perl. The behaviour is not identical to Perl, because PCRE will
2126        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
2127        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
2128        Perl does, and where it didn't before.
2130    24. Rewrite so as to remove the single use of %n from pcregrep because in some
2131        Windows environments %n is disabled by default.
2134    Version 7.4 21-Sep-07
2135    ---------------------
2137    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
2138        means that a class such as [\s] counted as "explicit reference to CR or
2139        LF". That isn't really right - the whole point of the change was to try to
2140        help when there was an actual mention of one of the two characters. So now
2141        the change happens only if \r or \n (or a literal CR or LF) character is
2142        encountered.
2144    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
2145        of both had grown to the point where there were only 3 bits left.
2146        Fortunately, there was spare space in the data structure, and so I have
2147        moved the internal flags into a new 16-bit field to free up more option
2148        bits.
2150    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
2151        but did not set the internal JCHANGED flag - either of these is enough to
2152        control the way the "get" function works - but the PCRE_INFO_JCHANGED
2153        facility is supposed to tell if (?J) was ever used, so now (?J) at the
2154        start sets both bits.
2156    4.  Added options (at build time, compile time, exec time) to change \R from
2157        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
2159    5.  doc/pcresyntax.html was missing from the distribution.
2161    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
2162        compatibility, even though it is no longer used.
2164    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
2165        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
2166        windows.h file is present (where different names are used). [This was
2167        reversed later after testing - see 16 below.]
2169    8.  Changed all #include <config.h> to #include "config.h". There were also
2170        some further <pcre.h> cases that I changed to "pcre.h".
2172    9.  When pcregrep was used with the --colour option, it missed the line ending
2173        sequence off the lines that it output.
2175    10. It was pointed out to me that arrays of string pointers cause lots of
2176        relocations when a shared library is dynamically loaded. A technique of
2177        using a single long string with a table of offsets can drastically reduce
2178        these. I have refactored PCRE in four places to do this. The result is
2179        dramatic:
2181          Originally:                          290
2182          After changing UCP table:            187
2183          After changing error message table:   43
2184          After changing table of "verbs"       36
2185          After changing table of Posix names   22
2187        Thanks to the folks working on Gregex for glib for this insight.
2189    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
2190        unicode-properties was also set.
2192    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
2194    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
2195        checked only for CRLF.
2197    14. Added casts to pcretest.c to avoid compiler warnings.
2199    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
2201    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
2202        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2203        entirely. This removes changes made in 7 above.
2205    17. The CMake files have been updated, and there is now more information about
2206        building with CMake in the NON-UNIX-USE document.
2209    Version 7.3 28-Aug-07
2210    ---------------------
2212     1. In the rejigging of the build system that eventually resulted in 7.1, the
2213        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
2214        brackets there is not right, since it causes compilers to look for an
2215        installed pcre.h, not the version that is in the source that is being
2216        compiled (which of course may be different). I have changed it back to:
2218          #include "pcre.h"
2220        I have a vague recollection that the change was concerned with compiling in
2221        different directories, but in the new build system, that is taken care of
2222        by the VPATH setting the Makefile.
2224     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
2225        when the subject happened to end in the byte 0x85 (e.g. if the last
2226        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
2227        characters but of course it shouldn't be taken as a newline when it is part
2228        of another character. The bug was that, for an unlimited repeat of . in
2229        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
2230        characters when looking for a newline.
2232     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
2234     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
2235        in debug output.
2237     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
2238        long printing in the pcrecpp unittest when running under MinGW.
2240     6. ESC_K was left out of the EBCDIC table.
2242     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
2243        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
2244        limit also applies to "virtual nesting" when a pattern is recursive, and in
2245        this case 1000 isn't so big. I have been able to remove this limit at the
2246        expense of backing off one optimization in certain circumstances. Normally,
2247        when pcre_exec() would call its internal match() function recursively and
2248        immediately return the result unconditionally, it uses a "tail recursion"
2249        feature to save stack. However, when a subpattern that can match an empty
2250        string has an unlimited repetition quantifier, it no longer makes this
2251        optimization. That gives it a stack frame in which to save the data for
2252        checking that an empty string has been matched. Previously this was taken
2253        from the 1000-entry workspace that had been reserved. So now there is no
2254        explicit limit, but more stack is used.
2256     8. Applied Daniel's patches to solve problems with the import/export magic
2257        syntax that is required for Windows, and which was going wrong for the
2258        pcreposix and pcrecpp parts of the library. These were overlooked when this
2259        problem was solved for the main library.
