ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 15 by nigel, Sat Feb 24 21:38:25 2007 UTC revision 979 by ph10, Sun Jun 17 19:08:41 2012 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.31 02-June-2012
5    -------------------------
7    1.  Fixing a wrong JIT test case and some compiler warnings.
9    2.  Removed a bashism from the RunTest script.
11    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
12        to unsigned type, result still unsigned" that was given by an MS compiler
13        on encountering the code "-sizeof(xxx)".
15    4.  Partial matching support is added to the JIT compiler.
17    5.  Fixed several bugs concerned with partial matching of items that consist
18        of more than one character:
20        (a) /^(..)\1/ did not partially match "aba" because checking references was
21            done on an "all or nothing" basis. This also applied to repeated
22            references.
24        (b) \R did not give a hard partial match if \r was found at the end of the
25            subject.
27        (c) \X did not give a hard partial match after matching one or more
28            characters at the end of the subject.
30        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
31            a partial match for the string "\r".
33        (e) When newline was set to CRLF, the metacharacter "." did not recognize
34            a partial match for a CR character at the end of the subject string.
36    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
37        running pcretest, the text "(JIT)" added to the output whenever JIT is
38        actually used to run the match.
40    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
41        or /S+[+] with a digit between 1 and 7.
43    8.  OP_NOT now supports any UTF character not just single-byte ones.
45    9.  (*MARK) control verb is now supported by the JIT compiler.
47    10. The command "./RunTest list" lists the available tests without actually
48        running any of them. (Because I keep forgetting what they all are.)
52    12. Applied a (slightly modified) user-supplied patch that improves performance
53        when the heap is used for recursion (compiled with --disable-stack-for-
54        recursion). Instead of malloc and free for each heap frame each time a
55        logical recursion happens, frames are retained on a chain and re-used where
56        possible. This sometimes gives as much as 30% improvement.
58    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
59        call.
61    14. As documented, (*COMMIT) is now confined to within a positive assertion.
63    15. It is now possible to link pcretest with libedit as an alternative to
64        libreadline.
66    16. (*COMMIT) control verb is now supported by the JIT compiler.
68    17. The Unicode data tables have been updated to Unicode 6.1.0.
70    18. Added --file-list option to pcregrep.
72    19. Added binary file support to pcregrep, including the -a, --binary-files,
73        -I, and --text options.
75    20. The madvise function is renamed for posix_madvise for QNX compatibility
76        reasons. Fixed by Giuseppe D'Angelo.
78    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
80    22. Changed the option for creating symbolic links for 16-bit man pages from
81        -s to -sf so that re-installing does not cause issues.
83    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
85    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
86        matches in certain environments (the workspace was not being correctly
87        retained). Also added to pcre_dfa_exec() a simple plausibility check on
88        some of the workspace data at the beginning of a restart.
90    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
91        was not doing so when it should - probably a typo introduced by SVN 528
92        (change 8.10/14).
94    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
95        \w+ when the character tables indicated that \x{c4} was a word character.
96        There were several related cases, all because the tests for doing a table
97        lookup were testing for characters less than 127 instead of 255.
99    27. If a pattern contains capturing parentheses that are not used in a match,
100        their slots in the ovector are set to -1. For those that are higher than
101        any matched groups, this happens at the end of processing. In the case when
102        there were back references that the ovector was too small to contain
103        (causing temporary malloc'd memory to be used during matching), and the
104        highest capturing number was not used, memory off the end of the ovector
105        was incorrectly being set to -1. (It was using the size of the temporary
106        memory instead of the true size.)
108    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
109        ovector size, it uses memory at the end of the block that it has got.
111    29. Check for an overlong MARK name and give an error at compile time. The
112        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
114    30. JIT compiler update.
116    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
117        Rill for the patch.
119    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
121    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
123    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
124        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
125        was enabled.
127    35. Improve JIT code generation for greedy plus quantifier.
129    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
130        1 to "aa" instead of to an empty string. The bug affected repeated groups
131        that could potentially match an empty string.
133    37. Optimizing single character iterators in JIT.
135    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
136        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
137        codepoints that are too big for the mode are faulted, and in a UTF mode,
138        disallowed codepoints are also faulted.
140    39. If PCRE was compiled with UTF support, in three places in the DFA
141        matcher there was code that should only have been obeyed in UTF mode, but
142        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
143        processing when bytes with values greater than 127 were present. In 16-bit
144        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
145        both cases the values are those that cannot be the first data item in a UTF
146        character. The three items that might have provoked this were recursions,
147        possessively repeated groups, and atomic groups.
150    Version 8.30 04-February-2012
151    -----------------------------
153    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
154        name is defined in ctype.h.
156    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
157        only in quite long subpatterns.
159    3.  Removed the function pcre_info(), which has been obsolete and deprecated
160        since it was replaced by pcre_fullinfo() in February 2000.
162    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
163        match a (*MARK), and the match failed at the start of the subject, a
164        reference to memory before the start of the subject could occur. This bug
165        was introduced by fix 17 of release 8.21.
167    5.  A reference to an unset group with zero minimum repetition was giving
168        totally wrong answers (in non-JavaScript-compatibility mode). For example,
169        /(another)?(\1?)test/ matched against "hello world test". This bug was
170        introduced in release 8.13.
172    6.  Add support for 16-bit character strings (a large amount of work involving
173        many changes and refactorings).
175    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
176        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
177        from a file.
179    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
180        rounding is not applied in this particular case).
182    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
183        if they appear, or are escaped, in patterns.
185    10. Get rid of a number of -Wunused-but-set-variable warnings.
187    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
188        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
189        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
190        also returns the mark "x". This bug applied to capturing parentheses,
191        non-capturing parentheses, and atomic parentheses. It also applied to some
192        assertions.
194    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
195        information out of configure.ac instead of relying on pcre.h.generic, which
196        is not stored in the repository.
198    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
199        -lreadline.
201    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
203    15. Applied Graycode's patch to put the top-level frame on the stack rather
204        than the heap when not using the stack for recursion. This gives a
205        performance improvement in many cases when recursion is not deep.
207    16. Experimental code added to "pcretest -C" to output the stack frame size.
210    Version 8.21 12-Dec-2011
211    ------------------------
213    1.  Updating the JIT compiler.
215    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
216        are added as well.
218    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
219        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
220        calling _pcre_jit_exec. Some extra comments are added.
222    4.  (*MARK) settings inside atomic groups that do not contain any capturing
223        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
224        was introduced by change 18 for 8.20.
226    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
227        ECMA-262 standard.
229    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
230        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
231        This bug was probably introduced by change 9 of 8.13.
233    7.  While fixing 6 above, I noticed that a number of other items were being
234        incorrectly rejected as "not fixed length". This arose partly because newer
235        opcodes had not been added to the fixed-length checking code. I have (a)
236        corrected the bug and added tests for these items, and (b) arranged for an
237        error to occur if an unknown opcode is encountered while checking for fixed
238        length instead of just assuming "not fixed length". The items that were
239        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
240        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
241        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
243    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
244        being incorrectly compiled and would have given unpredicatble results.
246    9.  A possessively repeated subpattern with minimum repeat count greater than
247        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
248        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
249        the first (A) could occur when it should not.
251    10. Add a cast and remove a redundant test from the code.
253    11. JIT should use pcre_malloc/pcre_free for allocation.
255    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
256        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
257        is anything other than /usr, -L is still shown).
259    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
261    14. Perl does not support \N without a following name in a [] class; PCRE now
262        also gives an error.
264    15. If a forward reference was repeated with an upper limit of around 2000,
265        it caused the error "internal error: overran compiling workspace". The
266        maximum number of forward references (including repeats) was limited by the
267        internal workspace, and dependent on the LINK_SIZE. The code has been
268        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
269        the default depends on LINK_SIZE. There is a new upper limit (for safety)
270        of around 200,000 forward references. While doing this, I also speeded up
271        the filling in of repeated forward references.
273    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
274        incorrectly expecting the subject to contain another "a" after the start.
276    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
277        in the match, the SKIP should be ignored. This was not happening; instead
278        the SKIP was being treated as NOMATCH. For patterns such as
279        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
280        tested.
282    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
283        now much more compatible with Perl, in particular in cases where the result
284        is a non-match for a non-anchored pattern. For example, if
285        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
286        "m", where previously it did not return a name. A side effect of this
287        change is that for partial matches, the last encountered mark name is
288        returned, as for non matches. A number of tests that were previously not
289        Perl-compatible have been moved into the Perl-compatible test files. The
290        refactoring has had the pleasing side effect of removing one argument from
291        the match() function, thus reducing its stack requirements.
293    19. If the /S+ option was used in pcretest to study a pattern using JIT,
294        subsequent uses of /S (without +) incorrectly behaved like /S+.
296    21. Retrieve executable code size support for the JIT compiler and fixing
297        some warnings.
299    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
300        not work when the shorter character appeared right at the end of the
301        subject string.
303    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
304        systems.
306    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
307        output it when the /M option is used in pcretest.
309    25. The CheckMan script was not being included in the distribution. Also, added
310        an explicit "perl" to run Perl scripts from the PrepareRelease script
311        because this is reportedly needed in Windows.
313    26. If study data was being save in a file and studying had not found a set of
314        "starts with" bytes for the pattern, the data written to the file (though
315        never used) was taken from uninitialized memory and so caused valgrind to
316        complain.
318    27. Updated RunTest.bat as provided by Sheri Pierce.
320    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
322    29. Computation of memory usage for the table of capturing group names was
323        giving an unnecessarily large value.
326    Version 8.20 21-Oct-2011
327    ------------------------
329    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
330        a POSIX class. After further experiments with Perl, which convinced me that
331        Perl has bugs and confusions, a closing square bracket is no longer allowed
332        in a POSIX name. This bug also affected patterns with classes that started
333        with full stops.
335    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
336        captured substring, but while checking the failing first alternative,
337        substring 1 is temporarily captured. If the output vector supplied to
338        pcre_exec() was not big enough for this capture, the yield of the function
339        was still zero ("insufficient space for captured substrings"). This cannot
340        be totally fixed without adding another stack variable, which seems a lot
341        of expense for a edge case. However, I have improved the situation in cases
342        such as /(a)(b)x|abc/ matched against "abc", where the return code
343        indicates that fewer than the maximum number of slots in the ovector have
344        been set.
346    3.  Related to (2) above: when there are more back references in a pattern than
347        slots in the output vector, pcre_exec() uses temporary memory during
348        matching, and copies in the captures as far as possible afterwards. It was
349        using the entire output vector, but this conflicts with the specification
350        that only 2/3 is used for passing back captured substrings. Now it uses
351        only the first 2/3, for compatibility. This is, of course, another edge
352        case.
354    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
355        main code base, and can be used by building with --enable-jit. When this is
356        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
357        runtime --no-jit option is given.
359    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
360        ovector, the return from the function was zero, implying that there were
361        other matches that did not fit. The correct "exactly full" value is now
362        returned.
364    6.  If a subpattern that was called recursively or as a subroutine contained
365        (*PRUNE) or any other control that caused it to give a non-standard return,
366        invalid errors such as "Error -26 (nested recursion at the same subject
367        position)" or even infinite loops could occur.
369    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
370        computing the minimum length on reaching *ACCEPT, and so ended up with the
371        wrong value of 1 rather than 0. Further investigation indicates that
372        computing a minimum subject length in the presence of *ACCEPT is difficult
373        (think back references, subroutine calls), and so I have changed the code
374        so that no minimum is registered for a pattern that contains *ACCEPT.
376    8.  If (*THEN) was present in the first (true) branch of a conditional group,
377        it was not handled as intended. [But see 16 below.]
379    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
380        Sheri Pierce.
382    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
383        the first byte in a match must be "a".
385    11. Change 17 for 8.13 increased the recursion depth for patterns like
386        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
387        pattern contains any instances of (*THEN). If it does not, the old
388        optimizations are restored. It would be nice to do this on a per-group
389        basis, but at the moment that is not feasible.
391    12. In some environments, the output of pcretest -C is CRLF terminated. This
392        broke RunTest's code that checks for the link size. A single white space
393        character after the value is now allowed for.
395    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
396        For "fr", it uses the Windows-specific input and output files.
398    14. If (*THEN) appeared in a group that was called recursively or as a
399        subroutine, it did not work as intended. [But see next item.]
401    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
402        pattern fragments (but not containing any | characters). If A and B are
403        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
404        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
405        D. In other words, Perl considers parentheses that do not contain any |
406        characters to be part of a surrounding alternative, whereas PCRE was
407        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
408        differently. PCRE now behaves in the same way as Perl, except in the case
409        of subroutine/recursion calls such as (?1) which have in any case always
410        been different (but PCRE had them first :-).
412    16. Related to 15 above: Perl does not treat the | in a conditional group as
413        creating alternatives. Such a group is treated in the same way as an
414        ordinary group without any | characters when processing (*THEN). PCRE has
415        been changed to match Perl's behaviour.
417    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
418        RunGrepTest script failed.
420    18. Change 22 for version 13 caused atomic groups to use more stack. This is
421        inevitable for groups that contain captures, but it can lead to a lot of
422        stack use in large patterns. The old behaviour has been restored for atomic
423        groups that do not contain any capturing parentheses.
425    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
426        suppress the check for a minimum subject length at run time. (If it was
427        given to pcre_exec() or pcre_dfa_exec() it did work.)
429    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
430        fail to work when decoding hex characters in data strings in EBCDIC
431        environments.
433    21. It appears that in at least one Mac OS environment, the isxdigit() function
434        is implemented as a macro that evaluates to its argument more than once,
435        contravening the C 90 Standard (I haven't checked a later standard). There
436        was an instance in pcretest which caused it to go wrong when processing
437        \x{...} escapes in subject strings. The has been rewritten to avoid using
438        things like p++ in the argument of isxdigit().
441    Version 8.13 16-Aug-2011
442    ------------------------
444    1.  The Unicode data tables have been updated to Unicode 6.0.0.
446    2.  Two minor typos in pcre_internal.h have been fixed.
448    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
449        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
450        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
452    4.  There were a number of related bugs in the code for matching backrefences
453        caselessly in UTF-8 mode when codes for the characters concerned were
454        different numbers of bytes. For example, U+023A and U+2C65 are an upper
455        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
456        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
457        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
458        2-byte code at the end of the subject (it thought there wasn't enough data
459        left).
461    5.  Comprehensive information about what went wrong is now returned by
462        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
463        as the output vector has at least 2 elements. The offset of the start of
464        the failing character and a reason code are placed in the vector.
466    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
467        now returned is for the first byte of the failing character, instead of the
468        last byte inspected. This is an incompatible change, but I hope it is small
469        enough not to be a problem. It makes the returned offset consistent with
470        pcre_exec() and pcre_dfa_exec().
472    7.  pcretest now gives a text phrase as well as the error number when
473        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
474        failure, the offset and reason code are output.
476    8.  When \R was used with a maximizing quantifier it failed to skip backwards
477        over a \r\n pair if the subsequent match failed. Instead, it just skipped
478        back over a single character (\n). This seems wrong (because it treated the
479        two characters as a single entity when going forwards), conflicts with the
480        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
481        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
482        has been changed.
