ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 5 by nigel, Sat Feb 24 21:38:05 2007 UTC revision 983 by ph10, Fri Jul 6 08:55:10 2012 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.31 06-July-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.
149    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
150        and pcregrep, because some OS require shared objects to be explicitly
151        passed to ld, causing the link step to fail if they are not.
153    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
154        mode, patterns that started with \h* or \R* might be incorrectly matched.
157    Version 8.30 04-February-2012
158    -----------------------------
160    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
161        name is defined in ctype.h.
163    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
164        only in quite long subpatterns.
166    3.  Removed the function pcre_info(), which has been obsolete and deprecated
167        since it was replaced by pcre_fullinfo() in February 2000.
169    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
170        match a (*MARK), and the match failed at the start of the subject, a
171        reference to memory before the start of the subject could occur. This bug
172        was introduced by fix 17 of release 8.21.
174    5.  A reference to an unset group with zero minimum repetition was giving
175        totally wrong answers (in non-JavaScript-compatibility mode). For example,
176        /(another)?(\1?)test/ matched against "hello world test". This bug was
177        introduced in release 8.13.
179    6.  Add support for 16-bit character strings (a large amount of work involving
180        many changes and refactorings).
182    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
183        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
184        from a file.
186    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
187        rounding is not applied in this particular case).
189    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
190        if they appear, or are escaped, in patterns.
192    10. Get rid of a number of -Wunused-but-set-variable warnings.
194    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
195        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
196        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
197        also returns the mark "x". This bug applied to capturing parentheses,
198        non-capturing parentheses, and atomic parentheses. It also applied to some
199        assertions.
201    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
202        information out of configure.ac instead of relying on pcre.h.generic, which
203        is not stored in the repository.
205    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
206        -lreadline.
208    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
210    15. Applied Graycode's patch to put the top-level frame on the stack rather
211        than the heap when not using the stack for recursion. This gives a
212        performance improvement in many cases when recursion is not deep.
214    16. Experimental code added to "pcretest -C" to output the stack frame size.
217    Version 8.21 12-Dec-2011
218    ------------------------
220    1.  Updating the JIT compiler.
222    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
223        are added as well.
225    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
226        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
227        calling _pcre_jit_exec. Some extra comments are added.
229    4.  (*MARK) settings inside atomic groups that do not contain any capturing
230        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
231        was introduced by change 18 for 8.20.
233    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
234        ECMA-262 standard.
236    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
237        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
238        This bug was probably introduced by change 9 of 8.13.
240    7.  While fixing 6 above, I noticed that a number of other items were being
241        incorrectly rejected as "not fixed length". This arose partly because newer
242        opcodes had not been added to the fixed-length checking code. I have (a)
243        corrected the bug and added tests for these items, and (b) arranged for an
244        error to occur if an unknown opcode is encountered while checking for fixed
245        length instead of just assuming "not fixed length". The items that were
246        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
247        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
248        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
250    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
251        being incorrectly compiled and would have given unpredicatble results.
253    9.  A possessively repeated subpattern with minimum repeat count greater than
254        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
255        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
256        the first (A) could occur when it should not.
258    10. Add a cast and remove a redundant test from the code.
260    11. JIT should use pcre_malloc/pcre_free for allocation.
262    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
263        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
264        is anything other than /usr, -L is still shown).
266    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
268    14. Perl does not support \N without a following name in a [] class; PCRE now
269        also gives an error.
271    15. If a forward reference was repeated with an upper limit of around 2000,
272        it caused the error "internal error: overran compiling workspace". The
273        maximum number of forward references (including repeats) was limited by the
274        internal workspace, and dependent on the LINK_SIZE. The code has been
275        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
276        the default depends on LINK_SIZE. There is a new upper limit (for safety)
277        of around 200,000 forward references. While doing this, I also speeded up
278        the filling in of repeated forward references.
280    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
281        incorrectly expecting the subject to contain another "a" after the start.
283    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
284        in the match, the SKIP should be ignored. This was not happening; instead
285        the SKIP was being treated as NOMATCH. For patterns such as
286        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
287        tested.
289    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
290        now much more compatible with Perl, in particular in cases where the result
291        is a non-match for a non-anchored pattern. For example, if
292        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
293        "m", where previously it did not return a name. A side effect of this
294        change is that for partial matches, the last encountered mark name is
295        returned, as for non matches. A number of tests that were previously not
296        Perl-compatible have been moved into the Perl-compatible test files. The
297        refactoring has had the pleasing side effect of removing one argument from
298        the match() function, thus reducing its stack requirements.
300    19. If the /S+ option was used in pcretest to study a pattern using JIT,
301        subsequent uses of /S (without +) incorrectly behaved like /S+.
303    21. Retrieve executable code size support for the JIT compiler and fixing
304        some warnings.
306    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
307        not work when the shorter character appeared right at the end of the
308        subject string.
310    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
311        systems.
313    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
314        output it when the /M option is used in pcretest.
316    25. The CheckMan script was not being included in the distribution. Also, added
317        an explicit "perl" to run Perl scripts from the PrepareRelease script
318        because this is reportedly needed in Windows.
320    26. If study data was being save in a file and studying had not found a set of
321        "starts with" bytes for the pattern, the data written to the file (though
322        never used) was taken from uninitialized memory and so caused valgrind to
323        complain.
325    27. Updated RunTest.bat as provided by Sheri Pierce.
327    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
329    29. Computation of memory usage for the table of capturing group names was
330        giving an unnecessarily large value.
333    Version 8.20 21-Oct-2011
334    ------------------------
336    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
337        a POSIX class. After further experiments with Perl, which convinced me that
338        Perl has bugs and confusions, a closing square bracket is no longer allowed
339        in a POSIX name. This bug also affected patterns with classes that started
340        with full stops.
342    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
343        captured substring, but while checking the failing first alternative,
344        substring 1 is temporarily captured. If the output vector supplied to
345        pcre_exec() was not big enough for this capture, the yield of the function
346        was still zero ("insufficient space for captured substrings"). This cannot
347        be totally fixed without adding another stack variable, which seems a lot
348        of expense for a edge case. However, I have improved the situation in cases
349        such as /(a)(b)x|abc/ matched against "abc", where the return code
350        indicates that fewer than the maximum number of slots in the ovector have
351        been set.
353    3.  Related to (2) above: when there are more back references in a pattern than
354        slots in the output vector, pcre_exec() uses temporary memory during
355        matching, and copies in the captures as far as possible afterwards. It was
356        using the entire output vector, but this conflicts with the specification
357        that only 2/3 is used for passing back captured substrings. Now it uses
358        only the first 2/3, for compatibility. This is, of course, another edge
359        case.
361    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
362        main code base, and can be used by building with --enable-jit. When this is
363        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
364        runtime --no-jit option is given.