2261     9. There were some crude static tests to avoid integer overflow when computing
2262        the size of patterns that contain repeated groups with explicit upper
2263        limits. As the maximum quantifier is 65535, the maximum group length was
2264        set at 30,000 so that the product of these two numbers did not overflow a
2265        32-bit integer. However, it turns out that people want to use groups that
2266        are longer than 30,000 bytes (though not repeat them that many times).
2267        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
2268        made it possible to implement the integer overflow checks in a much more
2269        dynamic way, which I have now done. The artificial limitation on group
2270        length has been removed - we now have only the limit on the total length of
2271        the compiled pattern, which depends on the LINK_SIZE setting.
2273    10. Fixed a bug in the documentation for get/copy named substring when
2274        duplicate names are permitted. If none of the named substrings are set, the
2275        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
2276        empty string.
2278    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
2279        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
2280        because the ] is interpreted as the first data character and the
2281        terminating ] is not found. PCRE has been made compatible with Perl in this
2282        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
2283        cause memory overwriting.
2285    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
2286        string has been matched (to stop an infinite loop). It was not recognizing
2287        a conditional subpattern that could match an empty string if that
2288        subpattern was within another subpattern. For example, it looped when
2289        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
2290        condition was not nested. This bug has been fixed.
2292    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
2293        past the start of the subject in the presence of bytes with the top bit
2294        set, for example "\x8aBCD".
2296    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
2297        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
2299    14. Optimized (?!) to (*FAIL).
2301    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
2302        This restricts code points to be within the range 0 to 0x10FFFF, excluding
2303        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
2304        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
2305        does: it's just the validity check that is more restrictive.
2307    16. Inserted checks for integer overflows during escape sequence (backslash)
2308        processing, and also fixed erroneous offset values for syntax errors during
2309        backslash processing.
2311    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
2312        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
2314    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
2315        caused an overrun.
2317    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
2318        something other than just ASCII characters) inside a group that had an
2319        unlimited repeat caused a loop at compile time (while checking to see
2320        whether the group could match an empty string).
2322    20. Debugging a pattern containing \p or \P could cause a crash. For example,
2323        [\P{Any}] did so. (Error in the code for printing property names.)
2325    21. An orphan \E inside a character class could cause a crash.
2327    22. A repeated capturing bracket such as (A)? could cause a wild memory
2328        reference during compilation.
2330    23. There are several functions in pcre_compile() that scan along a compiled
2331        expression for various reasons (e.g. to see if it's fixed length for look
2332        behind). There were bugs in these functions when a repeated \p or \P was
2333        present in the pattern. These operators have additional parameters compared
2334        with \d, etc, and these were not being taken into account when moving along
2335        the compiled data. Specifically:
2337        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2338            length.
2340        (b) An item such as \pL+ within a repeated group could cause crashes or
2341            loops.
2343        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2344            "reference to non-existent subpattern" error.
2346        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2348    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2349        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2351    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2353    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2354        character were causing crashes (broken optimization).
2356    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2357        \p or \P) caused a compile-time loop.
2359    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2360        break. For example, the unstudied pattern [\r\n]A does not match the string
2361        "\r\nA" because change 7.0/46 below moves the current point on by two
2362        characters after failing to match at the start. However, the pattern \nA
2363        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2364        the same is true. There doesn't seem any very clean way out of this, but
2365        what I have chosen to do makes the common cases work: PCRE now takes note
2366        of whether there can be an explicit match for \r or \n anywhere in the
2367        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2368        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2369        pattern has explicit CR or LF references.
2371    29. Added (*CR) etc for changing newline setting at start of pattern.
2374    Version 7.2 19-Jun-07
2375    ---------------------
2377     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
2378        which is apparently normally available under Windows.
2380     2. Re-jig the pcregrep tests with different newline settings in an attempt
2381        to make them independent of the local environment's newline setting.
2383     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
2385     4. Some of the "internals" tests were previously cut out when the link size
2386        was not 2, because the output contained actual offsets. The recent new
2387        "Z" feature of pcretest means that these can be cut out, making the tests
2388        usable with all link sizes.
2390     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
2391        stack recursion. This gives a massive performance boost under BSD, but just
2392        a small improvement under Linux. However, it saves one field in the frame
2393        in all cases.