484    9.  Some internal refactoring has changed the processing so that the handling
485        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
486        time (the PCRE_DOTALL option was changed this way some time ago: version
487        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
488        which was always a bit of a fudge. It also means that there is one less
489        argument for the match() function, which reduces its stack requirements
490        slightly. This change also fixes an incompatibility with Perl: the pattern
491        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
493    10. More internal refactoring has drastically reduced the number of recursive
494        calls to match() for possessively repeated groups such as (abc)++ when
495        using pcre_exec().
497    11. While implementing 10, a number of bugs in the handling of groups were
498        discovered and fixed:
500        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
501        (a|)*(?1) gave a compile-time internal error.
502        ((a|)+)+  did not notice that the outer group could match an empty string.
503        (^a|^)+   was not marked as anchored.
504        (.*a|.*)+ was not marked as matching at start or after a newline.
506    12. Yet more internal refactoring has removed another argument from the match()
507        function. Special calls to this function are now indicated by setting a
508        value in a variable in the "match data" data block.
510    13. Be more explicit in pcre_study() instead of relying on "default" for
511        opcodes that mean there is no starting character; this means that when new
512        ones are added and accidentally left out of pcre_study(), testing should
513        pick them up.
515    14. The -s option of pcretest has been documented for ages as being an old
516        synonym of -m (show memory usage). I have changed it to mean "force study
517        for every regex", that is, assume /S for every regex. This is similar to -i
518        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
519        using it. It makes it easier to run collections of tests with and without
520        study enabled, and thereby test pcre_study() more easily. All the standard
521        tests are now run with and without -s (but some patterns can be marked as
522        "never study" - see 20 below).
524    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
525        restoration of the capturing data to the outer values was not happening
526        correctly.
528    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
529        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
530        pattern had matched an empty string, and so incorrectly returned a no
531        match.
533    17. There was optimizing code for the last branch of non-capturing parentheses,
534        and also for the obeyed branch of a conditional subexpression, which used
535        tail recursion to cut down on stack usage. Unfortunately, now that there is
536        the possibility of (*THEN) occurring in these branches, tail recursion is
537        no longer possible because the return has to be checked for (*THEN). These
538        two optimizations have therefore been removed. [But see 8.20/11 above.]
540    18. If a pattern containing \R was studied, it was assumed that \R always
541        matched two bytes, thus causing the minimum subject length to be
542        incorrectly computed because \R can also match just one byte.
544    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
545        was incorrectly computed.
547    20. If /S is present twice on a test pattern in pcretest input, it now
548        *disables* studying, thereby overriding the use of -s on the command line
549        (see 14 above). This is necessary for one or two tests to keep the output
550        identical in both cases.
552    21. When (*ACCEPT) was used in an assertion that matched an empty string and
553        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
555    22. When an atomic group that contained a capturing parenthesis was
556        successfully matched, but the branch in which it appeared failed, the
557        capturing was not being forgotten if a higher numbered group was later
558        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
559        group 1 to "a", when in fact it should be unset. This applied to multi-
560        branched capturing and non-capturing groups, repeated or not, and also to
561        positive assertions (capturing in negative assertions does not happen
562        in PCRE) and also to nested atomic groups.
564    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
565        subject after a captured substring, to make it easier to tell which of a
566        number of identical substrings has been captured.
568    24. The way atomic groups are processed by pcre_exec() has been changed so that
569        if they are repeated, backtracking one repetition now resets captured
570        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
571        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
572        "aaa". Previously, it would have been "a". As part of this code
573        refactoring, the way recursive calls are handled has also been changed.
575    25. If an assertion condition captured any substrings, they were not passed
576        back unless some other capturing happened later. For example, if
577        (?(?=(a))a) was matched against "a", no capturing was returned.
579    26. When studying a pattern that contained subroutine calls or assertions,
580        the code for finding the minimum length of a possible match was handling
581        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
582        group 1 called group 2 while simultaneously a separate group 2 called group
583        1). A stack overflow occurred in this case. I have fixed this by limiting
584        the recursion depth to 10.
586    27. Updated RunTest.bat in the distribution to the version supplied by Tom
587        Fortmann. This supports explicit test numbers on the command line, and has
588        argument validation and error reporting.
590    28. An instance of \X with an unlimited repeat could fail if at any point the
591        first character it looked at was a mark character.
593    29. Some minor code refactoring concerning Unicode properties and scripts
594        should reduce the stack requirement of match() slightly.
596    30. Added the '=' option to pcretest to check the setting of unused capturing
597        slots at the end of the pattern, which are documented as being -1, but are
598        not included in the return count.
600    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
601        compiled something random. Now it gives a compile-time error (as does
602        Perl).
604    32. A *MARK encountered during the processing of a positive assertion is now
605        recorded and passed back (compatible with Perl).
607    33. If --only-matching or --colour was set on a pcregrep call whose pattern
608        had alternative anchored branches, the search for a second match in a line
609        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
610        matched the line "0102" twice. The same bug affected patterns that started
611        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
612        twice.
614    34. Previously, PCRE did not allow quantification of assertions. However, Perl
615        does, and because of capturing effects, quantifying parenthesized
616        assertions may at times be useful. Quantifiers are now allowed for
617        parenthesized assertions.
619    35. A minor code tidy in pcre_compile() when checking options for \R usage.
621    36. \g was being checked for fancy things in a character class, when it should
622        just be a literal "g".
624    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
625        appearance of a nested POSIX class supersedes an apparent external class.
626        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
627        unescaped square brackets may also appear as part of class names. For
628        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
629        more like Perl. (But see 8.20/1 above.)
631    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
632        was because it thought it was \N{name}, which is not supported).
634    39. Add minix to OS list not supporting the -S option in pcretest.
636    40. PCRE tries to detect cases of infinite recursion at compile time, but it
637        cannot analyze patterns in sufficient detail to catch mutual recursions
638        such as ((?1))((?2)). There is now a runtime test that gives an error if a
639        subgroup is called recursively as a subpattern for a second time at the
640        same position in the subject string. In previous releases this might have
641        been caught by the recursion limit, or it might have run out of stack.
643    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
644        happen only once. PCRE was, however incorrectly giving a compile time error
645        "recursive call could loop indefinitely" because it cannot analyze the
646        pattern in sufficient detail. The compile time test no longer happens when
647        PCRE is compiling a conditional subpattern, but actual runaway loops are
648        now caught at runtime (see 40 above).
650    42. It seems that Perl allows any characters other than a closing parenthesis
651        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
652        has been changed to be the same.
654    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
655        as not to get warnings when autogen.sh is called. Also changed
656        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
658    44. To help people who use pcregrep to scan files containing exceedingly long
659        lines, the following changes have been made:
661        (a) The default value of the buffer size parameter has been increased from
662            8K to 20K. (The actual buffer used is three times this size.)
664        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
665            PCRE is built.
667        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
668            to be set at run time.
670        (d) Numerical values in pcregrep options can be followed by K or M, for
671            example --buffer-size=50K.
673        (e) If a line being scanned overflows pcregrep's buffer, an error is now
674            given and the return code is set to 2.
676    45. Add a pointer to the latest mark to the callout data block.
678    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
679        partial match of an empty string instead of no match. This was specific to
680        the use of ".".
682    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
683        complete match instead of a partial match. This bug was dependent on both
684        the PCRE_UTF8 and PCRE_DOTALL options being set.
686    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
687        starting byte set, because \b was not being ignored.
690    Version 8.12 15-Jan-2011
691    ------------------------
693    1.  Fixed some typos in the markup of the man pages, and wrote a script that
694        checks for such things as part of the documentation building process.
696    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
697        --match-limit and --recursion-limit options (added for 8.11). In
698        particular, this made one of the standard tests fail. (The integer value
699        went into the wrong half of a long int.)
701    3.  If the --colour option was given to pcregrep with -v (invert match), it
702        did strange things, either producing crazy output, or crashing. It should,
703        of course, ignore a request for colour when reporting lines that do not
704        match.
706    4.  Another pcregrep bug caused similar problems if --colour was specified with
707        -M (multiline) and the pattern match finished with a line ending.
709    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
710        matched in multiline mode, the following line was shown as part of the
711        match. This seems wrong, so I have changed it.
713    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
714        the check for further matches in the same line (so they could be coloured)
715        to overrun the end of the current line. If another match was found, it was
716        incorrectly shown (and then shown again when found in the next line).
718    7.  If pcregrep was compiled under Windows, there was a reference to the
719        function pcregrep_exit() before it was defined. I am assuming this was
720        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
721        reported by a user. I've moved the definition above the reference.
724    Version 8.11 10-Dec-2010
725    ------------------------
727    1.  (*THEN) was not working properly if there were untried alternatives prior
728        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
729        backtracked to try for "b" instead of moving to the next alternative branch
730        at the same level (in this case, to look for "c"). The Perl documentation
731        is clear that when (*THEN) is backtracked onto, it goes to the "next
732        alternative in the innermost enclosing group".
734    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
735        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
736        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
737        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
738        (*THEN).
740    3.  If \s appeared in a character class, it removed the VT character from
741        the class, even if it had been included by some previous item, for example
742        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
743        of \s, but is part of the POSIX "space" class.)
745    4.  A partial match never returns an empty string (because you can always
746        match an empty string at the end of the subject); however the checking for
747        an empty string was starting at the "start of match" point. This has been
748        changed to the "earliest inspected character" point, because the returned
749        data for a partial match starts at this character. This means that, for
750        example, /(?<=abc)def/ gives a partial match for the subject "abc"
751        (previously it gave "no match").
753    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
754        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
755        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
756        has an implication that the given string is incomplete (because a partial
757        match is preferred over a full match). For this reason, these items now
758        give a partial match in this situation. [Aside: previously, the one case
759        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
760        match rather than a full match, which was wrong by the old rules, but is
761        now correct.]
763    6.  There was a bug in the handling of #-introduced comments, recognized when
764        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
765        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
766        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
767        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
768        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
769        places in pcre_compile().
771    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
772        comments when looking ahead for named forward references to subpatterns,
773        the only newline sequence it recognized was NL. It now handles newlines
774        according to the set newline convention.
776    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
777        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
778        cater for a lack of strerror(). These oversights have been fixed.
780    9.  Added --match-limit and --recursion-limit to pcregrep.
782    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
784    11. When the -o option was used, pcregrep was setting a return code of 1, even
785        when matches were found, and --line-buffered was not being honoured.
787    12. Added an optional parentheses number to the -o and --only-matching options
788        of pcregrep.
790    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
791        can match an empty string. The code to do it in pcretest and pcredemo
792        needed fixing:
794        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
795            only one byte after an empty string match just before CRLF (this case
796            just got forgotten; "any" and "anycrlf" were OK).
798        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
799            mode when an empty string match preceded an ASCII character followed by
800            a non-ASCII character. (The code for advancing by one character rather
801            than one byte was nonsense.)
803        (c) The pcredemo.c sample program did not have any code at all to handle
804            the cases when CRLF is a valid newline sequence.
806    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
807        as a starting offset was within the subject string. There is now a new
808        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
809        negative or greater than the length of the string. In order to test this,
810        pcretest is extended to allow the setting of negative starting offsets.
812    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
813        starting offset points to the beginning of a UTF-8 character was
814        unnecessarily clumsy. I tidied it up.
816    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
817        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
819    17. Nobody had reported that the --include_dir option, which was added in
820        release 7.7 should have been called --include-dir (hyphen, not underscore)
821        for compatibility with GNU grep. I have changed it to --include-dir, but
822        left --include_dir as an undocumented synonym, and the same for
823        --exclude-dir, though that is not available in GNU grep, at least as of
824        release 2.5.4.
826    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
827        characters from a string of bytes) have been redefined so as not to use
828        loops, in order to improve performance in some environments. At the same
829        time, I abstracted some of the common code into auxiliary macros to save
830        repetition (this should not affect the compiled code).
832    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
833        compile-time error is now given if \c is not followed by an ASCII
834        character, that is, a byte less than 128. (In EBCDIC mode, the code is
835        different, and any byte value is allowed.)
837    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
838        START_OPTIMIZE option, which is now allowed at compile time - but just
839        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
840        to pcregrep and other applications that have no direct access to PCRE
841        options. The new /Y option in pcretest sets this option when calling
842        pcre_compile().
844    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
845        back references. Groups containing recursive back references were forced to
846        be atomic by that change, but in the case of named groups, the amount of
847        memory required was incorrectly computed, leading to "Failed: internal
848        error: code overflow". This has been fixed.
850    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
851        pcretest.c, to avoid build problems in some Borland environments.
854    Version 8.10 25-Jun-2010
855    ------------------------
857    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
858        THEN.
860    2.  (*ACCEPT) was not working when inside an atomic group.
862    3.  Inside a character class, \B is treated as a literal by default, but
863        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
864        causes the error). The code is unchanged, but I tidied the documentation.
866    4.  Inside a character class, PCRE always treated \R and \X as literals,
867        whereas Perl faults them if its -w option is set. I have changed PCRE so
868        that it faults them when PCRE_EXTRA is set.
870    5.  Added support for \N, which always matches any character other than
871        newline. (It is the same as "." when PCRE_DOTALL is not set.)
873    6.  When compiling pcregrep with newer versions of gcc which may have
874        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
875        declared with attribute warn_unused_result" were given. Just casting the
876        result to (void) does not stop the warnings; a more elaborate fudge is
877        needed. I've used a macro to implement this.
879    7.  Minor change to pcretest.c to avoid a compiler warning.
881    8.  Added four artifical Unicode properties to help with an option to make
882        \s etc use properties (see next item). The new properties are: Xan
883        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
885    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
886        use Unicode properties. (*UCP) at the start of a pattern can be used to set
887        this option. Modified pcretest to add /W to test this facility. Added
888        REG_UCP to make it available via the POSIX interface.
890    10. Added --line-buffered to pcregrep.
892    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
893        studied, and the match started with a letter with a code point greater than
894        127 whose first byte was different to the first byte of the other case of
895        the letter, the other case of this starting letter was not recognized
896        (#976).
898    12. If a pattern that was studied started with a repeated Unicode property
899        test, for example, \p{Nd}+, there was the theoretical possibility of
900        setting up an incorrect bitmap of starting bytes, but fortunately it could
901        not have actually happened in practice until change 8 above was made (it
902        added property types that matched character-matching opcodes).
904    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
905        possible starting bytes for non-anchored patterns.
907    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
908        \R, and also a number of cases that involve Unicode properties, both
909        explicit and implicit when PCRE_UCP is set.
911    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
912        input, it could crash or give wrong results if characters with values
913        greater than 0xc0 were present in the subject string. (Detail: it assumed
914        UTF-8 input when processing these items.)
916    16. Added a lot of (int) casts to avoid compiler warnings in systems where
917        size_t is 64-bit (#991).
919    17. Added a check for running out of memory when PCRE is compiled with
920        --disable-stack-for-recursion (#990).
922    18. If the last data line in a file for pcretest does not have a newline on
923        the end, a newline was missing in the output.