366    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
367        ovector, the return from the function was zero, implying that there were
368        other matches that did not fit. The correct "exactly full" value is now
369        returned.
371    6.  If a subpattern that was called recursively or as a subroutine contained
372        (*PRUNE) or any other control that caused it to give a non-standard return,
373        invalid errors such as "Error -26 (nested recursion at the same subject
374        position)" or even infinite loops could occur.
376    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
377        computing the minimum length on reaching *ACCEPT, and so ended up with the
378        wrong value of 1 rather than 0. Further investigation indicates that
379        computing a minimum subject length in the presence of *ACCEPT is difficult
380        (think back references, subroutine calls), and so I have changed the code
381        so that no minimum is registered for a pattern that contains *ACCEPT.
383    8.  If (*THEN) was present in the first (true) branch of a conditional group,
384        it was not handled as intended. [But see 16 below.]
386    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
387        Sheri Pierce.
389    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
390        the first byte in a match must be "a".
392    11. Change 17 for 8.13 increased the recursion depth for patterns like
393        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
394        pattern contains any instances of (*THEN). If it does not, the old
395        optimizations are restored. It would be nice to do this on a per-group
396        basis, but at the moment that is not feasible.
398    12. In some environments, the output of pcretest -C is CRLF terminated. This
399        broke RunTest's code that checks for the link size. A single white space
400        character after the value is now allowed for.
402    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
403        For "fr", it uses the Windows-specific input and output files.
405    14. If (*THEN) appeared in a group that was called recursively or as a
406        subroutine, it did not work as intended. [But see next item.]
408    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
409        pattern fragments (but not containing any | characters). If A and B are
410        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
411        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
412        D. In other words, Perl considers parentheses that do not contain any |
413        characters to be part of a surrounding alternative, whereas PCRE was
414        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
415        differently. PCRE now behaves in the same way as Perl, except in the case
416        of subroutine/recursion calls such as (?1) which have in any case always
417        been different (but PCRE had them first :-).
419    16. Related to 15 above: Perl does not treat the | in a conditional group as
420        creating alternatives. Such a group is treated in the same way as an
421        ordinary group without any | characters when processing (*THEN). PCRE has
422        been changed to match Perl's behaviour.
424    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
425        RunGrepTest script failed.
427    18. Change 22 for version 13 caused atomic groups to use more stack. This is
428        inevitable for groups that contain captures, but it can lead to a lot of
429        stack use in large patterns. The old behaviour has been restored for atomic
430        groups that do not contain any capturing parentheses.
432    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
433        suppress the check for a minimum subject length at run time. (If it was
434        given to pcre_exec() or pcre_dfa_exec() it did work.)
436    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
437        fail to work when decoding hex characters in data strings in EBCDIC
438        environments.
440    21. It appears that in at least one Mac OS environment, the isxdigit() function
441        is implemented as a macro that evaluates to its argument more than once,
442        contravening the C 90 Standard (I haven't checked a later standard). There
443        was an instance in pcretest which caused it to go wrong when processing
444        \x{...} escapes in subject strings. The has been rewritten to avoid using
445        things like p++ in the argument of isxdigit().
448    Version 8.13 16-Aug-2011
449    ------------------------
451    1.  The Unicode data tables have been updated to Unicode 6.0.0.
453    2.  Two minor typos in pcre_internal.h have been fixed.
455    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
456        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
457        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
459    4.  There were a number of related bugs in the code for matching backrefences
460        caselessly in UTF-8 mode when codes for the characters concerned were
461        different numbers of bytes. For example, U+023A and U+2C65 are an upper
462        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
463        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
464        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
465        2-byte code at the end of the subject (it thought there wasn't enough data
466        left).
468    5.  Comprehensive information about what went wrong is now returned by
469        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
470        as the output vector has at least 2 elements. The offset of the start of
471        the failing character and a reason code are placed in the vector.
473    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
474        now returned is for the first byte of the failing character, instead of the
475        last byte inspected. This is an incompatible change, but I hope it is small
476        enough not to be a problem. It makes the returned offset consistent with
477        pcre_exec() and pcre_dfa_exec().
479    7.  pcretest now gives a text phrase as well as the error number when
480        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
481        failure, the offset and reason code are output.
483    8.  When \R was used with a maximizing quantifier it failed to skip backwards
484        over a \r\n pair if the subsequent match failed. Instead, it just skipped
485        back over a single character (\n). This seems wrong (because it treated the
486        two characters as a single entity when going forwards), conflicts with the
487        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
488        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
489        has been changed.
491    9.  Some internal refactoring has changed the processing so that the handling
492        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
493        time (the PCRE_DOTALL option was changed this way some time ago: version
494        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
495        which was always a bit of a fudge. It also means that there is one less
496        argument for the match() function, which reduces its stack requirements
497        slightly. This change also fixes an incompatibility with Perl: the pattern
498        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
500    10. More internal refactoring has drastically reduced the number of recursive
501        calls to match() for possessively repeated groups such as (abc)++ when
502        using pcre_exec().
504    11. While implementing 10, a number of bugs in the handling of groups were
505        discovered and fixed:
507        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
508        (a|)*(?1) gave a compile-time internal error.
509        ((a|)+)+  did not notice that the outer group could match an empty string.
510        (^a|^)+   was not marked as anchored.
511        (.*a|.*)+ was not marked as matching at start or after a newline.
513    12. Yet more internal refactoring has removed another argument from the match()
514        function. Special calls to this function are now indicated by setting a
515        value in a variable in the "match data" data block.
517    13. Be more explicit in pcre_study() instead of relying on "default" for
518        opcodes that mean there is no starting character; this means that when new
519        ones are added and accidentally left out of pcre_study(), testing should
520        pick them up.
522    14. The -s option of pcretest has been documented for ages as being an old
523        synonym of -m (show memory usage). I have changed it to mean "force study
524        for every regex", that is, assume /S for every regex. This is similar to -i
525        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
526        using it. It makes it easier to run collections of tests with and without
527        study enabled, and thereby test pcre_study() more easily. All the standard
528        tests are now run with and without -s (but some patterns can be marked as
529        "never study" - see 20 below).
531    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
532        restoration of the capturing data to the outer values was not happening
533        correctly.
535    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
536        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
537        pattern had matched an empty string, and so incorrectly returned a no
538        match.
540    17. There was optimizing code for the last branch of non-capturing parentheses,
541        and also for the obeyed branch of a conditional subexpression, which used
542        tail recursion to cut down on stack usage. Unfortunately, now that there is
543        the possibility of (*THEN) occurring in these branches, tail recursion is
544        no longer possible because the return has to be checked for (*THEN). These
545        two optimizations have therefore been removed. [But see 8.20/11 above.]
547    18. If a pattern containing \R was studied, it was assumed that \R always
548        matched two bytes, thus causing the minimum subject length to be
549        incorrectly computed because \R can also match just one byte.