2395     6. Added more features from the forthcoming Perl 5.10:
2397        (a) (?-n) (where n is a string of digits) is a relative subroutine or
2398            recursion call. It refers to the nth most recently opened parentheses.
2400        (b) (?+n) is also a relative subroutine call; it refers to the nth next
2401            to be opened parentheses.
2403        (c) Conditions that refer to capturing parentheses can be specified
2404            relatively, for example, (?(-2)... or (?(+3)...
2406        (d) \K resets the start of the current match so that everything before
2407            is not part of it.
2409        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2411        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2412            reference syntax.
2414        (g) (?| introduces a group in which the numbering of parentheses in each
2415            alternative starts with the same number.
2417        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2419     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2422     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2423        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2424        for detecting groups that can match an empty string.
2426     9. A pattern with a very large number of alternatives (more than several
2427        hundred) was running out of internal workspace during the pre-compile
2428        phase, where pcre_compile() figures out how much memory will be needed. A
2429        bit of new cunning has reduced the workspace needed for groups with
2430        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2431        workspace instead of running out of the 4096 that are available.
2433    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2435    11. Applied patch from Google to remove an optimization that didn't quite work.
2436        The report of the bug said:
2438          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2439          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2440          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2442    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2443        it matched the wrong number of bytes.
2446    Version 7.1 24-Apr-07
2447    ---------------------
2449     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
2450        that is more "standard", making use of automake and other Autotools. There
2451        is some re-arrangement of the files and adjustment of comments consequent
2452        on this.
2454     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
2455        for recursive directory scanning broke on some systems because the files
2456        are not scanned in any specific order and on different systems the order
2457        was different. A call to "sort" has been inserted into RunGrepTest for the
2458        approprate test as a short-term fix. In the longer term there may be an
2459        alternative.
2461     3. I had an email from Eric Raymond about problems translating some of PCRE's
2462        man pages to HTML (despite the fact that I distribute HTML pages, some
2463        people do their own conversions for various reasons). The problems
2464        concerned the use of low-level troff macros .br and .in. I have therefore
2465        removed all such uses from the man pages (some were redundant, some could
2466        be replaced by .nf/.fi pairs). The 132html script that I use to generate
2467        HTML has been updated to handle .nf/.fi and to complain if it encounters
2468        .br or .in.
2470     4. Updated comments in configure.ac that get placed in config.h.in and also
2471        arranged for config.h to be included in the distribution, with the name
2472        config.h.generic, for the benefit of those who have to compile without
2473        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
2475     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
2476        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
2477        makevp.bat for the current PCRE, using the additional files
2478        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
2480     6. A Windows user reported a minor discrepancy with test 2, which turned out
2481        to be caused by a trailing space on an input line that had got lost in his
2482        copy. The trailing space was an accident, so I've just removed it.
2484     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
2485        that is needed.
2487     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
2488        as "const" (a) because they are and (b) because it helps the PHP
2489        maintainers who have recently made a script to detect big data structures
2490        in the php code that should be moved to the .rodata section. I remembered
2491        to update Builducptable as well, so it won't revert if ucptable.h is ever
2492        re-created.
2494     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
2495        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
2496        order to be able to cut out the UTF-8 tables in the latter when UTF-8
2497        support is not required. This saves 1.5-2K of code, which is important in
2498        some applications.
2500        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
2501        so as not to refer to the tables, even though these functions will never be
2502        called when UTF-8 support is disabled. Otherwise there are problems with a
2503        shared library.
2505    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
2507        (a) It was defining its arguments as char * instead of void *.
2509        (b) It was assuming that all moves were upwards in memory; this was true
2510            a long time ago when I wrote it, but is no longer the case.
2512        The emulated memove() is provided for those environments that have neither
2513        memmove() nor bcopy(). I didn't think anyone used it these days, but that
2514        is clearly not the case, as these two bugs were recently reported.
2516    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
2517        and Detrail to create the HTML documentation, the .txt form of the man
2518        pages, and it removes trailing spaces from listed files. It also creates
2519        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
2520        case, it wraps all the #defines with #ifndefs. This script should be run
2521        before "make dist".
2523    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
2524        with Unicode property support.
2526        (a) For a maximizing quantifier, if the two different cases of the
2527            character were of different lengths in their UTF-8 codings (there are
2528            some cases like this - I found 11), and the matching function had to
2529            back up over a mixture of the two cases, it incorrectly assumed they
2530            were both the same length.