925    19. The default pcre_chartables.c file recognizes only ASCII characters (values
926        less than 128) in its various bitmaps. However, there is a facility for
927        generating tables according to the current locale when PCRE is compiled. It
928        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
929        characters, are recognized by isspace() and therefore were getting set in
930        these tables, and indeed these tables seem to approximate to ISO 8859. This
931        caused a problem in UTF-8 mode when pcre_study() was used to create a list
932        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
933        which of course cannot start UTF-8 characters. I have changed the code so
934        that only real ASCII characters (less than 128) and the correct starting
935        bytes for UTF-8 encodings are set for characters greater than 127 when in
936        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
937        altogether.)
939    20. Added the /T option to pcretest so as to be able to run tests with non-
940        standard character tables, thus making it possible to include the tests
941        used for 19 above in the standard set of tests.
943    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
944        reference to a subpattern the other side of a comment that contains an
945        opening parenthesis caused either an internal compiling error, or a
946        reference to the wrong subpattern.
949    Version 8.02 19-Mar-2010
950    ------------------------
952    1.  The Unicode data tables have been updated to Unicode 5.2.0.
954    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
955        configured.
957    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
958        original author of that file, following a query about its status.
960    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
961        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
963    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
964        quantifier applied to a forward-referencing subroutine call, could compile
965        incorrect code or give the error "internal error: previously-checked
966        referenced subpattern not found".
968    6.  Both MS Visual Studio and Symbian OS have problems with initializing
969        variables to point to external functions. For these systems, therefore,
970        pcre_malloc etc. are now initialized to local functions that call the
971        relevant global functions.
973    7.  There were two entries missing in the vectors called coptable and poptable
974        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
975        I've fixed the data, and added a kludgy way of testing at compile time that
976        the lengths are correct (equal to the number of opcodes).
978    8.  Following on from 7, I added a similar kludge to check the length of the
979        eint vector in pcreposix.c.
981    9.  Error texts for pcre_compile() are held as one long string to avoid too
982        much relocation at load time. To find a text, the string is searched,
983        counting zeros. There was no check for running off the end of the string,
984        which could happen if a new error number was added without updating the
985        string.
987    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
989    11. \K was not working if it appeared in an atomic group or in a group that
990        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
991        \K is "not well defined" if used in an assertion. PCRE now accepts it if
992        the assertion is positive, but not if it is negative.
994    12. Change 11 fortuitously reduced the size of the stack frame used in the
995        "match()" function of pcre_exec.c by one pointer. Forthcoming
996        implementation of support for (*MARK) will need an extra pointer on the
997        stack; I have reserved it now, so that the stack frame size does not
998        decrease.
1000    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1001        item in branch that calls a recursion is a subroutine call - as in the
1002        second branch in the above example - was incorrectly given the compile-
1003        time error "recursive call could loop indefinitely" because pcre_compile()
1004        was not correctly checking the subroutine for matching a non-empty string.
1006    14. The checks for overrunning compiling workspace could trigger after an
1007        overrun had occurred. This is a "should never occur" error, but it can be
1008        triggered by pathological patterns such as hundreds of nested parentheses.
1009        The checks now trigger 100 bytes before the end of the workspace.
1011    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1014    Version 8.01 19-Jan-2010
1015    ------------------------
1017    1.  If a pattern contained a conditional subpattern with only one branch (in
1018        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1019        computed the wrong minimum data length (which is of course zero for such
1020        subpatterns). This could cause incorrect "no match" results.
1022    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1023        the pattern is reset in the first branch, pcre_compile() failed with
1024        "internal error: code overflow at offset...". This happened only when
1025        the reset was to the original external option setting. (An optimization
1026        abstracts leading options settings into an external setting, which was the
1027        cause of this.)
1029    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1030        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1031        assertion pattern did not match (meaning that the assertion was true), it
1032        was incorrectly treated as false if the SKIP had been reached during the
1033        matching. This also applied to assertions used as conditions.
1035    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1036        assertion subpattern, including such a pattern used as a condition,
1037        unpredictable results occurred, instead of the error return
1040    5.  The C++ GlobalReplace function was not working like Perl for the special
1041        situation when an empty string is matched. It now does the fancy magic
1042        stuff that is necessary.
1044    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1045        removed. (These were left over from very, very early versions of PCRE.)
1047    7.  Some cosmetic changes to the code to make life easier when compiling it
1048        as part of something else:
1050        (a) Change DEBUG to PCRE_DEBUG.
1052        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1053            called "current" as "current_branch", to prevent a collision with the
1054            Linux macro when compiled as a kernel module.
1056        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1057            prevent a collision with the Linux macro when compiled as a kernel
1058            module.
1060    8.  In pcre_compile() there are some checks for integer overflows that used to
1061        cast potentially large values to (double). This has been changed to that
1062        when building, a check for int64_t is made, and if it is found, it is used
1063        instead, thus avoiding the use of floating point arithmetic. (There is no
1064        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1065        double.
1067    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1068        2005 (difference between two addresses compared to an unsigned value).
1070    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1071        custom one, because of the following reported problem in Windows:
1073          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1074              under Win32.
1075          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1076              therefore missing the function definition.
1077          - The compiler thus generates a "C" signature for the test function.
1078          - The linker fails to find the "C" function.
1079          - PCRE fails to configure if asked to do so against libbz2.
1081    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1082        messages were output:
1084          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1085          rerunning libtoolize, to keep the correct libtool macros in-tree.
1086          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1088        I have done both of these things.
1090    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1091        most of the time, it *can* run out if it is given a pattern that contains a
1092        runaway infinite recursion. I updated the discussion in the pcrestack man
1093        page.
1095    13. Now that we have gone to the x.xx style of version numbers, the minor
1096        version may start with zero. Using 08 or 09 is a bad idea because users
1097        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1098        interpreted as invalid octal numbers. I've updated the previous comment in
1099        configure.ac, and also added a check that gives an error if 08 or 09 are
1100        used.
1102    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1103        causing partial matching to fail when the end of the subject matched \W
1104        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1106    15. There were some discrepancies between the declarations in pcre_internal.h
1107        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1108        their definitions. The declarations used "const uschar *" and the
1109        definitions used USPTR. Even though USPTR is normally defined as "const
1110        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1111        reported that: "This difference in casting confuses some C++ compilers, for
1112        example, SunCC recognizes above declarations as different functions and
1113        generates broken code for hbpcre." I have changed the declarations to use
1114        USPTR.
1116    16. GNU libtool is named differently on some systems. The autogen.sh script now
1117        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1118        (FreeBSD).
1120    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1121        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1122        comment: "Figure out how to create a longlong from a string: strtoll and
1123        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1124        instance, but it only takes 2 args instead of 3!"
1126    18. A subtle bug concerned with back references has been fixed by a change of
1127        specification, with a corresponding code fix. A pattern such as
1128        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1129        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1130        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1131        same bug. Such groups have to be quantified to be useful, or contained
1132        inside another quantified group. (If there's no repetition, the reference
1133        can never match.) The problem arises because, having left the group and
1134        moved on to the rest of the pattern, a later failure that backtracks into
1135        the group uses the captured value from the final iteration of the group
1136        rather than the correct earlier one. I have fixed this in PCRE by forcing
1137        any group that contains a reference to itself to be an atomic group; that
1138        is, there cannot be any backtracking into it once it has completed. This is
1139        similar to recursive and subroutine calls.
1142    Version 8.00 19-Oct-09
1143    ----------------------
1145    1.  The table for translating pcre_compile() error codes into POSIX error codes
1146        was out-of-date, and there was no check on the pcre_compile() error code
1147        being within the table. This could lead to an OK return being given in
1148        error.
1150    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1151        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1152        in a Windows environment.
1154    3.  The pcregrep --count option prints the count for each file even when it is
1155        zero, as does GNU grep. However, pcregrep was also printing all files when
1156        --files-with-matches was added. Now, when both options are given, it prints
1157        counts only for those files that have at least one match. (GNU grep just
1158        prints the file name in this circumstance, but including the count seems
1159        more useful - otherwise, why use --count?) Also ensured that the
1160        combination -clh just lists non-zero counts, with no names.
1162    4.  The long form of the pcregrep -F option was incorrectly implemented as
1163        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1164        but it seems right to fix it, and I didn't think it was worth preserving
1165        the old behaviour.
1167    5.  The command line items --regex=pattern and --regexp=pattern were not
1168        recognized by pcregrep, which required --regex pattern or --regexp pattern
1169        (with a space rather than an '='). The man page documented the '=' forms,
1170        which are compatible with GNU grep; these now work.
1172    6.  No libpcreposix.pc file was created for pkg-config; there was just
1173        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1175    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1176        when UCP support is not needed, by modifying the Python script that
1177        generates it from Unicode data files. This should not matter if the module
1178        is correctly used as a library, but I received one complaint about 50K of
1179        unwanted data. My guess is that the person linked everything into his
1180        program rather than using a library. Anyway, it does no harm.
1182    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1183        was a minimum greater than 1 for a wide character in a possessive
1184        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1185        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1186        character. Chaos in the form of incorrect output or a compiling loop could
1187        result.
1189    9.  The restrictions on what a pattern can contain when partial matching is
1190        requested for pcre_exec() have been removed. All patterns can now be
1191        partially matched by this function. In addition, if there are at least two
1192        slots in the offset vector, the offset of the earliest inspected character
1193        for the match and the offset of the end of the subject are set in them when
1194        PCRE_ERROR_PARTIAL is returned.
1196    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1197        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1198        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1199        and may be more useful for multi-segment matching.
1201    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1202        used to be given if ever the end of the subject was reached; now it is
1203        given only if matching could not proceed because another character was
1204        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1205        string "Z", which now yields "no match" instead of "partial match". In the
1206        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1207        final character ended with (*FAIL).
1209    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1210        if the pattern had a "must contain" character that was already found in the
1211        earlier partial match, unless partial matching was again requested. For
1212        example, with the pattern /dog.(body)?/, the "must contain" character is
1213        "g". If the first part-match was for the string "dog", restarting with
1214        "sbody" failed. This bug has been fixed.
1216    13. The string returned by pcre_dfa_exec() after a partial match has been
1217        changed so that it starts at the first inspected character rather than the
1218        first character of the match. This makes a difference only if the pattern
1219        starts with a lookbehind assertion or \b or \B (\K is not supported by
1220        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1221        matching functions compatible, and I think it's the right thing to do.
1223    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1224        so that the demonstration program is easily available in environments where
1225        PCRE has not been installed from source.
1227    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1228        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1229        library.
1231    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1232        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1233        is not the first non-POSIX option to be added. Clearly some people find
1234        these options useful.
1236    17. If a caller to the POSIX matching function regexec() passes a non-zero
1237        value for nmatch with a NULL value for pmatch, the value of
1238        nmatch is forced to zero.
1240    18. RunGrepTest did not have a test for the availability of the -u option of
1241        the diff command, as RunTest does. It now checks in the same way as
1242        RunTest, and also checks for the -b option.
1244    19. If an odd number of negated classes containing just a single character
1245        interposed, within parentheses, between a forward reference to a named
1246        subpattern and the definition of the subpattern, compilation crashed with
1247        an internal error, complaining that it could not find the referenced
1248        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1249        [The bug was that it was starting one character too far in when skipping
1250        over the character class, thus treating the ] as data rather than
1251        terminating the class. This meant it could skip too much.]
1253    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1254        /g option in pcretest when the pattern contains \K, which makes it possible
1255        to have an empty string match not at the start, even when the pattern is
1256        anchored. Updated pcretest and pcredemo to use this option.
1258    21. If the maximum number of capturing subpatterns in a recursion was greater
1259        than the maximum at the outer level, the higher number was returned, but
1260        with unset values at the outer level. The correct (outer level) value is
1261        now given.
1263    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1264        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1265        make it do so. The string so far is captured, making this feature
1266        compatible with Perl.
1268    23. The tests have been re-organized, adding tests 11 and 12, to make it
1269        possible to check the Perl 5.10 features against Perl 5.10.
1271    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1272        pattern matches a fixed length string. PCRE did not allow this; now it
1273        does. Neither allows recursion.
1275    25. I finally figured out how to implement a request to provide the minimum
1276        length of subject string that was needed in order to match a given pattern.
1277        (It was back references and recursion that I had previously got hung up
1278        on.) This code has now been added to pcre_study(); it finds a lower bound
1279        to the length of subject needed. It is not necessarily the greatest lower
1280        bound, but using it to avoid searching strings that are too short does give
1281        some useful speed-ups. The value is available to calling programs via
1282        pcre_fullinfo().
1284    26. While implementing 25, I discovered to my embarrassment that pcretest had
1285        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1286        study optimizations had never been tested with that matching function.
1287        Oops. What is worse, even when it was passed study data, there was a bug in
1288        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1289        were also very few tests of studied patterns with pcre_dfa_exec().
1291    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1292        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1293        on the other side of the coin, they are no longer allowed to have different
1294        names, because these cannot be distinguished in PCRE, and this has caused
1295        confusion. (This is a difference from Perl.)
1297    28. When duplicate subpattern names are present (necessarily with different
1298        numbers, as required by 27 above), and a test is made by name in a
1299        conditional pattern, either for a subpattern having been matched, or for
1300        recursion in such a pattern, all the associated numbered subpatterns are
1301        tested, and the overall condition is true if the condition is true for any
1302        one of them. This is the way Perl works, and is also more like the way
1303        testing by number works.
1306    Version 7.9 11-Apr-09
1307    ---------------------
1309    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1310        (pcretest), all targets were linked against these libraries. This included
1311        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1312        libraries. This caused unwanted dependencies to be created. This problem
1313        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1314        pcretest is linked with readline.
1316    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1317        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1318        moved outside it again, because FALSE and TRUE are already defined in AIX,
1319        but BOOL is not.
1321    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1322        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1324    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1325        hyphens) following file names and line numbers when outputting matching
1326        lines. This is not true; no spaces are inserted. I have also clarified the
1327        wording for the --colour (or --color) option.
1329    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1330        was not coloured; this is different to GNU grep, so I have changed it to be
1331        the same.
1333    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1334        each matching line was coloured. Now it goes on to look for further matches
1335        of any of the test patterns, which is the same behaviour as GNU grep.
1337    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1338        doesn't make sense to accept an empty string match in pcregrep, so I have
1339        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1340        seems to be how GNU grep behaves.
1342    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1343        start or after a newline", because the conditional assertion was not being
1344        correctly handled. The rule now is that both the assertion and what follows
1345        in the first alternative must satisfy the test.
1347    9.  If auto-callout was enabled in a pattern with a conditional group whose
1348        condition was an assertion, PCRE could crash during matching, both with
1349        pcre_exec() and pcre_dfa_exec().
1351    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1352        used for matching.
1354    11. Unicode property support in character classes was not working for
1355        characters (bytes) greater than 127 when not in UTF-8 mode.
1357    12. Added the -M command line option to pcretest.
1359    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1361    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1363    16. Added comments and documentation about mis-use of no_arg in the C++
1364        wrapper.
1366    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1367        from Martin Jerabek that uses macro names for all relevant character and
1368        string constants.
1370    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1371        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1372        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1373        these, but not everybody uses configure.
1375    19. A conditional group that had only one branch was not being correctly
1376        recognized as an item that could match an empty string. This meant that an
1377        enclosing group might also not be so recognized, causing infinite looping
1378        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1379        with the subject "ab", where knowledge that the repeated group can match
1380        nothing is needed in order to break the loop.
1382    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1383        exec(), but without supplying a callout function, matching went wrong.