551    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
552        was incorrectly computed.
554    20. If /S is present twice on a test pattern in pcretest input, it now
555        *disables* studying, thereby overriding the use of -s on the command line
556        (see 14 above). This is necessary for one or two tests to keep the output
557        identical in both cases.
559    21. When (*ACCEPT) was used in an assertion that matched an empty string and
560        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
562    22. When an atomic group that contained a capturing parenthesis was
563        successfully matched, but the branch in which it appeared failed, the
564        capturing was not being forgotten if a higher numbered group was later
565        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
566        group 1 to "a", when in fact it should be unset. This applied to multi-
567        branched capturing and non-capturing groups, repeated or not, and also to
568        positive assertions (capturing in negative assertions does not happen
569        in PCRE) and also to nested atomic groups.
571    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
572        subject after a captured substring, to make it easier to tell which of a
573        number of identical substrings has been captured.
575    24. The way atomic groups are processed by pcre_exec() has been changed so that
576        if they are repeated, backtracking one repetition now resets captured
577        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
578        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
579        "aaa". Previously, it would have been "a". As part of this code
580        refactoring, the way recursive calls are handled has also been changed.
582    25. If an assertion condition captured any substrings, they were not passed
583        back unless some other capturing happened later. For example, if
584        (?(?=(a))a) was matched against "a", no capturing was returned.
586    26. When studying a pattern that contained subroutine calls or assertions,
587        the code for finding the minimum length of a possible match was handling
588        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
589        group 1 called group 2 while simultaneously a separate group 2 called group
590        1). A stack overflow occurred in this case. I have fixed this by limiting
591        the recursion depth to 10.
593    27. Updated RunTest.bat in the distribution to the version supplied by Tom
594        Fortmann. This supports explicit test numbers on the command line, and has
595        argument validation and error reporting.
597    28. An instance of \X with an unlimited repeat could fail if at any point the
598        first character it looked at was a mark character.
600    29. Some minor code refactoring concerning Unicode properties and scripts
601        should reduce the stack requirement of match() slightly.
603    30. Added the '=' option to pcretest to check the setting of unused capturing
604        slots at the end of the pattern, which are documented as being -1, but are
605        not included in the return count.
607    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
608        compiled something random. Now it gives a compile-time error (as does
609        Perl).
611    32. A *MARK encountered during the processing of a positive assertion is now
612        recorded and passed back (compatible with Perl).
614    33. If --only-matching or --colour was set on a pcregrep call whose pattern
615        had alternative anchored branches, the search for a second match in a line
616        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
617        matched the line "0102" twice. The same bug affected patterns that started
618        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
619        twice.
621    34. Previously, PCRE did not allow quantification of assertions. However, Perl
622        does, and because of capturing effects, quantifying parenthesized
623        assertions may at times be useful. Quantifiers are now allowed for
624        parenthesized assertions.
626    35. A minor code tidy in pcre_compile() when checking options for \R usage.
628    36. \g was being checked for fancy things in a character class, when it should
629        just be a literal "g".
631    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
632        appearance of a nested POSIX class supersedes an apparent external class.
633        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
634        unescaped square brackets may also appear as part of class names. For
635        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
636        more like Perl. (But see 8.20/1 above.)
638    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
639        was because it thought it was \N{name}, which is not supported).
641    39. Add minix to OS list not supporting the -S option in pcretest.
643    40. PCRE tries to detect cases of infinite recursion at compile time, but it
644        cannot analyze patterns in sufficient detail to catch mutual recursions
645        such as ((?1))((?2)). There is now a runtime test that gives an error if a
646        subgroup is called recursively as a subpattern for a second time at the
647        same position in the subject string. In previous releases this might have
648        been caught by the recursion limit, or it might have run out of stack.
650    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
651        happen only once. PCRE was, however incorrectly giving a compile time error
652        "recursive call could loop indefinitely" because it cannot analyze the
653        pattern in sufficient detail. The compile time test no longer happens when
654        PCRE is compiling a conditional subpattern, but actual runaway loops are
655        now caught at runtime (see 40 above).
657    42. It seems that Perl allows any characters other than a closing parenthesis
658        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
659        has been changed to be the same.
661    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
662        as not to get warnings when autogen.sh is called. Also changed
663        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
665    44. To help people who use pcregrep to scan files containing exceedingly long
666        lines, the following changes have been made:
668        (a) The default value of the buffer size parameter has been increased from
669            8K to 20K. (The actual buffer used is three times this size.)
671        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
672            PCRE is built.
674        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
675            to be set at run time.
677        (d) Numerical values in pcregrep options can be followed by K or M, for
678            example --buffer-size=50K.
680        (e) If a line being scanned overflows pcregrep's buffer, an error is now
681            given and the return code is set to 2.
683    45. Add a pointer to the latest mark to the callout data block.
685    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
686        partial match of an empty string instead of no match. This was specific to
687        the use of ".".
689    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
690        complete match instead of a partial match. This bug was dependent on both
691        the PCRE_UTF8 and PCRE_DOTALL options being set.
693    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
694        starting byte set, because \b was not being ignored.
697    Version 8.12 15-Jan-2011
698    ------------------------
700    1.  Fixed some typos in the markup of the man pages, and wrote a script that
701        checks for such things as part of the documentation building process.
703    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
704        --match-limit and --recursion-limit options (added for 8.11). In
705        particular, this made one of the standard tests fail. (The integer value
706        went into the wrong half of a long int.)
708    3.  If the --colour option was given to pcregrep with -v (invert match), it
709        did strange things, either producing crazy output, or crashing. It should,
710        of course, ignore a request for colour when reporting lines that do not
711        match.
713    4.  Another pcregrep bug caused similar problems if --colour was specified with
714        -M (multiline) and the pattern match finished with a line ending.
716    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
717        matched in multiline mode, the following line was shown as part of the
718        match. This seems wrong, so I have changed it.
720    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
721        the check for further matches in the same line (so they could be coloured)
722        to overrun the end of the current line. If another match was found, it was
723        incorrectly shown (and then shown again when found in the next line).
725    7.  If pcregrep was compiled under Windows, there was a reference to the
726        function pcregrep_exit() before it was defined. I am assuming this was
727        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
728        reported by a user. I've moved the definition above the reference.
731    Version 8.11 10-Dec-2010
732    ------------------------
734    1.  (*THEN) was not working properly if there were untried alternatives prior
735        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
736        backtracked to try for "b" instead of moving to the next alternative branch
737        at the same level (in this case, to look for "c"). The Perl documentation
738        is clear that when (*THEN) is backtracked onto, it goes to the "next
739        alternative in the innermost enclosing group".
741    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
742        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
743        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
744        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
745        (*THEN).