2532        (b) When PCRE was configured to use the heap rather than the stack for
2533            recursion during matching, it was not correctly preserving the data for
2534            the other case of a UTF-8 character when checking ahead for a match
2535            while processing a minimizing repeat. If the check also involved
2536            matching a wide character, but failed, corruption could cause an
2537            erroneous result when trying to check for a repeat of the original
2538            character.
2540    13. Some tidying changes to the testing mechanism:
2542        (a) The RunTest script now detects the internal link size and whether there
2543            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
2544            values substituted by "configure". (The RunGrepTest script already did
2545            this for UTF-8.) The configure.ac script no longer substitutes the
2546            relevant variables.
2548        (b) The debugging options /B and /D in pcretest show the compiled bytecode
2549            with length and offset values. This means that the output is different
2550            for different internal link sizes. Test 2 is skipped for link sizes
2551            other than 2 because of this, bypassing the problem. Unfortunately,
2552            there was also a test in test 3 (the locale tests) that used /B and
2553            failed for link sizes other than 2. Rather than cut the whole test out,
2554            I have added a new /Z option to pcretest that replaces the length and
2555            offset values with spaces. This is now used to make test 3 independent
2556            of link size. (Test 2 will be tidied up later.)
2558    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
2559        segmentation fault instead of returning the appropriate error message.
2561    15. In multiline mode when the newline sequence was set to "any", the pattern
2562        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
2563        This doesn't seem right; it now treats the CRLF combination as the line
2564        ending, and so does not match in that case. It's only a pattern such as ^$
2565        that would hit this one: something like ^ABC$ would have failed after \r
2566        and then tried again after \r\n.
2568    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
2569        in an attempt to make files that differ only in their line terminators
2570        compare equal. This works on Linux.
2572    17. Under certain error circumstances pcregrep might try to free random memory
2573        as it exited. This is now fixed, thanks to valgrind.
2575    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
2576        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
2577        was because its rules for how to advance for /g after matching an empty
2578        string at the end of a line did not allow for this case. They now check for
2579        it specially.
2581    20. pcretest is supposed to handle patterns and data of any length, by
2582        extending its buffers when necessary. It was getting this wrong when the
2583        buffer for a data line had to be extended.
2585    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
2586        CRLF as a newline sequence.
2588    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
2589        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
2590        I have nevertheless tidied it up.
2592    23. Added some casts to kill warnings from HP-UX ia64 compiler.
2594    24. Added a man page for pcre-config.
2597    Version 7.0 19-Dec-06
2598    ---------------------
2600     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
2601        moving to gcc 4.1.1.
2603     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
2604        sys/time.h, which is documented as needed for this function. It doesn't
2605        seem to matter on Linux, but it showed up on some releases of OS X.
2607     3. It seems that there are systems where bytes whose values are greater than
2608        127 match isprint() in the "C" locale. The "C" locale should be the
2609        default when a C program starts up. In most systems, only ASCII printing
2610        characters match isprint(). This difference caused the output from pcretest
2611        to vary, making some of the tests fail. I have changed pcretest so that:
2613        (a) When it is outputting text in the compiled version of a pattern, bytes
2614            other than 32-126 are always shown as hex escapes.
2616        (b) When it is outputting text that is a matched part of a subject string,
2617            it does the same, unless a different locale has been set for the match
2618            (using the /L modifier). In this case, it uses isprint() to decide.
2620     4. Fixed a major bug that caused incorrect computation of the amount of memory
2621        required for a compiled pattern when options that changed within the
2622        pattern affected the logic of the preliminary scan that determines the
2623        length. The relevant options are -x, and -i in UTF-8 mode. The result was
2624        that the computed length was too small. The symptoms of this bug were
2625        either the PCRE error "internal error: code overflow" from pcre_compile(),
2626        or a glibc crash with a message such as "pcretest: free(): invalid next
2627        size (fast)". Examples of patterns that provoked this bug (shown in
2628        pcretest format) are:
2630          /(?-x: )/x
2631          /(?x)(?-x: \s*#\s*)/
2632          /((?i)[\x{c0}])/8
2633          /(?i:[\x{c0}])/8
2635        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
2636        is now done differently.
2638     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
2639        wrapper classes; (b) implement a new function in the C++ scanner that is
2640        more efficient than the old way of doing things because it avoids levels of
2641        recursion in the regex matching; (c) add a paragraph to the documentation
2642        for the FullMatch() function.