1385    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1386        leak if the size of the offset vector was greater than 30. When the vector
1387        is smaller, the saved offsets during recursion go onto a local stack
1388        vector, but for larger vectors malloc() is used. It was failing to free
1389        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1390        error, in fact).
1392    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1393        heapframe that is used only when UTF-8 support is enabled. This caused no
1394        problem, but was untidy.
1396    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1397        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1398        included within another project.
1400    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1401        slightly modified by me:
1403          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1404              not building pcregrep.
1406          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1407              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1409    25. Forward references, both numeric and by name, in patterns that made use of
1410        duplicate group numbers, could behave incorrectly or give incorrect errors,
1411        because when scanning forward to find the reference group, PCRE was not
1412        taking into account the duplicate group numbers. A pattern such as
1413        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1415    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1416        the feature of a custom pointer more persuasive (as requested by a user).
1418    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1419        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1420        pre-defined.
1422    28. Added support for (*UTF8) at the start of a pattern.
1424    29. Arrange for flags added by the "release type" setting in CMake to be shown
1425        in the configuration summary.
1428    Version 7.8 05-Sep-08
1429    ---------------------
1431    1.  Replaced UCP searching code with optimized version as implemented for Ad
1432        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1433        stage table and inline lookup instead of a function, giving speed ups of 2
1434        to 5 times on some simple patterns that I tested. Permission was given to
1435        distribute the MultiStage2.py script that generates the tables (it's not in
1436        the tarball, but is in the Subversion repository).
1438    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1439        scripts.
1441    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1442        a group with a zero qualifier. The result of the study could be incorrect,
1443        or the function might crash, depending on the pattern.
1445    4.  Caseless matching was not working for non-ASCII characters in back
1446        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1447        It now works when Unicode Property Support is available.
1449    5.  In pcretest, an escape such as \x{de} in the data was always generating
1450        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1451        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1452        truncation.
1454    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1456    7.  Added two (int) casts to pcregrep when printing the difference of two
1457        pointers, in case they are 64-bit values.
1459    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1460        test 2 if it fails.
1462    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1463        and a #define of that name to empty if it is not externally set. This is to
1464        allow users of MSVC to set it if necessary.
1466    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1467        the convenience functions in the pcre_get.c source file.
1469    11. An option change at the start of a pattern that had top-level alternatives
1470        could cause overwriting and/or a crash. This command provoked a crash in
1471        some environments:
1473          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1475        This potential security problem was recorded as CVE-2008-2371.
1477    12. For a pattern where the match had to start at the beginning or immediately
1478        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1479        pcre_dfa_exec() could read past the end of the passed subject if there was
1480        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1481        pcretest so that it places the subject at the end of its malloc-ed buffer.
1483    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1484        exec() might read past the end of the data buffer in UTF-8 mode.
1486    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1487        the data contained the byte 0x85 as part of a UTF-8 character within its
1488        first line. This applied both to normal and DFA matching.
1490    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1491        /^[^d]*?$/8 failed to match "abc".
1493    16. Added a missing copyright notice to pcrecpp_internal.h.
1495    17. Make it more clear in the documentation that values returned from
1496        pcre_exec() in ovector are byte offsets, not character counts.
1498    18. Tidied a few places to stop certain compilers from issuing warnings.
1500    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1501        supplied by Stefan Weber. I made a further small update for 7.8 because
1502        there is a change of source arrangements: the pcre_searchfuncs.c module is
1503        replaced by pcre_ucd.c.
1506    Version 7.7 07-May-08
1507    ---------------------
1509    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1510        a string to a long long, pretend we don't even have a long long." This is
1511        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1513    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1514        pre-7.6 versions, which defined a global no_arg variable instead of putting
1515        it in the RE class. (See also #8 below.)
1517    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1518        Lopes.
1520    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1522        (1) The include/exclude patterns were being applied to the whole pathnames
1523            of files, instead of just to the final components.
1525        (2) If there was more than one level of directory, the subdirectories were
1526            skipped unless they satisfied the include/exclude conditions. This is
1527            inconsistent with GNU grep (and could even be seen as contrary to the
1528            pcregrep specification - which I improved to make it absolutely clear).
1529            The action now is always to scan all levels of directory, and just
1530            apply the include/exclude patterns to regular files.
1532    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1533        --exclude_dir in the tests to avoid scanning .svn directories.
1535    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1536        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1537        doesn't support NULs in patterns.
1539    7.  Added some missing "const"s to declarations of static tables in
1540        pcre_compile.c and pcre_dfa_exec.c.
1542    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1543        caused by fix #2  above. (Subsequently also a second patch to fix the
1544        first patch. And a third patch - this was a messy problem.)
1546    9.  Applied Craig's patch to remove the use of push_back().
1548    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1549        matching function regexec().
1551    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1552        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1553        references. PCRE supports relative numbers with this syntax (I don't think
1554        Oniguruma does).
1556    12. Previously, a group with a zero repeat such as (...){0} was completely
1557        omitted from the compiled regex. However, this means that if the group
1558        was called as a subroutine from elsewhere in the pattern, things went wrong
1559        (an internal error was given). Such groups are now left in the compiled
1560        pattern, with a new opcode that causes them to be skipped at execution
1561        time.
1563    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1564        to the way PCRE behaves:
1566        (a) A lone ] character is dis-allowed (Perl treats it as data).
1568        (b) A back reference to an unmatched subpattern matches an empty string
1569            (Perl fails the current match path).
1571        (c) A data ] in a character class must be notated as \] because if the
1572            first data character in a class is ], it defines an empty class. (In
1573            Perl it is not possible to have an empty class.) The empty class []
1574            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1575            The negative empty class [^] matches any one character, independently
1576            of the DOTALL setting.
1578    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1579        non-existent subpattern following a character class starting with ']' and
1580        containing () gave an internal compiling error instead of "reference to
1581        non-existent subpattern". Fortunately, when the pattern did exist, the
1582        compiled code was correct. (When scanning forwards to check for the
1583        existencd of the subpattern, it was treating the data ']' as terminating
1584        the class, so got the count wrong. When actually compiling, the reference
1585        was subsequently set up correctly.)
1587    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1588        it was being rejected as not supported by pcre_dfa_exec(), even though
1589        other assertions are supported. I have made pcre_dfa_exec() support
1590        (*FAIL).
1592    16. The implementation of 13c above involved the invention of a new opcode,
1593        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1594        cannot be changed at match time, I realized I could make a small
1595        improvement to matching performance by compiling OP_ALLANY instead of
1596        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1597        on the OP_ANY path.
1599    17. Compiling pcretest on Windows with readline support failed without the
1600        following two fixes: (1) Make the unistd.h include conditional on
1601        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1603    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1604        ncurses library to be included for pcretest when ReadLine support is
1605        requested, but also to allow for it to be overridden. This patch came from
1606        Daniel Bergström.
1608    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1609        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1610        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1611        spotting this.
1614    Version 7.6 28-Jan-08
1615    ---------------------
1617    1.  A character class containing a very large number of characters with
1618        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1619        overflow.
1621    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1622        HAVE_LONG_LONG is not defined.
1624    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1625        bring them up to date and include new features. This patch includes:
1627        - Fixed PH's badly added libz and libbz2 support.
1628        - Fixed a problem with static linking.
1629        - Added pcredemo. [But later removed - see 7 below.]
1630        - Fixed dftables problem and added an option.
1631        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1632            HAVE_LONG_LONG.
1633        - Added readline support for pcretest.
1634        - Added an listing of the option settings after cmake has run.
1636    4.  A user submitted a patch to Makefile that makes it easy to create
1637        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1638        Makefile.am that cause it to include this special target, without
1639        affecting anything else. Note that the same mingw target plus all
1640        the other distribution libraries and programs are now supported
1641        when configuring with CMake (see 6 below) instead of with
1642        Configure/Make.
1644    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1645        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1646        exported in the Windows port". It has not yet been confirmed that the patch
1647        solves the problem, but it does no harm.
1649    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1650        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1651        with CMake, and also correct the comment about stack recursion.
1653    7.  Remove the automatic building of pcredemo from the ./configure system and
1654        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1655        of a program that users should build themselves after PCRE is installed, so
1656        building it automatically is not really right. What is more, it gave
1657        trouble in some build environments.
1659    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1662    Version 7.5 10-Jan-08
1663    ---------------------
1665    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1666        values in parens when parsing an RE using the C++ wrapper."
1668    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1669        Characters greater than 255 were excluded from the class instead of being
1670        included.
1672    3.  The same bug as (2) above applied to negated POSIX classes such as
1673        [:^space:].
1675    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1676        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1677        I have changed it.
1679    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1680        first named subpattern) and a construct such as (?&a) would reference the
1681        first named subpattern whose name started with "a" (in other words, the
1682        length check was missing). Both these problems are fixed. "Subpattern name
1683        expected" is now given for (?&) (a zero-length name), and this patch also
1684        makes it give the same error for \k'' (previously it complained that that
1685        was a reference to a non-existent subpattern).
1687    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1688        this is right because (?- can be followed by option settings as well as by
1689        digits. I have, however, made the messages clearer.
1691    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1692        than the number used in the conditional) now cause a compile-time error.
1693        This is actually not compatible with Perl, which accepts such patterns, but
1694        treats the conditional as always being FALSE (as PCRE used to), but it
1695        seems to me that giving a diagnostic is better.
1697    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1698        and messages.
1700    9.  Fix two occurrences of "backslash" in comments that should have been
1701        "backspace".
1703    10. Remove two redundant lines of code that can never be obeyed (their function
1704        was moved elsewhere).
1706    11. The program that makes PCRE's Unicode character property table had a bug
1707        which caused it to generate incorrect table entries for sequences of
1708        characters that have the same character type, but are in different scripts.
1709        It amalgamated them into a single range, with the script of the first of
1710        them. In other words, some characters were in the wrong script. There were
1711        thirteen such cases, affecting characters in the following ranges:
1713          U+002b0 - U+002c1
1714          U+0060c - U+0060d
1715          U+0061e - U+00612
1716          U+0064b - U+0065e
1717          U+0074d - U+0076d
1718          U+01800 - U+01805
1719          U+01d00 - U+01d77
1720          U+01d9b - U+01dbf
1721          U+0200b - U+0200f
1722          U+030fc - U+030fe
1723          U+03260 - U+0327f
1724          U+0fb46 - U+0fbb1
1725          U+10450 - U+1049d
1727    12. The -o option (show only the matching part of a line) for pcregrep was not
1728        compatible with GNU grep in that, if there was more than one match in a
1729        line, it showed only the first of them. It now behaves in the same way as
1730        GNU grep.
1732    13. If the -o and -v options were combined for pcregrep, it printed a blank
1733        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1734        does the same. The return code can be used to tell if there were any
1735        non-matching lines.
1737    14. Added --file-offsets and --line-offsets to pcregrep.
1739    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1740        infinitely looping recursion. The bug was that positive lookaheads were not
1741        being skipped when checking for a possible empty match (negative lookaheads
1742        and both kinds of lookbehind were skipped).
1744    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1745        inclusion of <windows.h> to before rather than after the definition of
1746        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1748    17. Specifying a possessive quantifier with a specific limit for a Unicode
1749        character property caused pcre_compile() to compile bad code, which led at
1750        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1751        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1752        caused the error; without that there was no problem.
1754    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1756    19. Added --enable-pcretest-libreadline.
1758    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1759        RE::GlobalReplace(). As a result, the number of replacements returned was
1760        double what it should be. I removed one of the increments, but Craig sent a
1761        later patch that removed the other one (the right fix) and added unit tests
1762        that check the return values (which was not done before).
1764    21. Several CMake things:
1766        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1767            the names libpcre and libpcreposix, not just pcre and pcreposix.
1769        (2) The above change means that pcretest and pcregrep are now correctly
1770            linked with the newly-built libraries, not previously installed ones.
1774    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1775        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1776        UTF-8 newline character). The key issue is that the pattern starts .*;
1777        this means that the match must be either at the beginning, or after a
1778        newline. The bug was in the code for advancing after a failed match and
1779        checking that the new position followed a newline. It was not taking
1780        account of UTF-8 characters correctly.
1782    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1783        character classes. PCRE was not treating the sequence [:...:] as a
1784        character class unless the ... were all letters. Perl, however, seems to
1785        allow any characters between [: and :], though of course it rejects as
1786        unknown any "names" that contain non-letters, because all the known class
1787        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1788        for example, whereas PCRE did not - it did not recognize a POSIX character
1789        class. This seemed a bit dangerous, so the code has been changed to be
1790        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1791        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1792        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1793        Perl does, and where it didn't before.
1795    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1796        Windows environments %n is disabled by default.
1799    Version 7.4 21-Sep-07
1800    ---------------------
1802    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1803        means that a class such as [\s] counted as "explicit reference to CR or
1804        LF". That isn't really right - the whole point of the change was to try to
1805        help when there was an actual mention of one of the two characters. So now
1806        the change happens only if \r or \n (or a literal CR or LF) character is
1807        encountered.
1809    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1810        of both had grown to the point where there were only 3 bits left.
1811        Fortunately, there was spare space in the data structure, and so I have
1812        moved the internal flags into a new 16-bit field to free up more option
1813        bits.
1815    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1816        but did not set the internal JCHANGED flag - either of these is enough to
1817        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1818        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1819        start sets both bits.
1821    4.  Added options (at build time, compile time, exec time) to change \R from
1822        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1824    5.  doc/pcresyntax.html was missing from the distribution.
1826    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1827        compatibility, even though it is no longer used.
1829    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1830        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1831        windows.h file is present (where different names are used). [This was
1832        reversed later after testing - see 16 below.]
1834    8.  Changed all #include <config.h> to #include "config.h". There were also
1835        some further <pcre.h> cases that I changed to "pcre.h".
1837    9.  When pcregrep was used with the --colour option, it missed the line ending
1838        sequence off the lines that it output.
1840    10. It was pointed out to me that arrays of string pointers cause lots of
1841        relocations when a shared library is dynamically loaded. A technique of
1842        using a single long string with a table of offsets can drastically reduce
1843        these. I have refactored PCRE in four places to do this. The result is
1844        dramatic:
1846          Originally:                          290
1847          After changing UCP table:            187
1848          After changing error message table:   43
1849          After changing table of "verbs"       36
1850          After changing table of Posix names   22
1852        Thanks to the folks working on Gregex for glib for this insight.
1854    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1855        unicode-properties was also set.
1857    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1859    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1860        checked only for CRLF.
1862    14. Added casts to pcretest.c to avoid compiler warnings.
1864    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1866    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1867        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1868        entirely. This removes changes made in 7 above.
1870    17. The CMake files have been updated, and there is now more information about
1871        building with CMake in the NON-UNIX-USE document.
1874    Version 7.3 28-Aug-07
1875    ---------------------
1877     1. In the rejigging of the build system that eventually resulted in 7.1, the
1878        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1879        brackets there is not right, since it causes compilers to look for an
1880        installed pcre.h, not the version that is in the source that is being
1881        compiled (which of course may be different). I have changed it back to:
1883          #include "pcre.h"
1885        I have a vague recollection that the change was concerned with compiling in
1886        different directories, but in the new build system, that is taken care of
1887        by the VPATH setting the Makefile.