747    3.  If \s appeared in a character class, it removed the VT character from
748        the class, even if it had been included by some previous item, for example
749        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
750        of \s, but is part of the POSIX "space" class.)
752    4.  A partial match never returns an empty string (because you can always
753        match an empty string at the end of the subject); however the checking for
754        an empty string was starting at the "start of match" point. This has been
755        changed to the "earliest inspected character" point, because the returned
756        data for a partial match starts at this character. This means that, for
757        example, /(?<=abc)def/ gives a partial match for the subject "abc"
758        (previously it gave "no match").
760    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
761        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
762        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
763        has an implication that the given string is incomplete (because a partial
764        match is preferred over a full match). For this reason, these items now
765        give a partial match in this situation. [Aside: previously, the one case
766        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
767        match rather than a full match, which was wrong by the old rules, but is
768        now correct.]
770    6.  There was a bug in the handling of #-introduced comments, recognized when
771        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
772        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
773        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
774        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
775        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
776        places in pcre_compile().
778    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
779        comments when looking ahead for named forward references to subpatterns,
780        the only newline sequence it recognized was NL. It now handles newlines
781        according to the set newline convention.
783    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
784        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
785        cater for a lack of strerror(). These oversights have been fixed.
787    9.  Added --match-limit and --recursion-limit to pcregrep.
789    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
791    11. When the -o option was used, pcregrep was setting a return code of 1, even
792        when matches were found, and --line-buffered was not being honoured.
794    12. Added an optional parentheses number to the -o and --only-matching options
795        of pcregrep.
797    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
798        can match an empty string. The code to do it in pcretest and pcredemo
799        needed fixing:
801        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
802            only one byte after an empty string match just before CRLF (this case
803            just got forgotten; "any" and "anycrlf" were OK).
805        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
806            mode when an empty string match preceded an ASCII character followed by
807            a non-ASCII character. (The code for advancing by one character rather
808            than one byte was nonsense.)
810        (c) The pcredemo.c sample program did not have any code at all to handle
811            the cases when CRLF is a valid newline sequence.
813    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
814        as a starting offset was within the subject string. There is now a new
815        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
816        negative or greater than the length of the string. In order to test this,
817        pcretest is extended to allow the setting of negative starting offsets.
819    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
820        starting offset points to the beginning of a UTF-8 character was
821        unnecessarily clumsy. I tidied it up.
823    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
824        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
826    17. Nobody had reported that the --include_dir option, which was added in
827        release 7.7 should have been called --include-dir (hyphen, not underscore)
828        for compatibility with GNU grep. I have changed it to --include-dir, but
829        left --include_dir as an undocumented synonym, and the same for
830        --exclude-dir, though that is not available in GNU grep, at least as of
831        release 2.5.4.
833    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
834        characters from a string of bytes) have been redefined so as not to use
835        loops, in order to improve performance in some environments. At the same
836        time, I abstracted some of the common code into auxiliary macros to save
837        repetition (this should not affect the compiled code).
839    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
840        compile-time error is now given if \c is not followed by an ASCII
841        character, that is, a byte less than 128. (In EBCDIC mode, the code is
842        different, and any byte value is allowed.)
844    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
845        START_OPTIMIZE option, which is now allowed at compile time - but just
846        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
847        to pcregrep and other applications that have no direct access to PCRE
848        options. The new /Y option in pcretest sets this option when calling
849        pcre_compile().
851    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
852        back references. Groups containing recursive back references were forced to
853        be atomic by that change, but in the case of named groups, the amount of
854        memory required was incorrectly computed, leading to "Failed: internal
855        error: code overflow". This has been fixed.
857    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
858        pcretest.c, to avoid build problems in some Borland environments.
861    Version 8.10 25-Jun-2010
862    ------------------------
864    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
865        THEN.
867    2.  (*ACCEPT) was not working when inside an atomic group.
869    3.  Inside a character class, \B is treated as a literal by default, but
870        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
871        causes the error). The code is unchanged, but I tidied the documentation.
873    4.  Inside a character class, PCRE always treated \R and \X as literals,
874        whereas Perl faults them if its -w option is set. I have changed PCRE so
875        that it faults them when PCRE_EXTRA is set.
877    5.  Added support for \N, which always matches any character other than
878        newline. (It is the same as "." when PCRE_DOTALL is not set.)
880    6.  When compiling pcregrep with newer versions of gcc which may have
881        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
882        declared with attribute warn_unused_result" were given. Just casting the
883        result to (void) does not stop the warnings; a more elaborate fudge is
884        needed. I've used a macro to implement this.
886    7.  Minor change to pcretest.c to avoid a compiler warning.
888    8.  Added four artifical Unicode properties to help with an option to make
889        \s etc use properties (see next item). The new properties are: Xan
890        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
892    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
893        use Unicode properties. (*UCP) at the start of a pattern can be used to set
894        this option. Modified pcretest to add /W to test this facility. Added
895        REG_UCP to make it available via the POSIX interface.
897    10. Added --line-buffered to pcregrep.
899    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
900        studied, and the match started with a letter with a code point greater than
901        127 whose first byte was different to the first byte of the other case of
902        the letter, the other case of this starting letter was not recognized
903        (#976).
905    12. If a pattern that was studied started with a repeated Unicode property
906        test, for example, \p{Nd}+, there was the theoretical possibility of
907        setting up an incorrect bitmap of starting bytes, but fortunately it could
908        not have actually happened in practice until change 8 above was made (it
909        added property types that matched character-matching opcodes).
911    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
912        possible starting bytes for non-anchored patterns.
914    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
915        \R, and also a number of cases that involve Unicode properties, both
916        explicit and implicit when PCRE_UCP is set.
918    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
919        input, it could crash or give wrong results if characters with values
920        greater than 0xc0 were present in the subject string. (Detail: it assumed
921        UTF-8 input when processing these items.)
923    16. Added a lot of (int) casts to avoid compiler warnings in systems where
924        size_t is 64-bit (#991).
926    17. Added a check for running out of memory when PCRE is compiled with
927        --disable-stack-for-recursion (#990).
929    18. If the last data line in a file for pcretest does not have a newline on
930        the end, a newline was missing in the output.
932    19. The default pcre_chartables.c file recognizes only ASCII characters (values
933        less than 128) in its various bitmaps. However, there is a facility for
934        generating tables according to the current locale when PCRE is compiled. It
935        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
936        characters, are recognized by isspace() and therefore were getting set in
937        these tables, and indeed these tables seem to approximate to ISO 8859. This
938        caused a problem in UTF-8 mode when pcre_study() was used to create a list
939        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
940        which of course cannot start UTF-8 characters. I have changed the code so
941        that only real ASCII characters (less than 128) and the correct starting
942        bytes for UTF-8 encodings are set for characters greater than 127 when in
943        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
944        altogether.)