2644     6. The escape sequence \n was being treated as whatever was defined as
2645        "newline". Not only was this contrary to the documentation, which states
2646        that \n is character 10 (hex 0A), but it also went horribly wrong when
2647        "newline" was defined as CRLF. This has been fixed.
2649     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
2650        was being set to -1 for the "end of line" case (supposedly a value that no
2651        character can have). Though this value is never used (the check for end of
2652        line is "zero bytes in current character"), it caused compiler complaints.
2653        I've changed it to 0xffffffff.
2655     8. In pcre_version.c, the version string was being built by a sequence of
2656        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
2657        string (as it is for production releases) called a macro with an empty
2658        argument. The C standard says the result of this is undefined. The gcc
2659        compiler treats it as an empty string (which was what was wanted) but it is
2660        reported that Visual C gives an error. The source has been hacked around to
2661        avoid this problem.
2663     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
2664        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
2665        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
2666        of them did).
2668    10. Originally, pcretest opened its input and output without "b"; then I was
2669        told that "b" was needed in some environments, so it was added for release
2670        5.0 to both the input and output. (It makes no difference on Unix-like
2671        systems.) Later I was told that it is wrong for the input on Windows. I've
2672        now abstracted the modes into two macros, to make it easier to fiddle with
2673        them, and removed "b" from the input mode under Windows.
2675    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
2677    12. Added -help and --help to pcretest as an official way of being reminded
2678        of the options.
2680    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
2681        and pcrecpp.cc because they annoy compilers at high warning levels.
2683    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
2685    15. Fixed an occurrence of == in configure.ac that should have been = (shell
2686        scripts are not C programs :-) and which was not noticed because it works
2687        on Linux.
2689    16. pcretest is supposed to handle any length of pattern and data line (as one
2690        line or as a continued sequence of lines) by extending its input buffer if
2691        necessary. This feature was broken for very long pattern lines, leading to
2692        a string of junk being passed to pcre_compile() if the pattern was longer
2693        than about 50K.
2695    17. I have done a major re-factoring of the way pcre_compile() computes the
2696        amount of memory needed for a compiled pattern. Previously, there was code
2697        that made a preliminary scan of the pattern in order to do this. That was
2698        OK when PCRE was new, but as the facilities have expanded, it has become
2699        harder and harder to keep it in step with the real compile phase, and there
2700        have been a number of bugs (see for example, 4 above). I have now found a
2701        cunning way of running the real compile function in a "fake" mode that
2702        enables it to compute how much memory it would need, while actually only
2703        ever using a few hundred bytes of working memory and without too many
2704        tests of the mode. This should make future maintenance and development
2705        easier. A side effect of this work is that the limit of 200 on the nesting
2706        depth of parentheses has been removed (though this was never a serious
2707        limitation, I suspect). However, there is a downside: pcre_compile() now
2708        runs more slowly than before (30% or more, depending on the pattern). I
2709        hope this isn't a big issue. There is no effect on runtime performance.
2711    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
2712        newline (only possible for the last line of a file) and it was a
2713        pattern that set a locale (followed by /Lsomething), pcretest crashed.
2715    19. Added additional timing features to pcretest. (1) The -tm option now times
2716        matching only, not compiling. (2) Both -t and -tm can be followed, as a
2717        separate command line item, by a number that specifies the number of
2718        repeats to use when timing. The default is 50000; this gives better
2719        precision, but takes uncomfortably long for very large patterns.
2721    20. Extended pcre_study() to be more clever in cases where a branch of a
2722        subpattern has no definite first character. For example, (a*|b*)[cd] would
2723        previously give no result from pcre_study(). Now it recognizes that the
2724        first character must be a, b, c, or d.
2726    21. There was an incorrect error "recursive call could loop indefinitely" if
2727        a subpattern (or the entire pattern) that was being tested for matching an
2728        empty string contained only one non-empty item after a nested subpattern.
2729        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
2730        incorrectly, because the \d was being skipped in the check.
2732    22. The pcretest program now has a new pattern option /B and a command line
2733        option -b, which is equivalent to adding /B to every pattern. This causes
2734        it to show the compiled bytecode, without the additional information that
2735        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
2736        is the same as /B/I).
2738    23. A new optimization is now able automatically to treat some sequences such
2739        as a*b as a*+b. More specifically, if something simple (such as a character
2740        or a simple class like \d) has an unlimited quantifier, and is followed by
2741        something that cannot possibly match the quantified thing, the quantifier
2742        is automatically "possessified".