1889     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1890        when the subject happened to end in the byte 0x85 (e.g. if the last
1891        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1892        characters but of course it shouldn't be taken as a newline when it is part
1893        of another character. The bug was that, for an unlimited repeat of . in
1894        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1895        characters when looking for a newline.
1897     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1899     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1900        in debug output.
1902     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1903        long printing in the pcrecpp unittest when running under MinGW.
1905     6. ESC_K was left out of the EBCDIC table.
1907     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1908        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1909        limit also applies to "virtual nesting" when a pattern is recursive, and in
1910        this case 1000 isn't so big. I have been able to remove this limit at the
1911        expense of backing off one optimization in certain circumstances. Normally,
1912        when pcre_exec() would call its internal match() function recursively and
1913        immediately return the result unconditionally, it uses a "tail recursion"
1914        feature to save stack. However, when a subpattern that can match an empty
1915        string has an unlimited repetition quantifier, it no longer makes this
1916        optimization. That gives it a stack frame in which to save the data for
1917        checking that an empty string has been matched. Previously this was taken
1918        from the 1000-entry workspace that had been reserved. So now there is no
1919        explicit limit, but more stack is used.
1921     8. Applied Daniel's patches to solve problems with the import/export magic
1922        syntax that is required for Windows, and which was going wrong for the
1923        pcreposix and pcrecpp parts of the library. These were overlooked when this
1924        problem was solved for the main library.
1926     9. There were some crude static tests to avoid integer overflow when computing
1927        the size of patterns that contain repeated groups with explicit upper
1928        limits. As the maximum quantifier is 65535, the maximum group length was
1929        set at 30,000 so that the product of these two numbers did not overflow a
1930        32-bit integer. However, it turns out that people want to use groups that
1931        are longer than 30,000 bytes (though not repeat them that many times).
1932        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1933        made it possible to implement the integer overflow checks in a much more
1934        dynamic way, which I have now done. The artificial limitation on group
1935        length has been removed - we now have only the limit on the total length of
1936        the compiled pattern, which depends on the LINK_SIZE setting.
1938    10. Fixed a bug in the documentation for get/copy named substring when
1939        duplicate names are permitted. If none of the named substrings are set, the
1940        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1941        empty string.
1943    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1944        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1945        because the ] is interpreted as the first data character and the
1946        terminating ] is not found. PCRE has been made compatible with Perl in this
1947        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1948        cause memory overwriting.
1950    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1951        string has been matched (to stop an infinite loop). It was not recognizing
1952        a conditional subpattern that could match an empty string if that
1953        subpattern was within another subpattern. For example, it looped when
1954        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1955        condition was not nested. This bug has been fixed.
1957    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1958        past the start of the subject in the presence of bytes with the top bit
1959        set, for example "\x8aBCD".
1961    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1962        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1964    14. Optimized (?!) to (*FAIL).
1966    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1967        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1968        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1969        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1970        does: it's just the validity check that is more restrictive.
1972    16. Inserted checks for integer overflows during escape sequence (backslash)
1973        processing, and also fixed erroneous offset values for syntax errors during
1974        backslash processing.
1976    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1977        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1979    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1980        caused an overrun.
1982    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1983        something other than just ASCII characters) inside a group that had an
1984        unlimited repeat caused a loop at compile time (while checking to see
1985        whether the group could match an empty string).
1987    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1988        [\P{Any}] did so. (Error in the code for printing property names.)
1990    21. An orphan \E inside a character class could cause a crash.
1992    22. A repeated capturing bracket such as (A)? could cause a wild memory
1993        reference during compilation.
1995    23. There are several functions in pcre_compile() that scan along a compiled
1996        expression for various reasons (e.g. to see if it's fixed length for look
1997        behind). There were bugs in these functions when a repeated \p or \P was
1998        present in the pattern. These operators have additional parameters compared
1999        with \d, etc, and these were not being taken into account when moving along
2000        the compiled data. Specifically:
2002        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2003            length.
2005        (b) An item such as \pL+ within a repeated group could cause crashes or
2006            loops.
2008        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2009            "reference to non-existent subpattern" error.
2011        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2013    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2014        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2016    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2018    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2019        character were causing crashes (broken optimization).
2021    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2022        \p or \P) caused a compile-time loop.
2024    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2025        break. For example, the unstudied pattern [\r\n]A does not match the string
2026        "\r\nA" because change 7.0/46 below moves the current point on by two
2027        characters after failing to match at the start. However, the pattern \nA
2028        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2029        the same is true. There doesn't seem any very clean way out of this, but
2030        what I have chosen to do makes the common cases work: PCRE now takes note
2031        of whether there can be an explicit match for \r or \n anywhere in the
2032        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2033        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2034        pattern has explicit CR or LF references.
2036    29. Added (*CR) etc for changing newline setting at start of pattern.
2039    Version 7.2 19-Jun-07
2040    ---------------------
2042     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
2043        which is apparently normally available under Windows.
2045     2. Re-jig the pcregrep tests with different newline settings in an attempt
2046        to make them independent of the local environment's newline setting.
2048     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
2050     4. Some of the "internals" tests were previously cut out when the link size
2051        was not 2, because the output contained actual offsets. The recent new
2052        "Z" feature of pcretest means that these can be cut out, making the tests
2053        usable with all link sizes.
2055     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
2056        stack recursion. This gives a massive performance boost under BSD, but just
2057        a small improvement under Linux. However, it saves one field in the frame
2058        in all cases.
2060     6. Added more features from the forthcoming Perl 5.10:
2062        (a) (?-n) (where n is a string of digits) is a relative subroutine or
2063            recursion call. It refers to the nth most recently opened parentheses.
2065        (b) (?+n) is also a relative subroutine call; it refers to the nth next
2066            to be opened parentheses.
2068        (c) Conditions that refer to capturing parentheses can be specified
2069            relatively, for example, (?(-2)... or (?(+3)...
2071        (d) \K resets the start of the current match so that everything before
2072            is not part of it.
2074        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2076        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2077            reference syntax.
2079        (g) (?| introduces a group in which the numbering of parentheses in each
2080            alternative starts with the same number.
2082        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2084     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2087     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2088        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2089        for detecting groups that can match an empty string.
2091     9. A pattern with a very large number of alternatives (more than several
2092        hundred) was running out of internal workspace during the pre-compile
2093        phase, where pcre_compile() figures out how much memory will be needed. A
2094        bit of new cunning has reduced the workspace needed for groups with
2095        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2096        workspace instead of running out of the 4096 that are available.
2098    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2100    11. Applied patch from Google to remove an optimization that didn't quite work.
2101        The report of the bug said:
2103          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2104          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2105          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2107    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2108        it matched the wrong number of bytes.
2111    Version 7.1 24-Apr-07
2112    ---------------------
2114     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
2115        that is more "standard", making use of automake and other Autotools. There
2116        is some re-arrangement of the files and adjustment of comments consequent
2117        on this.
2119     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
2120        for recursive directory scanning broke on some systems because the files
2121        are not scanned in any specific order and on different systems the order
2122        was different. A call to "sort" has been inserted into RunGrepTest for the
2123        approprate test as a short-term fix. In the longer term there may be an
2124        alternative.
2126     3. I had an email from Eric Raymond about problems translating some of PCRE's
2127        man pages to HTML (despite the fact that I distribute HTML pages, some
2128        people do their own conversions for various reasons). The problems
2129        concerned the use of low-level troff macros .br and .in. I have therefore
2130        removed all such uses from the man pages (some were redundant, some could
2131        be replaced by .nf/.fi pairs). The 132html script that I use to generate
2132        HTML has been updated to handle .nf/.fi and to complain if it encounters
2133        .br or .in.
2135     4. Updated comments in configure.ac that get placed in config.h.in and also
2136        arranged for config.h to be included in the distribution, with the name
2137        config.h.generic, for the benefit of those who have to compile without
2138        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
2140     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
2141        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
2142        makevp.bat for the current PCRE, using the additional files
2143        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
2145     6. A Windows user reported a minor discrepancy with test 2, which turned out
2146        to be caused by a trailing space on an input line that had got lost in his
2147        copy. The trailing space was an accident, so I've just removed it.
2149     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
2150        that is needed.
2152     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
2153        as "const" (a) because they are and (b) because it helps the PHP
2154        maintainers who have recently made a script to detect big data structures
2155        in the php code that should be moved to the .rodata section. I remembered
2156        to update Builducptable as well, so it won't revert if ucptable.h is ever
2157        re-created.
2159     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
2160        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
2161        order to be able to cut out the UTF-8 tables in the latter when UTF-8
2162        support is not required. This saves 1.5-2K of code, which is important in
2163        some applications.
2165        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
2166        so as not to refer to the tables, even though these functions will never be
2167        called when UTF-8 support is disabled. Otherwise there are problems with a
2168        shared library.
2170    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
2172        (a) It was defining its arguments as char * instead of void *.
2174        (b) It was assuming that all moves were upwards in memory; this was true
2175            a long time ago when I wrote it, but is no longer the case.
2177        The emulated memove() is provided for those environments that have neither
2178        memmove() nor bcopy(). I didn't think anyone used it these days, but that
2179        is clearly not the case, as these two bugs were recently reported.
2181    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
2182        and Detrail to create the HTML documentation, the .txt form of the man
2183        pages, and it removes trailing spaces from listed files. It also creates
2184        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
2185        case, it wraps all the #defines with #ifndefs. This script should be run
2186        before "make dist".
2188    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
2189        with Unicode property support.
2191        (a) For a maximizing quantifier, if the two different cases of the
2192            character were of different lengths in their UTF-8 codings (there are
2193            some cases like this - I found 11), and the matching function had to
2194            back up over a mixture of the two cases, it incorrectly assumed they
2195            were both the same length.
2197        (b) When PCRE was configured to use the heap rather than the stack for
2198            recursion during matching, it was not correctly preserving the data for
2199            the other case of a UTF-8 character when checking ahead for a match
2200            while processing a minimizing repeat. If the check also involved
2201            matching a wide character, but failed, corruption could cause an
2202            erroneous result when trying to check for a repeat of the original
2203            character.
2205    13. Some tidying changes to the testing mechanism:
2207        (a) The RunTest script now detects the internal link size and whether there
2208            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
2209            values substituted by "configure". (The RunGrepTest script already did
2210            this for UTF-8.) The configure.ac script no longer substitutes the
2211            relevant variables.
2213        (b) The debugging options /B and /D in pcretest show the compiled bytecode
2214            with length and offset values. This means that the output is different
2215            for different internal link sizes. Test 2 is skipped for link sizes
2216            other than 2 because of this, bypassing the problem. Unfortunately,
2217            there was also a test in test 3 (the locale tests) that used /B and
2218            failed for link sizes other than 2. Rather than cut the whole test out,
2219            I have added a new /Z option to pcretest that replaces the length and
2220            offset values with spaces. This is now used to make test 3 independent
2221            of link size. (Test 2 will be tidied up later.)
2223    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
2224        segmentation fault instead of returning the appropriate error message.
2226    15. In multiline mode when the newline sequence was set to "any", the pattern
2227        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
2228        This doesn't seem right; it now treats the CRLF combination as the line
2229        ending, and so does not match in that case. It's only a pattern such as ^$
2230        that would hit this one: something like ^ABC$ would have failed after \r
2231        and then tried again after \r\n.
2233    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
2234        in an attempt to make files that differ only in their line terminators
2235        compare equal. This works on Linux.
2237    17. Under certain error circumstances pcregrep might try to free random memory
2238        as it exited. This is now fixed, thanks to valgrind.
2240    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
2241        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
2242        was because its rules for how to advance for /g after matching an empty
2243        string at the end of a line did not allow for this case. They now check for
2244        it specially.
2246    20. pcretest is supposed to handle patterns and data of any length, by
2247        extending its buffers when necessary. It was getting this wrong when the
2248        buffer for a data line had to be extended.
2250    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
2251        CRLF as a newline sequence.
2253    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
2254        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
2255        I have nevertheless tidied it up.
2257    23. Added some casts to kill warnings from HP-UX ia64 compiler.
2259    24. Added a man page for pcre-config.
2262    Version 7.0 19-Dec-06
2263    ---------------------
2265     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
2266        moving to gcc 4.1.1.
2268     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
2269        sys/time.h, which is documented as needed for this function. It doesn't
2270        seem to matter on Linux, but it showed up on some releases of OS X.
2272     3. It seems that there are systems where bytes whose values are greater than
2273        127 match isprint() in the "C" locale. The "C" locale should be the
2274        default when a C program starts up. In most systems, only ASCII printing
2275        characters match isprint(). This difference caused the output from pcretest
2276        to vary, making some of the tests fail. I have changed pcretest so that:
2278        (a) When it is outputting text in the compiled version of a pattern, bytes
2279            other than 32-126 are always shown as hex escapes.
2281        (b) When it is outputting text that is a matched part of a subject string,
2282            it does the same, unless a different locale has been set for the match
2283            (using the /L modifier). In this case, it uses isprint() to decide.
2285     4. Fixed a major bug that caused incorrect computation of the amount of memory
2286        required for a compiled pattern when options that changed within the
2287        pattern affected the logic of the preliminary scan that determines the
2288        length. The relevant options are -x, and -i in UTF-8 mode. The result was
2289        that the computed length was too small. The symptoms of this bug were
2290        either the PCRE error "internal error: code overflow" from pcre_compile(),
2291        or a glibc crash with a message such as "pcretest: free(): invalid next
2292        size (fast)". Examples of patterns that provoked this bug (shown in
2293        pcretest format) are:
2295          /(?-x: )/x
2296          /(?x)(?-x: \s*#\s*)/
2297          /((?i)[\x{c0}])/8
2298          /(?i:[\x{c0}])/8
2300        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
2301        is now done differently.
2303     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
2304        wrapper classes; (b) implement a new function in the C++ scanner that is
2305        more efficient than the old way of doing things because it avoids levels of
2306        recursion in the regex matching; (c) add a paragraph to the documentation
2307        for the FullMatch() function.
2309     6. The escape sequence \n was being treated as whatever was defined as
2310        "newline". Not only was this contrary to the documentation, which states
2311        that \n is character 10 (hex 0A), but it also went horribly wrong when
2312        "newline" was defined as CRLF. This has been fixed.
2314     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
2315        was being set to -1 for the "end of line" case (supposedly a value that no
2316        character can have). Though this value is never used (the check for end of
2317        line is "zero bytes in current character"), it caused compiler complaints.
2318        I've changed it to 0xffffffff.
2320     8. In pcre_version.c, the version string was being built by a sequence of
2321        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
2322        string (as it is for production releases) called a macro with an empty
2323        argument. The C standard says the result of this is undefined. The gcc
2324        compiler treats it as an empty string (which was what was wanted) but it is
2325        reported that Visual C gives an error. The source has been hacked around to
2326        avoid this problem.
2328     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
2329        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
2330        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
2331        of them did).
2333    10. Originally, pcretest opened its input and output without "b"; then I was
2334        told that "b" was needed in some environments, so it was added for release
2335        5.0 to both the input and output. (It makes no difference on Unix-like
2336        systems.) Later I was told that it is wrong for the input on Windows. I've
2337        now abstracted the modes into two macros, to make it easier to fiddle with
2338        them, and removed "b" from the input mode under Windows.