946    20. Added the /T option to pcretest so as to be able to run tests with non-
947        standard character tables, thus making it possible to include the tests
948        used for 19 above in the standard set of tests.
950    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
951        reference to a subpattern the other side of a comment that contains an
952        opening parenthesis caused either an internal compiling error, or a
953        reference to the wrong subpattern.
956    Version 8.02 19-Mar-2010
957    ------------------------
959    1.  The Unicode data tables have been updated to Unicode 5.2.0.
961    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
962        configured.
964    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
965        original author of that file, following a query about its status.
967    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
968        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
970    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
971        quantifier applied to a forward-referencing subroutine call, could compile
972        incorrect code or give the error "internal error: previously-checked
973        referenced subpattern not found".
975    6.  Both MS Visual Studio and Symbian OS have problems with initializing
976        variables to point to external functions. For these systems, therefore,
977        pcre_malloc etc. are now initialized to local functions that call the
978        relevant global functions.
980    7.  There were two entries missing in the vectors called coptable and poptable
981        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
982        I've fixed the data, and added a kludgy way of testing at compile time that
983        the lengths are correct (equal to the number of opcodes).
985    8.  Following on from 7, I added a similar kludge to check the length of the
986        eint vector in pcreposix.c.
988    9.  Error texts for pcre_compile() are held as one long string to avoid too
989        much relocation at load time. To find a text, the string is searched,
990        counting zeros. There was no check for running off the end of the string,
991        which could happen if a new error number was added without updating the
992        string.
994    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
996    11. \K was not working if it appeared in an atomic group or in a group that
997        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
998        \K is "not well defined" if used in an assertion. PCRE now accepts it if
999        the assertion is positive, but not if it is negative.
1001    12. Change 11 fortuitously reduced the size of the stack frame used in the
1002        "match()" function of pcre_exec.c by one pointer. Forthcoming
1003        implementation of support for (*MARK) will need an extra pointer on the
1004        stack; I have reserved it now, so that the stack frame size does not
1005        decrease.
1007    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1008        item in branch that calls a recursion is a subroutine call - as in the
1009        second branch in the above example - was incorrectly given the compile-
1010        time error "recursive call could loop indefinitely" because pcre_compile()
1011        was not correctly checking the subroutine for matching a non-empty string.
1013    14. The checks for overrunning compiling workspace could trigger after an
1014        overrun had occurred. This is a "should never occur" error, but it can be
1015        triggered by pathological patterns such as hundreds of nested parentheses.
1016        The checks now trigger 100 bytes before the end of the workspace.
1018    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1021    Version 8.01 19-Jan-2010
1022    ------------------------
1024    1.  If a pattern contained a conditional subpattern with only one branch (in
1025        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1026        computed the wrong minimum data length (which is of course zero for such
1027        subpatterns). This could cause incorrect "no match" results.
1029    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1030        the pattern is reset in the first branch, pcre_compile() failed with
1031        "internal error: code overflow at offset...". This happened only when
1032        the reset was to the original external option setting. (An optimization
1033        abstracts leading options settings into an external setting, which was the
1034        cause of this.)
1036    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1037        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1038        assertion pattern did not match (meaning that the assertion was true), it
1039        was incorrectly treated as false if the SKIP had been reached during the
1040        matching. This also applied to assertions used as conditions.
1042    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1043        assertion subpattern, including such a pattern used as a condition,
1044        unpredictable results occurred, instead of the error return
1047    5.  The C++ GlobalReplace function was not working like Perl for the special
1048        situation when an empty string is matched. It now does the fancy magic
1049        stuff that is necessary.
1051    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1052        removed. (These were left over from very, very early versions of PCRE.)
1054    7.  Some cosmetic changes to the code to make life easier when compiling it
1055        as part of something else:
1057        (a) Change DEBUG to PCRE_DEBUG.
1059        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1060            called "current" as "current_branch", to prevent a collision with the
1061            Linux macro when compiled as a kernel module.
1063        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1064            prevent a collision with the Linux macro when compiled as a kernel
1065            module.
1067    8.  In pcre_compile() there are some checks for integer overflows that used to
1068        cast potentially large values to (double). This has been changed to that
1069        when building, a check for int64_t is made, and if it is found, it is used
1070        instead, thus avoiding the use of floating point arithmetic. (There is no
1071        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1072        double.
1074    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1075        2005 (difference between two addresses compared to an unsigned value).
1077    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1078        custom one, because of the following reported problem in Windows:
1080          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1081              under Win32.
1082          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1083              therefore missing the function definition.
1084          - The compiler thus generates a "C" signature for the test function.
1085          - The linker fails to find the "C" function.
1086          - PCRE fails to configure if asked to do so against libbz2.
1088    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1089        messages were output:
1091          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1092          rerunning libtoolize, to keep the correct libtool macros in-tree.
1093          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1095        I have done both of these things.
1097    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1098        most of the time, it *can* run out if it is given a pattern that contains a
1099        runaway infinite recursion. I updated the discussion in the pcrestack man
1100        page.
1102    13. Now that we have gone to the x.xx style of version numbers, the minor
1103        version may start with zero. Using 08 or 09 is a bad idea because users
1104        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1105        interpreted as invalid octal numbers. I've updated the previous comment in
1106        configure.ac, and also added a check that gives an error if 08 or 09 are
1107        used.
1109    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1110        causing partial matching to fail when the end of the subject matched \W
1111        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1113    15. There were some discrepancies between the declarations in pcre_internal.h
1114        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1115        their definitions. The declarations used "const uschar *" and the
1116        definitions used USPTR. Even though USPTR is normally defined as "const
1117        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1118        reported that: "This difference in casting confuses some C++ compilers, for
1119        example, SunCC recognizes above declarations as different functions and
1120        generates broken code for hbpcre." I have changed the declarations to use
1121        USPTR.
1123    16. GNU libtool is named differently on some systems. The autogen.sh script now
1124        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1125        (FreeBSD).
1127    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1128        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1129        comment: "Figure out how to create a longlong from a string: strtoll and
1130        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1131        instance, but it only takes 2 args instead of 3!"
1133    18. A subtle bug concerned with back references has been fixed by a change of
1134        specification, with a corresponding code fix. A pattern such as
1135        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1136        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1137        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1138        same bug. Such groups have to be quantified to be useful, or contained
1139        inside another quantified group. (If there's no repetition, the reference
1140        can never match.) The problem arises because, having left the group and
1141        moved on to the rest of the pattern, a later failure that backtracks into
1142        the group uses the captured value from the final iteration of the group
1143        rather than the correct earlier one. I have fixed this in PCRE by forcing
1144        any group that contains a reference to itself to be an atomic group; that
1145        is, there cannot be any backtracking into it once it has completed. This is
1146        similar to recursive and subroutine calls.