2744    24. A recursive reference to a subpattern whose number was greater than 39
2745        went wrong under certain circumstances in UTF-8 mode. This bug could also
2746        have affected the operation of pcre_study().
2748    25. Realized that a little bit of performance could be had by replacing
2749        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
2751    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
2753    27. Possessive quantifiers such as a++ were previously implemented by turning
2754        them into atomic groups such as ($>a+). Now they have their own opcodes,
2755        which improves performance. This includes the automatically created ones
2756        from 23 above.
2758    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
2759        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
2760        the first matched character to be a colon. This applied both to named and
2761        numbered groups.
2763    29. The ucpinternal.h header file was missing its idempotency #ifdef.
2765    30. I was sent a "project" file called libpcre.a.dev which I understand makes
2766        building PCRE on Windows easier, so I have included it in the distribution.
2768    31. There is now a check in pcretest against a ridiculously large number being
2769        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
2770        loop, the loop is abandoned.
2772    32. Forward references to subpatterns in conditions such as (?(2)...) where
2773        subpattern 2 is defined later cause pcre_compile() to search forwards in
2774        the pattern for the relevant set of parentheses. This search went wrong
2775        when there were unescaped parentheses in a character class, parentheses
2776        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2778    33. "Subroutine" calls and backreferences were previously restricted to
2779        referencing subpatterns earlier in the regex. This restriction has now
2780        been removed.
2782    34. Added a number of extra features that are going to be in Perl 5.10. On the
2783        whole, these are just syntactic alternatives for features that PCRE had
2784        previously implemented using the Python syntax or my own invention. The
2785        other formats are all retained for compatibility.
2787        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2788            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2789            also .NET compatible.
2791        (b) A recursion or subroutine call to a named group can now be defined as
2792            (?&name) as well as (?P>name).
2794        (c) A backreference to a named group can now be defined as \k<name> or
2795            \k'name' as well as (?P=name). The new forms, as well as being in Perl
2796            5.10, are also .NET compatible.
2798        (d) A conditional reference to a named group can now use the syntax
2799            (?(<name>) or (?('name') as well as (?(name).
2801        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2802            groups (named and numbered) that are never evaluated inline, but can be
2803            called as "subroutines" from elsewhere. In effect, the DEFINE condition
2804            is always false. There may be only one alternative in such a group.
2806        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2807            as the simple (?(R). The condition is true only if the most recent
2808            recursion is that of the given number or name. It does not search out
2809            through the entire recursion stack.
2811        (g) The escape \gN or \g{N} has been added, where N is a positive or
2812            negative number, specifying an absolute or relative reference.
2814    35. Tidied to get rid of some further signed/unsigned compiler warnings and
2815        some "unreachable code" warnings.
2817    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2818        things, this adds five new scripts.
2820    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2821        There were also incompatibilities regarding the handling of \Q..\E inside
2822        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2823        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2825    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2826        matches an empty string, and forcibly breaks the loop. There were bugs in
2827        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2828        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2829        separate and independent bugs (that affected different cases) have been
2830        fixed.
2832    39. Refactored the code to abolish the use of different opcodes for small
2833        capturing bracket numbers. This is a tidy that I avoided doing when I
2834        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2835        The new approach is not only tidier, it makes it possible to reduce the
2836        memory needed to fix the previous bug (38).
2838    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2839        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2840        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2841        mode.
2843    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2844        report.
2846    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2847        copy construction and assignment in the C++ wrapper.
2849    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2850        couple of bugs that could have given wrong results in the "--newline=crlf"
2851        case.
2853    44. Added a number of casts and did some reorganization of signed/unsigned int
2854        variables following suggestions from Dair Grant. Also renamed the variable
2855        "this" as "item" because it is a C++ keyword.
2857    45. Arranged for dftables to add
2859          #include "pcre_internal.h"
2861        to pcre_chartables.c because without it, gcc 4.x may remove the array
2862        definition from the final binary if PCRE is built into a static library and
2863        dead code stripping is activated.
2865    46. For an unanchored pattern, if a match attempt fails at the start of a
2866        newline sequence, and the newline setting is CRLF or ANY, and the next two
2867        characters are CRLF, advance by two characters instead of one.
2870  Version 6.7 04-Jul-06  Version 6.7 04-Jul-06
2871  ---------------------  ---------------------

Removed from v.91  
changed lines
  Added in v.1328

  ViewVC Help
Powered by ViewVC 1.1.5