2340    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
2342    12. Added -help and --help to pcretest as an official way of being reminded
2343        of the options.
2345    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
2346        and pcrecpp.cc because they annoy compilers at high warning levels.
2348    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
2350    15. Fixed an occurrence of == in configure.ac that should have been = (shell
2351        scripts are not C programs :-) and which was not noticed because it works
2352        on Linux.
2354    16. pcretest is supposed to handle any length of pattern and data line (as one
2355        line or as a continued sequence of lines) by extending its input buffer if
2356        necessary. This feature was broken for very long pattern lines, leading to
2357        a string of junk being passed to pcre_compile() if the pattern was longer
2358        than about 50K.
2360    17. I have done a major re-factoring of the way pcre_compile() computes the
2361        amount of memory needed for a compiled pattern. Previously, there was code
2362        that made a preliminary scan of the pattern in order to do this. That was
2363        OK when PCRE was new, but as the facilities have expanded, it has become
2364        harder and harder to keep it in step with the real compile phase, and there
2365        have been a number of bugs (see for example, 4 above). I have now found a
2366        cunning way of running the real compile function in a "fake" mode that
2367        enables it to compute how much memory it would need, while actually only
2368        ever using a few hundred bytes of working memory and without too many
2369        tests of the mode. This should make future maintenance and development
2370        easier. A side effect of this work is that the limit of 200 on the nesting
2371        depth of parentheses has been removed (though this was never a serious
2372        limitation, I suspect). However, there is a downside: pcre_compile() now
2373        runs more slowly than before (30% or more, depending on the pattern). I
2374        hope this isn't a big issue. There is no effect on runtime performance.
2376    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
2377        newline (only possible for the last line of a file) and it was a
2378        pattern that set a locale (followed by /Lsomething), pcretest crashed.
2380    19. Added additional timing features to pcretest. (1) The -tm option now times
2381        matching only, not compiling. (2) Both -t and -tm can be followed, as a
2382        separate command line item, by a number that specifies the number of
2383        repeats to use when timing. The default is 50000; this gives better
2384        precision, but takes uncomfortably long for very large patterns.
2386    20. Extended pcre_study() to be more clever in cases where a branch of a
2387        subpattern has no definite first character. For example, (a*|b*)[cd] would
2388        previously give no result from pcre_study(). Now it recognizes that the
2389        first character must be a, b, c, or d.
2391    21. There was an incorrect error "recursive call could loop indefinitely" if
2392        a subpattern (or the entire pattern) that was being tested for matching an
2393        empty string contained only one non-empty item after a nested subpattern.
2394        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
2395        incorrectly, because the \d was being skipped in the check.
2397    22. The pcretest program now has a new pattern option /B and a command line
2398        option -b, which is equivalent to adding /B to every pattern. This causes
2399        it to show the compiled bytecode, without the additional information that
2400        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
2401        is the same as /B/I).
2403    23. A new optimization is now able automatically to treat some sequences such
2404        as a*b as a*+b. More specifically, if something simple (such as a character
2405        or a simple class like \d) has an unlimited quantifier, and is followed by
2406        something that cannot possibly match the quantified thing, the quantifier
2407        is automatically "possessified".
2409    24. A recursive reference to a subpattern whose number was greater than 39
2410        went wrong under certain circumstances in UTF-8 mode. This bug could also
2411        have affected the operation of pcre_study().
2413    25. Realized that a little bit of performance could be had by replacing
2414        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
2416    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
2418    27. Possessive quantifiers such as a++ were previously implemented by turning
2419        them into atomic groups such as ($>a+). Now they have their own opcodes,
2420        which improves performance. This includes the automatically created ones
2421        from 23 above.
2423    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
2424        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
2425        the first matched character to be a colon. This applied both to named and
2426        numbered groups.
2428    29. The ucpinternal.h header file was missing its idempotency #ifdef.
2430    30. I was sent a "project" file called libpcre.a.dev which I understand makes
2431        building PCRE on Windows easier, so I have included it in the distribution.
2433    31. There is now a check in pcretest against a ridiculously large number being
2434        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
2435        loop, the loop is abandoned.
2437    32. Forward references to subpatterns in conditions such as (?(2)...) where
2438        subpattern 2 is defined later cause pcre_compile() to search forwards in
2439        the pattern for the relevant set of parentheses. This search went wrong
2440        when there were unescaped parentheses in a character class, parentheses
2441        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2443    33. "Subroutine" calls and backreferences were previously restricted to
2444        referencing subpatterns earlier in the regex. This restriction has now
2445        been removed.
2447    34. Added a number of extra features that are going to be in Perl 5.10. On the
2448        whole, these are just syntactic alternatives for features that PCRE had
2449        previously implemented using the Python syntax or my own invention. The
2450        other formats are all retained for compatibility.
2452        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2453            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2454            also .NET compatible.
2456        (b) A recursion or subroutine call to a named group can now be defined as
2457            (?&name) as well as (?P>name).
2459        (c) A backreference to a named group can now be defined as \k<name> or
2460            \k'name' as well as (?P=name). The new forms, as well as being in Perl
2461            5.10, are also .NET compatible.
2463        (d) A conditional reference to a named group can now use the syntax
2464            (?(<name>) or (?('name') as well as (?(name).
2466        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2467            groups (named and numbered) that are never evaluated inline, but can be
2468            called as "subroutines" from elsewhere. In effect, the DEFINE condition
2469            is always false. There may be only one alternative in such a group.
2471        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2472            as the simple (?(R). The condition is true only if the most recent
2473            recursion is that of the given number or name. It does not search out
2474            through the entire recursion stack.
2476        (g) The escape \gN or \g{N} has been added, where N is a positive or
2477            negative number, specifying an absolute or relative reference.
2479    35. Tidied to get rid of some further signed/unsigned compiler warnings and
2480        some "unreachable code" warnings.
2482    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2483        things, this adds five new scripts.
2485    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2486        There were also incompatibilities regarding the handling of \Q..\E inside
2487        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2488        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2490    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2491        matches an empty string, and forcibly breaks the loop. There were bugs in
2492        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2493        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2494        separate and independent bugs (that affected different cases) have been
2495        fixed.
2497    39. Refactored the code to abolish the use of different opcodes for small
2498        capturing bracket numbers. This is a tidy that I avoided doing when I
2499        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2500        The new approach is not only tidier, it makes it possible to reduce the
2501        memory needed to fix the previous bug (38).
2503    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2504        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2505        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2506        mode.
2508    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2509        report.
2511    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2512        copy construction and assignment in the C++ wrapper.
2514    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2515        couple of bugs that could have given wrong results in the "--newline=crlf"
2516        case.
2518    44. Added a number of casts and did some reorganization of signed/unsigned int
2519        variables following suggestions from Dair Grant. Also renamed the variable
2520        "this" as "item" because it is a C++ keyword.
2522    45. Arranged for dftables to add
2524          #include "pcre_internal.h"
2526        to pcre_chartables.c because without it, gcc 4.x may remove the array
2527        definition from the final binary if PCRE is built into a static library and
2528        dead code stripping is activated.
2530    46. For an unanchored pattern, if a match attempt fails at the start of a
2531        newline sequence, and the newline setting is CRLF or ANY, and the next two
2532        characters are CRLF, advance by two characters instead of one.
2535    Version 6.7 04-Jul-06
2536    ---------------------
2538     1. In order to handle tests when input lines are enormously long, pcretest has
2539        been re-factored so that it automatically extends its buffers when
2540        necessary. The code is crude, but this _is_ just a test program. The
2541        default size has been increased from 32K to 50K.
2543     2. The code in pcre_study() was using the value of the re argument before
2544        testing it for NULL. (Of course, in any sensible call of the function, it
2545        won't be NULL.)
2547     3. The memmove() emulation function in pcre_internal.h, which is used on
2548        systems that lack both memmove() and bcopy() - that is, hardly ever -
2549        was missing a "static" storage class specifier.
2551     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2552        containing an extended class (one that cannot be represented by a bitmap
2553        because it contains high-valued characters or Unicode property items, e.g.
2554        [\pZ]). Almost always one would set UTF-8 mode when processing such a
2555        pattern, but PCRE should not loop if you do not (it no longer does).
2556        [Detail: two cases were found: (a) a repeated subpattern containing an
2557        extended class; (b) a recursive reference to a subpattern that followed a
2558        previous extended class. It wasn't skipping over the extended class
2559        correctly when UTF-8 mode was not set.]
2561     5. A negated single-character class was not being recognized as fixed-length
2562        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2563        compile error "lookbehind assertion is not fixed length".
2565     6. The RunPerlTest auxiliary script was showing an unexpected difference
2566        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2567        write a Perl script that can interpret lines of an input file either as
2568        byte characters or as UTF-8, which is what "perltest" was being required to
2569        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2570        can't do is switch easily at run time between having the "use utf8;" pragma
2571        or not. In the end, I fudged it by using the RunPerlTest script to insert
2572        "use utf8;" explicitly for the UTF-8 tests.
2574     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2575        the end of the subject string, contrary to the documentation and to what
2576        Perl does. This was true of both matching functions. Now it matches only at
2577        the start of the subject and immediately after *internal* newlines.
2579     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2580        a pointer to an int instead of a pointer to an unsigned long int. This
2581        caused problems on 64-bit systems.
2583     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2584        instance of the 'standard' template library not being so standard".
2586    10. There was no check on the number of named subpatterns nor the maximum
2587        length of a subpattern name. The product of these values is used to compute
2588        the size of the memory block for a compiled pattern. By supplying a very
2589        long subpattern name and a large number of named subpatterns, the size
2590        computation could be caused to overflow. This is now prevented by limiting
2591        the length of names to 32 characters, and the number of named subpatterns
2592        to 10,000.
2594    11. Subpatterns that are repeated with specific counts have to be replicated in
2595        the compiled pattern. The size of memory for this was computed from the
2596        length of the subpattern and the repeat count. The latter is limited to
2597        65535, but there was no limit on the former, meaning that integer overflow
2598        could in principle occur. The compiled length of a repeated subpattern is
2599        now limited to 30,000 bytes in order to prevent this.
2601    12. Added the optional facility to have named substrings with the same name.
2603    13. Added the ability to use a named substring as a condition, using the
2604        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2605        are numbers (not recommended). Forward references are permitted.
2607    14. Added forward references in named backreferences (if you see what I mean).
2609    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2610        pattern could run off the end of the subject. For example, the pattern
2611        "(?s)(.{1,5})"8 did this with the subject "ab".
2613    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2614        PCRE_CASELESS was set when matching characters that were quantified with ?
2615        or *.
2617    17. A character class other than a single negated character that had a minimum
2618        but no maximum quantifier - for example [ab]{6,} - was not handled
2619        correctly by pce_dfa_exec(). It would match only one character.
2621    18. A valid (though odd) pattern that looked like a POSIX character
2622        class but used an invalid character after [ (for example [[,abc,]]) caused
2623        pcre_compile() to give the error "Failed: internal error: code overflow" or
2624        in some cases to crash with a glibc free() error. This could even happen if
2625        the pattern terminated after [[ but there just happened to be a sequence of
2626        letters, a binary zero, and a closing ] in the memory that followed.
2628    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2629        over the years. Originally (before any Unicode support), just the bottom 8
2630        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2631        output from "man perlunicode" includes this:
2633          The regular expression compiler produces polymorphic opcodes.  That
2634          is, the pattern adapts to the data and automatically switches to
2635          the Unicode character scheme when presented with Unicode data--or
2636          instead uses a traditional byte scheme when presented with byte
2637          data.
2639        Sadly, a wide octal escape does not cause a switch, and in a string with
2640        no other multibyte characters, these octal escapes are treated as before.
2641        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2642        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2643        Unicode string.
2645        I have not perpetrated such confusion in PCRE. Up till now, it took just
2646        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2647        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2648        translate to the appropriate multibyte character.
2650    29. Applied some refactoring to reduce the number of warnings from Microsoft
2651        and Borland compilers. This has included removing the fudge introduced
2652        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2653        a warning about an unused variable.
2655    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2656        characters since release 4.0, because Perl (from release 5.004) does not.
2657        [Or at least, is documented not to: some releases seem to be in conflict
2658        with the documentation.] However, when a pattern was studied with
2659        pcre_study() and all its branches started with \s, PCRE still included VT
2660        as a possible starting character. Of course, this did no harm; it just
2661        caused an unnecessary match attempt.
2663    22. Removed a now-redundant internal flag bit that recorded the fact that case
2664        dependency changed within the pattern. This was once needed for "required
2665        byte" processing, but is no longer used. This recovers a now-scarce options
2666        bit. Also moved the least significant internal flag bit to the most-
2667        significant bit of the word, which was not previously used (hangover from
2668        the days when it was an int rather than a uint) to free up another bit for
2669        the future.
2671    23. Added support for CRLF line endings as well as CR and LF. As well as the
2672        default being selectable at build time, it can now be changed at runtime
2673        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2674        specify that it is scanning data with non-default line endings.
2676    24. Changed the definition of CXXLINK to make it agree with the definition of
2677        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2679    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2680        recursions. This makes a big different to stack usage for some patterns.
2682    26. If a subpattern containing a named recursion or subroutine reference such
2683        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2684        the space required for the compiled pattern went wrong and gave too small a
2685        value. Depending on the environment, this could lead to "Failed: internal
2686        error: code overflow at offset 49" or "glibc detected double free or
2687        corruption" errors.
2689    27. Applied patches from Google (a) to support the new newline modes and (b) to
2690        advance over multibyte UTF-8 characters in GlobalReplace.
2692    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2693        difference for some implementation of PCRE in some Windows version.
2695    29. Added some extra testing facilities to pcretest:
2697        \q<number>   in a data line sets the "match limit" value
2698        \Q<number>   in a data line sets the "match recursion limt" value
2699        -S <number>  sets the stack size, where <number> is in megabytes
2701        The -S option isn't available for Windows.
2704    Version 6.6 06-Feb-06
2705    ---------------------
2707     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2708        in pcreposix.h. I have copied the definition from pcre.h.
2710     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2711        because pcre.h is no longer a built file.
2713     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2714        not normally included in the compiled code.
2717    Version 6.5 01-Feb-06
2718    ---------------------
2720     1. When using the partial match feature with pcre_dfa_exec(), it was not
2721        anchoring the second and subsequent partial matches at the new starting
2722        point. This could lead to incorrect results. For example, with the pattern
2723        /1234/, partially matching against "123" and then "a4" gave a match.
2725     2. Changes to pcregrep:
2727        (a) All non-match returns from pcre_exec() were being treated as failures
2728            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2729            error message is output. Some extra information is given for the
2730            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2731            probably the only errors that are likely to be caused by users (by
2732            specifying a regex that has nested indefinite repeats, for instance).
2733            If there are more than 20 of these errors, pcregrep is abandoned.
2735        (b) A binary zero was treated as data while matching, but terminated the
2736            output line if it was written out. This has been fixed: binary zeroes
2737            are now no different to any other data bytes.
2739        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2740            used to set a locale for matching. The --locale=xxxx long option has
2741            been added (no short equivalent) to specify a locale explicitly on the
2742            pcregrep command, overriding the environment variables.