1149    Version 8.00 19-Oct-09
1150    ----------------------
1152    1.  The table for translating pcre_compile() error codes into POSIX error codes
1153        was out-of-date, and there was no check on the pcre_compile() error code
1154        being within the table. This could lead to an OK return being given in
1155        error.
1157    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1158        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1159        in a Windows environment.
1161    3.  The pcregrep --count option prints the count for each file even when it is
1162        zero, as does GNU grep. However, pcregrep was also printing all files when
1163        --files-with-matches was added. Now, when both options are given, it prints
1164        counts only for those files that have at least one match. (GNU grep just
1165        prints the file name in this circumstance, but including the count seems
1166        more useful - otherwise, why use --count?) Also ensured that the
1167        combination -clh just lists non-zero counts, with no names.
1169    4.  The long form of the pcregrep -F option was incorrectly implemented as
1170        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1171        but it seems right to fix it, and I didn't think it was worth preserving
1172        the old behaviour.
1174    5.  The command line items --regex=pattern and --regexp=pattern were not
1175        recognized by pcregrep, which required --regex pattern or --regexp pattern
1176        (with a space rather than an '='). The man page documented the '=' forms,
1177        which are compatible with GNU grep; these now work.
1179    6.  No libpcreposix.pc file was created for pkg-config; there was just
1180        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1182    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1183        when UCP support is not needed, by modifying the Python script that
1184        generates it from Unicode data files. This should not matter if the module
1185        is correctly used as a library, but I received one complaint about 50K of
1186        unwanted data. My guess is that the person linked everything into his
1187        program rather than using a library. Anyway, it does no harm.
1189    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1190        was a minimum greater than 1 for a wide character in a possessive
1191        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1192        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1193        character. Chaos in the form of incorrect output or a compiling loop could
1194        result.
1196    9.  The restrictions on what a pattern can contain when partial matching is
1197        requested for pcre_exec() have been removed. All patterns can now be
1198        partially matched by this function. In addition, if there are at least two
1199        slots in the offset vector, the offset of the earliest inspected character
1200        for the match and the offset of the end of the subject are set in them when
1201        PCRE_ERROR_PARTIAL is returned.
1203    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1204        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1205        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1206        and may be more useful for multi-segment matching.
1208    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1209        used to be given if ever the end of the subject was reached; now it is
1210        given only if matching could not proceed because another character was
1211        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1212        string "Z", which now yields "no match" instead of "partial match". In the
1213        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1214        final character ended with (*FAIL).
1216    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1217        if the pattern had a "must contain" character that was already found in the
1218        earlier partial match, unless partial matching was again requested. For
1219        example, with the pattern /dog.(body)?/, the "must contain" character is
1220        "g". If the first part-match was for the string "dog", restarting with
1221        "sbody" failed. This bug has been fixed.
1223    13. The string returned by pcre_dfa_exec() after a partial match has been
1224        changed so that it starts at the first inspected character rather than the
1225        first character of the match. This makes a difference only if the pattern
1226        starts with a lookbehind assertion or \b or \B (\K is not supported by
1227        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1228        matching functions compatible, and I think it's the right thing to do.
1230    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1231        so that the demonstration program is easily available in environments where
1232        PCRE has not been installed from source.
1234    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1235        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1236        library.
1238    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1239        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1240        is not the first non-POSIX option to be added. Clearly some people find
1241        these options useful.
1243    17. If a caller to the POSIX matching function regexec() passes a non-zero
1244        value for nmatch with a NULL value for pmatch, the value of
1245        nmatch is forced to zero.
1247    18. RunGrepTest did not have a test for the availability of the -u option of
1248        the diff command, as RunTest does. It now checks in the same way as
1249        RunTest, and also checks for the -b option.
1251    19. If an odd number of negated classes containing just a single character
1252        interposed, within parentheses, between a forward reference to a named
1253        subpattern and the definition of the subpattern, compilation crashed with
1254        an internal error, complaining that it could not find the referenced
1255        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1256        [The bug was that it was starting one character too far in when skipping
1257        over the character class, thus treating the ] as data rather than
1258        terminating the class. This meant it could skip too much.]
1260    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1261        /g option in pcretest when the pattern contains \K, which makes it possible
1262        to have an empty string match not at the start, even when the pattern is
1263        anchored. Updated pcretest and pcredemo to use this option.
1265    21. If the maximum number of capturing subpatterns in a recursion was greater
1266        than the maximum at the outer level, the higher number was returned, but
1267        with unset values at the outer level. The correct (outer level) value is
1268        now given.
1270    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1271        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1272        make it do so. The string so far is captured, making this feature
1273        compatible with Perl.
1275    23. The tests have been re-organized, adding tests 11 and 12, to make it
1276        possible to check the Perl 5.10 features against Perl 5.10.
1278    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1279        pattern matches a fixed length string. PCRE did not allow this; now it
1280        does. Neither allows recursion.
1282    25. I finally figured out how to implement a request to provide the minimum
1283        length of subject string that was needed in order to match a given pattern.
1284        (It was back references and recursion that I had previously got hung up
1285        on.) This code has now been added to pcre_study(); it finds a lower bound
1286        to the length of subject needed. It is not necessarily the greatest lower
1287        bound, but using it to avoid searching strings that are too short does give
1288        some useful speed-ups. The value is available to calling programs via
1289        pcre_fullinfo().
1291    26. While implementing 25, I discovered to my embarrassment that pcretest had
1292        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1293        study optimizations had never been tested with that matching function.
1294        Oops. What is worse, even when it was passed study data, there was a bug in
1295        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1296        were also very few tests of studied patterns with pcre_dfa_exec().
1298    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1299        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1300        on the other side of the coin, they are no longer allowed to have different
1301        names, because these cannot be distinguished in PCRE, and this has caused
1302        confusion. (This is a difference from Perl.)
1304    28. When duplicate subpattern names are present (necessarily with different
1305        numbers, as required by 27 above), and a test is made by name in a
1306        conditional pattern, either for a subpattern having been matched, or for
1307        recursion in such a pattern, all the associated numbered subpatterns are
1308        tested, and the overall condition is true if the condition is true for any
1309        one of them. This is the way Perl works, and is also more like the way
1310        testing by number works.
1313    Version 7.9 11-Apr-09
1314    ---------------------
1316    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1317        (pcretest), all targets were linked against these libraries. This included
1318        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1319        libraries. This caused unwanted dependencies to be created. This problem
1320        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1321        pcretest is linked with readline.
1323    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1324        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1325        moved outside it again, because FALSE and TRUE are already defined in AIX,
1326        but BOOL is not.
1328    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1329        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1331    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1332        hyphens) following file names and line numbers when outputting matching
1333        lines. This is not true; no spaces are inserted. I have also clarified the
1334        wording for the --colour (or --color) option.
1336    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1337        was not coloured; this is different to GNU grep, so I have changed it to be
1338        the same.