2744        (d) When -B was used with -n, some line numbers in the output were one less
2745            than they should have been.
2747        (e) Added the -o (--only-matching) option.
2749        (f) If -A or -C was used with -c (count only), some lines of context were
2750            accidentally printed for the final match.
2752        (g) Added the -H (--with-filename) option.
2754        (h) The combination of options -rh failed to suppress file names for files
2755            that were found from directory arguments.
2757        (i) Added the -D (--devices) and -d (--directories) options.
2759        (j) Added the -F (--fixed-strings) option.
2761        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2763        (l) Added the --colo(u)r option.
2765        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2766            is not present by default.
2768     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2769        items such as (?R) or (?1), when the recursion could match a number of
2770        alternatives. If it matched one of the alternatives, but subsequently,
2771        outside the recursion, there was a failure, the code tried to back up into
2772        the recursion. However, because of the way PCRE is implemented, this is not
2773        possible, and the result was an incorrect result from the match.
2775        In order to prevent this happening, the specification of recursion has
2776        been changed so that all such subpatterns are automatically treated as
2777        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2779     4. I had overlooked the fact that, in some locales, there are characters for
2780        which isalpha() is true but neither isupper() nor islower() are true. In
2781        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2782        and ordfeminine) are like this. This affected the treatment of \w and \W
2783        when they appeared in character classes, but not when they appeared outside
2784        a character class. The bit map for "word" characters is now created
2785        separately from the results of isalnum() instead of just taking it from the
2786        upper, lower, and digit maps. (Plus the underscore character, of course.)
2788     5. The above bug also affected the handling of POSIX character classes such as
2789        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2790        permanent tables. Instead, the bit maps for such a class were previously
2791        created as the appropriate unions of the upper, lower, and digit bitmaps.
2792        Now they are created by subtraction from the [[:word:]] class, which has
2793        its own bitmap.
2795     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2796        It is created by subtracting the vertical space characters (\x09, \x0a,
2797        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2798        subtraction was done in the overall bitmap for a character class, meaning
2799        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2800        be recognized. This bug has been fixed.
2802     7. Patches from the folks at Google:
2804          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2805          real life, but is still worth protecting against".
2807          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2808          regular expressions".
2810          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2811          have it.
2813          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2814          "configure" and the latter not, in order to fix a problem somebody had
2815          with compiling the Arg class on HP-UX.
2817          (e) Improve the error-handling of the C++ wrapper a little bit.
2819          (f) New tests for checking recursion limiting.
2821     8. The pcre_memmove() function, which is used only if the environment does not
2822        have a standard memmove() function (and is therefore rarely compiled),
2823        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2824        returning a result (though PCRE never actually uses the result).
2826     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2827        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2828        returned instead of calling malloc() with an overflowing number that would
2829        most likely cause subsequent chaos.
2831    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2833    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2834        with this option is matched, the nmatch and pmatch options of regexec() are
2835        ignored.
2837    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2838        provided in case anyone wants to the the POSIX interface with UTF-8
2839        strings.
2841    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2842        C++ linking (needed for some HP-UX environments).
2844    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2845        (unused parameter) and in the pcre_printint() function (omitted "default"
2846        switch label when the default is to do nothing).
2848    15. Added some code to make it possible, when PCRE is compiled as a C++
2849        library, to replace subject pointers for pcre_exec() with a smart pointer
2850        class, thus making it possible to process discontinuous strings.
2852    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2853        much the same function. They were added by different people who were trying
2854        to make PCRE easy to compile on non-Unix systems. It has been suggested
2855        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2856        for compiling on Windows systems. I have therefore replaced it with
2857        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2858        defaults to "extern" for C or "extern C" for C++, which works fine on
2859        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2860        SCOPE with something explicit in config.h. In addition:
2862        (a) pcreposix.h still had just "extern" instead of either of these macros;
2863            I have replaced it with PCRE_DATA_SCOPE.
2865        (b) Functions such as _pcre_xclass(), which are internal to the library,
2866            but external in the C sense, all had PCRE_EXPORT in their definitions.
2867            This is apparently wrong for the Windows case, so I have removed it.
2868            (It makes no difference on Unix-like systems.)
2870    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2871        of recursive calls to match(). This is different to MATCH_LIMIT because
2872        that limits the total number of calls to match(), not all of which increase
2873        the depth of recursion. Limiting the recursion depth limits the amount of
2874        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2875        when PCRE is compiled, and changed at run time. A patch from Google adds
2876        this functionality to the C++ interface.
2878    18. Changes to the handling of Unicode character properties:
2880        (a) Updated the table to Unicode 4.1.0.
2882        (b) Recognize characters that are not in the table as "Cn" (undefined).
2884        (c) I revised the way the table is implemented to a much improved format
2885            which includes recognition of ranges. It now supports the ranges that
2886            are defined in UnicodeData.txt, and it also amalgamates other
2887            characters into ranges. This has reduced the number of entries in the
2888            table from around 16,000 to around 3,000, thus reducing its size
2889            considerably. I realized I did not need to use a tree structure after
2890            all - a binary chop search is just as efficient. Having reduced the
2891            number of entries, I extended their size from 6 bytes to 8 bytes to
2892            allow for more data.
2894        (d) Added support for Unicode script names via properties such as \p{Han}.
2896    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2897        matching that character.
2899    20. When matching a repeated Unicode property with a minimum greater than zero,
2900        (for example \pL{2,}), PCRE could look past the end of the subject if it
2901        reached it while seeking the minimum number of characters. This could
2902        happen only if some of the characters were more than one byte long, because
2903        there is a check for at least the minimum number of bytes.
2905    21. Refactored the implementation of \p and \P so as to be more general, to
2906        allow for more different types of property in future. This has changed the
2907        compiled form incompatibly. Anybody with saved compiled patterns that use
2908        \p or \P will have to recompile them.
2910    22. Added "Any" and "L&" to the supported property types.
2912    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2913        but give a compile time error if the value is greater than 0xff.
2915    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2916        accidentally not being installed or uninstalled.
2918    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2919        made were to insert the current release number. This seemed silly, because
2920        it made things harder for people building PCRE on systems that don't run
2921        "configure". I have turned pcre.h into a distributed file, no longer built
2922        by "configure", with the version identification directly included. There is
2923        no longer a pcre.h.in file.
2925        However, this change necessitated a change to the pcre-config script as
2926        well. It is built from pcre-config.in, and one of the substitutions was the
2927        release number. I have updated configure.ac so that ./configure now finds
2928        the release number by grepping pcre.h.
2930    26. Added the ability to run the tests under valgrind.
2933    Version 6.4 05-Sep-05
2934    ---------------------
2936     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2937        "--" to be printed when multiple files were scanned, even when none of the
2938        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2939        consider it to be a bug, and have restored the previous behaviour.
2941     2. A couple of code tidies to get rid of compiler warnings.
2943     3. The pcretest program used to cheat by referring to symbols in the library
2944        whose names begin with _pcre_. These are internal symbols that are not
2945        really supposed to be visible externally, and in some environments it is
2946        possible to suppress them. The cheating is now confined to including
2947        certain files from the library's source, which is a bit cleaner.
2949     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2950        file's purpose clearer.
2952     5. Reorganized pcre_ucp_findchar().
2955    Version 6.3 15-Aug-05
2956    ---------------------
2958     1. The file libpcre.pc.in did not have general read permission in the tarball.
2960     2. There were some problems when building without C++ support:
2962        (a) If C++ support was not built, "make install" and "make test" still
2963            tried to test it.
2965        (b) There were problems when the value of CXX was explicitly set. Some
2966            changes have been made to try to fix these, and ...
2968        (c) --disable-cpp can now be used to explicitly disable C++ support.
2970        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2971            backslash in a target when C++ was disabled. This confuses some
2972            versions of "make", apparently. Using an intermediate variable solves
2973            this. (Same for CPP_LOBJ.)
2975     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2976        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2977        necessary on certain architectures.
2979     4. Added a setting of -export-symbols-regex to the link command to remove
2980        those symbols that are exported in the C sense, but actually are local
2981        within the library, and not documented. Their names all begin with
2982        "_pcre_". This is not a perfect job, because (a) we have to except some
2983        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2984        available (and never for static libraries). I have made a note to try to
2985        find a way round (a) in the future.
2988    Version 6.2 01-Aug-05
2989    ---------------------
2991     1. There was no test for integer overflow of quantifier values. A construction
2992        such as {1111111111111111} would give undefined results. What is worse, if
2993        a minimum quantifier for a parenthesized subpattern overflowed and became
2994        negative, the calculation of the memory size went wrong. This could have
2995        led to memory overwriting.
2997     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2999     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
3000        operating environments where this matters.
3002     4. Applied Giuseppe Maxia's patch to add additional features for controlling
3003        PCRE options from within the C++ wrapper.
3005     5. Named capturing subpatterns were not being correctly counted when a pattern
3006        was compiled. This caused two problems: (a) If there were more than 100
3007        such subpatterns, the calculation of the memory needed for the whole
3008        compiled pattern went wrong, leading to an overflow error. (b) Numerical
3009        back references of the form \12, where the number was greater than 9, were
3010        not recognized as back references, even though there were sufficient
3011        previous subpatterns.
3013     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
3014        versions of gcc, e.g. 2.95.4.
3017    Version 6.1 21-Jun-05
3018    ---------------------
3020     1. There was one reference to the variable "posix" in pcretest.c that was not
3021        surrounded by "#if !defined NOPOSIX".
3023     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
3024        the cross-check on the old pcre_info() function, for the benefit of the
3025        cut-down version of PCRE that is currently imported into Exim.
3027     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
3028        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
3029        patterns that start (?i) but I don't think that matters. The use of (?i) is
3030        just an example; this all applies to the other options as well.
3032     4. Since libtool seems to echo the compile commands it is issuing, the output
3033        from "make" can be reduced a bit by putting "@" in front of each libtool
3034        compile command.
3036     5. Patch from the folks at Google for configure.in to be a bit more thorough
3037        in checking for a suitable C++ installation before trying to compile the
3038        C++ stuff. This should fix a reported problem when a compiler was present,
3039        but no suitable headers.
3041     6. The man pages all had just "PCRE" as their title. I have changed them to
3042        be the relevant file name. I have also arranged that these names are
3043        retained in the file doc/pcre.txt, which is a concatenation in text format
3044        of all the man pages except the little individual ones for each function.
3046     7. The NON-UNIX-USE file had not been updated for the different set of source
3047        files that come with release 6. I also added a few comments about the C++
3048        wrapper.
3051    Version 6.0 07-Jun-05
3052    ---------------------
3054     1. Some minor internal re-organization to help with my DFA experiments.
3056     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
3057        didn't matter for the library itself when fully configured, but did matter
3058        when compiling without UCP support, or within Exim, where the ucp files are
3059        not imported.
3061     3. Refactoring of the library code to split up the various functions into
3062        different source modules. The addition of the new DFA matching code (see
3063        below) to a single monolithic source would have made it really too
3064        unwieldy, quite apart from causing all the code to be include in a
3065        statically linked application, when only some functions are used. This is
3066        relevant even without the DFA addition now that patterns can be compiled in
3067        one application and matched in another.
3069        The downside of splitting up is that there have to be some external
3070        functions and data tables that are used internally in different modules of
3071        the library but which are not part of the API. These have all had their
3072        names changed to start with "_pcre_" so that they are unlikely to clash
3073        with other external names.
3075     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
3076        a different (DFA) algorithm. Although it is slower than the original
3077        function, it does have some advantages for certain types of matching
3078        problem.
3080     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
3081        including restarting after a partial match.
3083     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
3084        defined when compiling for Windows was sent to me. I have put it into the
3085        code, though I have no means of testing or verifying it.
3087     7. Added the pcre_refcount() auxiliary function.
3089     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
3090        match before or at the first newline in the subject string. In pcretest,
3091        the /f option on a pattern can be used to set this.
3093     9. A repeated \w when used in UTF-8 mode with characters greater than 256
3094        would behave wrongly. This has been present in PCRE since release 4.0.
3096    10. A number of changes to the pcregrep command:
3098        (a) Refactored how -x works; insert ^(...)$ instead of setting
3099            PCRE_ANCHORED and checking the length, in preparation for adding
3100            something similar for -w.
3102        (b) Added the -w (match as a word) option.
3104        (c) Refactored the way lines are read and buffered so as to have more
3105            than one at a time available.
3107        (d) Implemented a pcregrep test script.
3109        (e) Added the -M (multiline match) option. This allows patterns to match
3110            over several lines of the subject. The buffering ensures that at least
3111            8K, or the rest of the document (whichever is the shorter) is available
3112            for matching (and similarly the previous 8K for lookbehind assertions).
3114        (f) Changed the --help output so that it now says
3116              -w, --word-regex(p)
3118            instead of two lines, one with "regex" and the other with "regexp"
3119            because that confused at least one person since the short forms are the
3120            same. (This required a bit of code, as the output is generated
3121            automatically from a table. It wasn't just a text change.)
3123        (g) -- can be used to terminate pcregrep options if the next thing isn't an
3124            option but starts with a hyphen. Could be a pattern or a path name
3125            starting with a hyphen, for instance.
3127        (h) "-" can be given as a file name to represent stdin.
3129        (i) When file names are being printed, "(standard input)" is used for
3130            the standard input, for compatibility with GNU grep. Previously
3131            "<stdin>" was used.
3133        (j) The option --label=xxx can be used to supply a name to be used for
3134            stdin when file names are being printed. There is no short form.
3136        (k) Re-factored the options decoding logic because we are going to add
3137            two more options that take data. Such options can now be given in four
3138            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
3140        (l) Added the -A, -B, and -C options for requesting that lines of context
3141            around matches be printed.
3143        (m) Added the -L option to print the names of files that do not contain
3144            any matching lines, that is, the complement of -l.
3146        (n) The return code is 2 if any file cannot be opened, but pcregrep does
3147            continue to scan other files.
3149        (o) The -s option was incorrectly implemented. For compatibility with other
3150            greps, it now suppresses the error message for a non-existent or non-
3151            accessible file (but not the return code). There is a new option called
3152            -q that suppresses the output of matching lines, which was what -s was
3153            previously doing.
3155        (p) Added --include and --exclude options to specify files for inclusion
3156            and exclusion when recursing.
3158    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
3159        Hopefully, it now does.
3161    12. Missing cast in pcre_study().
3163    13. Added an "uninstall" target to the makefile.
3165    14. Replaced "extern" in the function prototypes in Makefile.in with
3166        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
3167        world, but is set differently for Windows.
3169    15. Added a second compiling function called pcre_compile2(). The only
3170        difference is that it has an extra argument, which is a pointer to an
3171        integer error code. When there is a compile-time failure, this is set
3172        non-zero, in addition to the error test pointer being set to point to an
3173        error message. The new argument may be NULL if no error number is required
3174        (but then you may as well call pcre_compile(), which is now just a
3175        wrapper). This facility is provided because some applications need a
3176        numeric error indication, but it has also enabled me to tidy up the way
3177        compile-time errors are handled in the POSIX wrapper.
3179    16. Added VPATH=.libs to the makefile; this should help when building with one
3180        prefix path and installing with another. (Or so I'm told by someone who
3181        knows more about this stuff than I do.)
3183    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
3184        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
3185        match everything, including newlines. This is not POSIX-compatible, but
3186        somebody wanted the feature. From pcretest it can be activated by using
3187        both the P and the s flags.
3189    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
3191    19. libpcre.pc was being incorrectly installed as executable.
3193    20. A couple of places in pcretest check for end-of-line by looking for '\n';
3194        it now also looks for '\r' so that it will work unmodified on Windows.
3196    21. Added Google's contributed C++ wrapper to the distribution.
3198    22. Added some untidy missing memory free() calls in pcretest, to keep
3199        Electric Fence happy when testing.
3203    Version 5.0 13-Sep-04
3204    ---------------------
3206     1. Internal change: literal characters are no longer packed up into items
3207        containing multiple characters in a single byte-string. Each character
3208        is now matched using a separate opcode. However, there may be more than one
3209        byte in the character in UTF-8 mode.
3211     2. The pcre_callout_block structure has two new fields: pattern_position and
3212        next_item_length. These contain the offset in the pattern to the next match
3213        item, and its length, respectively.
3215     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
3216        insertion of callouts before each pattern item. Added the /C option to
3217        pcretest to make use of this.
3219     4. On the advice of a Windows user, the lines
3221          #if defined(_WIN32) || defined(WIN32)
3222          _setmode( _fileno( stdout ), 0x8000 );
3223          #endif  /* defined(_WIN32) || defined(WIN32) */
3225        have been added to the source of pcretest. This apparently does useful
3226        magic in relation to line terminators.
3228     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
3229        for the benefit of those environments where the "b" makes a difference.
3231     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
3232        to know about it. I have put a hack into configure.in that adds in code
3233        to set GCC=yes if CC=icc. This seems to end up at a point in the
3234        generated configure script that is early enough to affect the setting of
3235        compiler options, which is what is needed, but I have no means of testing
3236        whether it really works. (The user who reported this had patched the
3237        generated configure script, which of course I cannot do.)
3239        LATER: After change 22 below (new libtool files), the configure script
3240        seems to know about icc (and also ecc). Therefore, I have commented out
3241        this hack in configure.in.
3243     7. Added support for pkg-config (2 patches were sent in).
3245     8. Negated POSIX character classes that used a combination of internal tables
3246        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
3247        [[:^ascii]]. Typically, they would match almost any characters. The other
3248        POSIX classes were not broken in this way.
3250     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
3251        to find the match, as PCRE was deluded into thinking that the match had to
3252        start at the start point or following a newline. The same bug applied to
3253        patterns with negative forward assertions or any backward assertions
3254        preceding ".*" at the start, unless the pattern required a fixed first
3255        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
3257    10. In UTF-8 mode, when moving forwards in the subject after a failed match
3258        starting at the last subject character, bytes beyond the end of the subject
3259        string were read.
3261    11. Renamed the variable "class" as "classbits" to make life easier for C++
3262        users. (Previously there was a macro definition, but it apparently wasn't
3263        enough.)
3265    12. Added the new field "tables" to the extra data so that tables can be passed
3266        in at exec time, or the internal tables can be re-selected. This allows
3267        a compiled regex to be saved and re-used at a later time by a different
3268        program that might have everything at different addresses.
3270    13. Modified the pcre-config script so that, when run on Solaris, it shows a
3271        -R library as well as a -L library.
3273    14. The debugging options of pcretest (-d on the command line or D on a
3274        pattern) showed incorrect output for anything following an extended class
3275        that contained multibyte characters and which was followed by a quantifier.
3277    15. Added optional support for general category Unicode character properties
3278        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
3279        support. It adds about 90K to the size of the library. The meanings of the
3280        inbuilt class escapes such as \d and \s have NOT been changed.
3282    16. Updated pcredemo.c to include calls to free() to release the memory for the
3283        compiled pattern.
3285    17. The generated file chartables.c was being created in the source directory
3286        instead of in the building directory. This caused the build to fail if the
3287        source directory was different from the building directory, and was
3288        read-only.
3290    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
3291        file. No doubt somebody will tell me if they don't make sense... Also added
3292        Dan Mooney's comments about building on OpenVMS.
3294    19. Added support for partial matching via the PCRE_PARTIAL option for
3295        pcre_exec() and the \P data escape in pcretest.
3297    20. Extended pcretest with 3 new pattern features:
3299        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
3300              write the compiled pattern to the file whose name is "rest-of-line".
3301              This is a straight binary dump of the data, with the saved pointer to
3302              the character tables forced to be NULL. The study data, if any, is
3303              written too. After writing, pcretest reads a new pattern.
3305        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
3306              compiled pattern from the given file. There must not be any
3307              occurrences of "<" in the file name (pretty unlikely); if there are,
3308              pcretest will instead treat the initial "<" as a pattern delimiter.
3309              After reading in the pattern, pcretest goes on to read data lines as
3310              usual.
3312        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
3313              and 16-bit fields in a compiled pattern, to simulate a pattern that
3314              was compiled on a host of opposite endianness.
3316    21. The pcre-exec() function can now cope with patterns that were compiled on
3317        hosts of opposite endianness, with this restriction:
3319          As for any compiled expression that is saved and used later, the tables
3320          pointer field cannot be preserved; the extra_data field in the arguments
3321          to pcre_exec() should be used to pass in a tables address if a value
3322          other than the default internal tables were used at compile time.
3324    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
3325        now diagnosed as an error. Previously, most of the time, a negative number
3326        would have been treated as zero, but if in addition "ovector" was passed as
3327        NULL, a crash could occur.
3329    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
3330        new versions from the libtool 1.5 distribution (the last one is a copy of
3331        a file called libtool.m4). This seems to have fixed the need to patch
3332        "configure" to support Darwin 1.3 (which I used to do). However, I still
3333        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
3334        workstation).
3336    24. Changed the PCRE licence to be the more standard "BSD" licence.
3339    Version 4.5 01-Dec-03
3340    ---------------------
3342     1. There has been some re-arrangement of the code for the match() function so
3343        that it can be compiled in a version that does not call itself recursively.
3344        Instead, it keeps those local variables that need separate instances for
3345        each "recursion" in a frame on the heap, and gets/frees frames whenever it
3346        needs to "recurse". Keeping track of where control must go is done by means
3347        of setjmp/longjmp. The whole thing is implemented by a set of macros that
3348        hide most of the details from the main code, and operates only if
3349        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
3350        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
3351        operating.
3353        To make it easier for callers to provide specially tailored get/free
3354        functions for this usage, two new functions, pcre_stack_malloc, and
3355        pcre_stack_free, are used. They are always called in strict stacking order,
3356        and the size of block requested is always the same.
3358        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
3359        PCRE has been compiled to use the stack or the heap for recursion. The
3360        -C option of pcretest uses this to show which version is compiled.
3362        A new data escape \S, is added to pcretest; it causes the amounts of store
3363        obtained and freed by both kinds of malloc/free at match time to be added
3364        to the output.
3366     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
3367        what's available on my current Linux desktop machine.
3369     3. When matching a UTF-8 string, the test for a valid string at the start has
3370        been extended. If start_offset is not zero, PCRE now checks that it points
3371        to a byte that is the start of a UTF-8 character. If not, it returns
3372        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
3373        this is necessary because there may be backward assertions in the pattern.
3374        When matching the same subject several times, it may save resources to use
3375        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
3377     4. The code for checking the validity of UTF-8 strings has been tightened so
3378        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
3379        containing "overlong sequences".
3381     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
3382        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
3383        should have been "&", but it just so happened that all the cases this let
3384        through by mistake were picked up later in the function.
3386     6. I had used a variable called "isblank" - this is a C99 function, causing
3387        some compilers to warn. To avoid this, I renamed it (as "blankclass").
3389     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
3390        prompting; (b) run "./pcretest /dev/null" at the start of the test script
3391        so the version is shown; (c) stop "make test" echoing "./RunTest".
3393     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
3395     9. The prototype for memmove() for systems that don't have it was using
3396        size_t, but the inclusion of the header that defines size_t was later. I've
3397        moved the #includes for the C headers earlier to avoid this.
3399    10. Added some adjustments to the code to make it easier to compiler on certain
3400        special systems:
3402          (a) Some "const" qualifiers were missing.
3403          (b) Added the macro EXPORT before all exported functions; by default this
3404              is defined to be empty.
3405          (c) Changed the dftables auxiliary program (that builds chartables.c) so
3406              that it reads its output file name as an argument instead of writing
3407              to the standard output and assuming this can be redirected.
3409    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
3410        class containing characters with values greater than 255, PCRE compilation
3411        went into a loop.
3413    12. A recursive reference to a subpattern that was within another subpattern
3414        that had a minimum quantifier of zero caused PCRE to crash. For example,
3415        (x(y(?2))z)? provoked this bug with a subject that got as far as the
3416        recursion. If the recursively-called subpattern itself had a zero repeat,
3417        that was OK.
3419    13. In pcretest, the buffer for reading a data line was set at 30K, but the
3420        buffer into which it was copied (for escape processing) was still set at
3421        1024, so long lines caused crashes.
3423    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
3424        "internal error: code overflow...". This applied to any character class
3425        that was followed by a possessive quantifier.
3427    15. Modified the Makefile to add libpcre.la as a prerequisite for
3428        libpcreposix.la because I was told this is needed for a parallel build to
3429        work.
3431    16. If a pattern that contained .* following optional items at the start was
3432        studied, the wrong optimizing data was generated, leading to matching
3433        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
3434        matching string must start with a or b or c. The correct conclusion for
3435        this pattern is that a match can start with any character.
3438    Version 4.4 13-Aug-03
3439    ---------------------
3441     1. In UTF-8 mode, a character class containing characters with values between
3442        127 and 255 was not handled correctly if the compiled pattern was studied.
3443        In fixing this, I have also improved the studying algorithm for such
3444        classes (slightly).
3446     2. Three internal functions had redundant arguments passed to them. Removal
3447        might give a very teeny performance improvement.
3449     3. Documentation bug: the value of the capture_top field in a callout is *one
3450        more than* the number of the hightest numbered captured substring.
3452     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
3453        in incorrectly linking with a previously installed version. They now link
3454        explicitly with libpcre.la.
3456     5. configure.in no longer needs to recognize Cygwin specially.
3458     6. A problem in pcre.in for Windows platforms is fixed.
3460     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
3461        pcretest, it used to include the size of the study block as part of its
3462        output. Unfortunately, the structure contains a field that has a different
3463        size on different hardware architectures. This meant that the tests that
3464        showed this size failed. As the block is currently always of a fixed size,
3465        this information isn't actually particularly useful in pcretest output, so
3466        I have just removed it.
3468     8. Three pre-processor statements accidentally did not start in column 1.
3469        Sadly, there are *still* compilers around that complain, even though
3470        standard C has not required this for well over a decade. Sigh.
3472     9. In pcretest, the code for checking callouts passed small integers in the
3473        callout_data field, which is a void * field. However, some picky compilers
3474        complained about the casts involved for this on 64-bit systems. Now
3475        pcretest passes the address of the small integer instead, which should get
3476        rid of the warnings.
3478    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
3479        both compile and run time, and gives an error if an invalid UTF-8 sequence
3480        is found. There is a option for disabling this check in cases where the
3481        string is known to be correct and/or the maximum performance is wanted.
3483    11. In response to a bug report, I changed one line in Makefile.in from
3485            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
3486        to
3487            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
3489        to look similar to other lines, but I have no way of telling whether this
3490        is the right thing to do, as I do not use Windows. No doubt I'll get told
3491        if it's wrong...
3494    Version 4.3 21-May-03
3495    ---------------------
3497    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
3498       Makefile.
3500    2. Some refactoring to improve the quality of the code:
3502       (i)   The utf8_table... variables are now declared "const".
3504       (ii)  The code for \cx, which used the "case flipping" table to upper case
3505             lower case letters, now just substracts 32. This is ASCII-specific,
3506             but the whole concept of \cx is ASCII-specific, so it seems
3507             reasonable.
3509       (iii) PCRE was using its character types table to recognize decimal and
3510             hexadecimal digits in the pattern. This is silly, because it handles
3511             only 0-9, a-f, and A-F, but the character types table is locale-
3512             specific, which means strange things might happen. A private
3513             table is now used for this - though it costs 256 bytes, a table is
3514             much faster than multiple explicit tests. Of course, the standard
3515             character types table is still used for matching digits in subject
3516             strings against \d.
3518       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
3519             ending in _t are). So I've renamed it as ESC_tee.
3521    3. The first argument for regexec() in the POSIX wrapper should have been
3522       defined as "const".
3524    4. Changed pcretest to use malloc() for its buffers so that they can be
3525       Electric Fenced for debugging.
3527    5. There were several places in the code where, in UTF-8 mode, PCRE would try
3528       to read one or more bytes before the start of the subject string. Often this
3529       had no effect on PCRE's behaviour, but in some circumstances it could
3530       provoke a segmentation fault.
3532    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
3533       to try to read one or more bytes before the start of the subject string.
3535    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3536       UTF-8 support could misbehave in various ways if the subject string
3537       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3538       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3539       back over UTF-8 characters.)
3542    Version 4.2 14-Apr-03
3543    ---------------------
3545    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
3547    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
3548         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
3549         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
3550         [WIN_PREFIX]: new variable, "cyg" for Cygwin
3551         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
3552           and BUILD_EXEEXT
3553         Note: automatic setting of the BUILD variables is not yet working
3554         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
3555           compile-time but not at link-time
3556         [LINK]: use for linking executables only
3557         make different versions for Windows and non-Windows
3558         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
3559           libraries
3560         [LINK_FOR_BUILD]: new variable
3561         [OBJEXT]: use throughout
3562         [EXEEXT]: use throughout
3563         <winshared>: new target
3564         <wininstall>: new target
3565         <dftables.o>: use native compiler
3566         <dftables>: use native linker
3567         <install>: handle Windows platform correctly
3568         <clean>: ditto
3569         <check>: ditto
3570         copy DLL to top builddir before testing
3572       As part of these changes, -no-undefined was removed again. This was reported
3573       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3574       in any case.
3576    3. Some tidies to get rid of compiler warnings:
3578       . In the match_data structure, match_limit was an unsigned long int, whereas
3579         match_call_count was an int. I've made them both unsigned long ints.
3581       . In pcretest the fact that a const uschar * doesn't automatically cast to
3582         a void * provoked a warning.
3584       . Turning on some more compiler warnings threw up some "shadow" variables
3585         and a few more missing casts.
3587    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3588       option, a class that contained a single character with a value between 128
3589       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3591    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3592       option, a class that contained several characters, but with at least one
3593       whose value was between 128 and 255 caused PCRE to crash.
3596    Version 4.1 12-Mar-03
3597    ---------------------
3599    1. Compiling with gcc -pedantic found a couple of places where casts were
3600    needed, and a string in dftables.c that was longer than standard compilers are
3601    required to support.
3603    2. Compiling with Sun's compiler found a few more places where the code could
3604    be tidied up in order to avoid warnings.
3606    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3607    first of these names is deprecated in the latest Autoconf in favour of the name
3608    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3609    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3610    analogy I have changed it to CFLAGS_FOR_BUILD.
3612    4. Added -no-undefined to the linking command in the Makefile, because this is
3613    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the