1340    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1341        each matching line was coloured. Now it goes on to look for further matches
1342        of any of the test patterns, which is the same behaviour as GNU grep.
1344    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1345        doesn't make sense to accept an empty string match in pcregrep, so I have
1346        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1347        seems to be how GNU grep behaves.
1349    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1350        start or after a newline", because the conditional assertion was not being
1351        correctly handled. The rule now is that both the assertion and what follows
1352        in the first alternative must satisfy the test.
1354    9.  If auto-callout was enabled in a pattern with a conditional group whose
1355        condition was an assertion, PCRE could crash during matching, both with
1356        pcre_exec() and pcre_dfa_exec().
1358    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1359        used for matching.
1361    11. Unicode property support in character classes was not working for
1362        characters (bytes) greater than 127 when not in UTF-8 mode.
1364    12. Added the -M command line option to pcretest.
1366    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1368    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1370    16. Added comments and documentation about mis-use of no_arg in the C++
1371        wrapper.
1373    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1374        from Martin Jerabek that uses macro names for all relevant character and
1375        string constants.
1377    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1378        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1379        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1380        these, but not everybody uses configure.
1382    19. A conditional group that had only one branch was not being correctly
1383        recognized as an item that could match an empty string. This meant that an
1384        enclosing group might also not be so recognized, causing infinite looping
1385        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1386        with the subject "ab", where knowledge that the repeated group can match
1387        nothing is needed in order to break the loop.
1389    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1390        exec(), but without supplying a callout function, matching went wrong.
1392    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1393        leak if the size of the offset vector was greater than 30. When the vector
1394        is smaller, the saved offsets during recursion go onto a local stack
1395        vector, but for larger vectors malloc() is used. It was failing to free
1396        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1397        error, in fact).
1399    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1400        heapframe that is used only when UTF-8 support is enabled. This caused no
1401        problem, but was untidy.
1403    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1404        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1405        included within another project.
1407    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1408        slightly modified by me:
1410          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1411              not building pcregrep.
1413          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1414              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1416    25. Forward references, both numeric and by name, in patterns that made use of
1417        duplicate group numbers, could behave incorrectly or give incorrect errors,
1418        because when scanning forward to find the reference group, PCRE was not
1419        taking into account the duplicate group numbers. A pattern such as
1420        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1422    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1423        the feature of a custom pointer more persuasive (as requested by a user).
1425    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1426        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1427        pre-defined.
1429    28. Added support for (*UTF8) at the start of a pattern.
1431    29. Arrange for flags added by the "release type" setting in CMake to be shown
1432        in the configuration summary.
1435    Version 7.8 05-Sep-08
1436    ---------------------
1438    1.  Replaced UCP searching code with optimized version as implemented for Ad
1439        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1440        stage table and inline lookup instead of a function, giving speed ups of 2
1441        to 5 times on some simple patterns that I tested. Permission was given to
1442        distribute the MultiStage2.py script that generates the tables (it's not in
1443        the tarball, but is in the Subversion repository).
1445    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1446        scripts.
1448    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1449        a group with a zero qualifier. The result of the study could be incorrect,
1450        or the function might crash, depending on the pattern.
1452    4.  Caseless matching was not working for non-ASCII characters in back
1453        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1454        It now works when Unicode Property Support is available.
1456    5.  In pcretest, an escape such as \x{de} in the data was always generating
1457        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1458        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1459        truncation.
1461    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1463    7.  Added two (int) casts to pcregrep when printing the difference of two
1464        pointers, in case they are 64-bit values.
1466    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1467        test 2 if it fails.
1469    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1470        and a #define of that name to empty if it is not externally set. This is to
1471        allow users of MSVC to set it if necessary.
1473    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1474        the convenience functions in the pcre_get.c source file.
1476    11. An option change at the start of a pattern that had top-level alternatives
1477        could cause overwriting and/or a crash. This command provoked a crash in
1478        some environments:
1480          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1482        This potential security problem was recorded as CVE-2008-2371.
1484    12. For a pattern where the match had to start at the beginning or immediately
1485        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1486        pcre_dfa_exec() could read past the end of the passed subject if there was
1487        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1488        pcretest so that it places the subject at the end of its malloc-ed buffer.
1490    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1491        exec() might read past the end of the data buffer in UTF-8 mode.
1493    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1494        the data contained the byte 0x85 as part of a UTF-8 character within its
1495        first line. This applied both to normal and DFA matching.
1497    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1498        /^[^d]*?$/8 failed to match "abc".
1500    16. Added a missing copyright notice to pcrecpp_internal.h.
1502    17. Make it more clear in the documentation that values returned from
1503        pcre_exec() in ovector are byte offsets, not character counts.
1505    18. Tidied a few places to stop certain compilers from issuing warnings.
1507    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1508        supplied by Stefan Weber. I made a further small update for 7.8 because
1509        there is a change of source arrangements: the pcre_searchfuncs.c module is
1510        replaced by pcre_ucd.c.
1513    Version 7.7 07-May-08
1514    ---------------------
1516    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1517        a string to a long long, pretend we don't even have a long long." This is
1518        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1520    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1521        pre-7.6 versions, which defined a global no_arg variable instead of putting
1522        it in the RE class. (See also #8 below.)
1524    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1525        Lopes.
1527    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1529        (1) The include/exclude patterns were being applied to the whole pathnames
1530            of files, instead of just to the final components.
1532        (2) If there was more than one level of directory, the subdirectories were
1533            skipped unless they satisfied the include/exclude conditions. This is
1534            inconsistent with GNU grep (and could even be seen as contrary to the
1535            pcregrep specification - which I improved to make it absolutely clear).
1536            The action now is always to scan all levels of directory, and just
1537            apply the include/exclude patterns to regular files.
1539    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1540        --exclude_dir in the tests to avoid scanning .svn directories.
1542    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1543        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1544        doesn't support NULs in patterns.
1546    7.  Added some missing "const"s to declarations of static tables in
1547        pcre_compile.c and pcre_dfa_exec.c.
1549    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1550        caused by fix #2  above. (Subsequently also a second patch to fix the
1551        first patch. And a third patch - this was a messy problem.)
1553    9.  Applied Craig's patch to remove the use of push_back().
1555    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1556        matching function regexec().
1558    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1559        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1560        references. PCRE supports relative numbers with this syntax (I don't think
1561        Oniguruma does).
1563    12. Previously, a group with a zero repeat such as (...){0} was completely
1564        omitted from the compiled regex. However, this means that if the group
1565        was called as a subroutine from elsewhere in the pattern, things went wrong
1566        (an internal error was given). Such groups are now left in the compiled
1567        pattern, with a new opcode that causes them to be skipped at execution
1568        time.
1570    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1571        to the way PCRE behaves:
1573        (a) A lone ] character is dis-allowed (Perl treats it as data).
1575        (b) A back reference to an unmatched subpattern matches an empty string
1576            (Perl fails the current match path).
1578        (c) A data ] in a character class must be notated as \] because if the
1579            first data character in a class is ], it defines an empty class. (In
1580            Perl it is not possible to have an empty class.) The empty class []
1581            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1582            The negative empty class [^] matches any one character, independently
1583            of the DOTALL setting.
1585    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1586        non-existent subpattern following a character class starting with ']' and
1587        containing () gave an internal compiling error instead of "reference to
1588        non-existent subpattern". Fortunately, when the pattern did exist, the
1589        compiled code was correct. (When scanning forwards to check for the
1590        existencd of the subpattern, it was treating the data ']' as terminating
1591        the class, so got the count wrong. When actually compiling, the reference
1592        was subsequently set up correctly.)
1594    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1595        it was being rejected as not supported by pcre_dfa_exec(), even though
1596        other assertions are supported. I have made pcre_dfa_exec() support
1597        (*FAIL).
1599    16. The implementation of 13c above involved the invention of a new opcode,
1600        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1601        cannot be changed at match time, I realized I could make a small
1602        improvement to matching performance by compiling OP_ALLANY instead of
1603        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1604        on the OP_ANY path.
1606    17. Compiling pcretest on Windows with readline support failed without the
1607        following two fixes: (1) Make the unistd.h include conditional on
1608        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1610    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1611        ncurses library to be included for pcretest when ReadLine support is
1612        requested, but also to allow for it to be overridden. This patch came from
1613        Daniel Bergström.
1615    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1616        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1617        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1618        spotting this.
1621    Version 7.6 28-Jan-08
1622    ---------------------
1624    1.  A character class containing a very large number of characters with
1625        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1626        overflow.
1628    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1629        HAVE_LONG_LONG is not defined.
1631    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1632        bring them up to date and include new features. This patch includes:
1634        - Fixed PH's badly added libz and libbz2 support.
1635        - Fixed a problem with static linking.
1636        - Added pcredemo. [But later removed - see 7 below.]
1637        - Fixed dftables problem and added an option.
1638        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1639            HAVE_LONG_LONG.
1640        - Added readline support for pcretest.
1641        - Added an listing of the option settings after cmake has run.
1643    4.  A user submitted a patch to Makefile that makes it easy to create
1644        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1645        Makefile.am that cause it to include this special target, without
1646        affecting anything else. Note that the same mingw target plus all
1647        the other distribution libraries and programs are now supported
1648        when configuring with CMake (see 6 below) instead of with
1649        Configure/Make.
1651    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1652        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1653        exported in the Windows port". It has not yet been confirmed that the patch
1654        solves the problem, but it does no harm.
1656    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1657        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1658        with CMake, and also correct the comment about stack recursion.
1660    7.  Remove the automatic building of pcredemo from the ./configure system and
1661        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1662        of a program that users should build themselves after PCRE is installed, so
1663        building it automatically is not really right. What is more, it gave
1664        trouble in some build environments.
1666    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1669    Version 7.5 10-Jan-08
1670    ---------------------
1672    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1673        values in parens when parsing an RE using the C++ wrapper."
1675    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1676        Characters greater than 255 were excluded from the class instead of being
1677        included.
1679    3.  The same bug as (2) above applied to negated POSIX classes such as
1680        [:^space:].
1682    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1683        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1684        I have changed it.
1686    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1687        first named subpattern) and a construct such as (?&a) would reference the
1688        first named subpattern whose name started with "a" (in other words, the
1689        length check was missing). Both these problems are fixed. "Subpattern name
1690        expected" is now given for (?&) (a zero-length name), and this patch also
1691        makes it give the same error for \k'' (previously it complained that that
1692        was a reference to a non-existent subpattern).
1694    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1695        this is right because (?- can be followed by option settings as well as by
1696        digits. I have, however, made the messages clearer.
1698    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1699        than the number used in the conditional) now cause a compile-time error.
1700        This is actually not compatible with Perl, which accepts such patterns, but
1701        treats the conditional as always being FALSE (as PCRE used to), but it
1702        seems to me that giving a diagnostic is better.
1704    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1705        and messages.
1707    9.  Fix two occurrences of "backslash" in comments that should have been
1708        "backspace".
1710    10. Remove two redundant lines of code that can never be obeyed (their function
1711        was moved elsewhere).
1713    11. The program that makes PCRE's Unicode character property table had a bug
1714        which caused it to generate incorrect table entries for sequences of
1715        characters that have the same character type, but are in different scripts.
1716        It amalgamated them into a single range, with the script of the first of
1717        them. In other words, some characters were in the wrong script. There were
1718        thirteen such cases, affecting characters in the following ranges:
1720          U+002b0 - U+002c1
1721          U+0060c - U+0060d
1722          U+0061e - U+00612
1723          U+0064b - U+0065e
1724          U+0074d - U+0076d
1725          U+01800 - U+01805
1726          U+01d00 - U+01d77
1727          U+01d9b - U+01dbf
1728          U+0200b - U+0200f
1729          U+030fc - U+030fe
1730          U+03260 - U+0327f
1731          U+0fb46 - U+0fbb1
1732          U+10450 - U+1049d
1734    12. The -o option (show only the matching part of a line) for pcregrep was not
1735        compatible with GNU grep in that, if there was more than one match in a
1736        line, it showed only the first of them. It now behaves in the same way as
1737        GNU grep.
1739    13. If the -o and -v options were combined for pcregrep, it printed a blank
1740        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1741        does the same. The return code can be used to tell if there were any
1742        non-matching lines.
1744    14. Added --file-offsets and --line-offsets to pcregrep.
1746    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1747        infinitely looping recursion. The bug was that positive lookaheads were not
1748        being skipped when checking for a possible empty match (negative lookaheads
1749        and both kinds of lookbehind were skipped).
1751    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1752        inclusion of <windows.h> to before rather than after the definition of
1753        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1755    17. Specifying a possessive quantifier with a specific limit for a Unicode
1756        character property caused pcre_compile() to compile bad code, which led at
1757        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1758        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1759        caused the error; without that there was no problem.
1761    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1763    19. Added --enable-pcretest-libreadline.
1765    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1766        RE::GlobalReplace(). As a result, the number of replacements returned was
1767        double what it should be. I removed one of the increments, but Craig sent a
1768        later patch that removed the other one (the right fix) and added unit tests
1769        that check the return values (which was not done before).
1771    21. Several CMake things:
1773        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1774            the names libpcre and libpcreposix, not just pcre and pcreposix.
1776        (2) The above change means that pcretest and pcregrep are now correctly
1777            linked with the newly-built libraries, not previously installed ones.
1781    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1782        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1783        UTF-8 newline character). The key issue is that the pattern starts .*;
1784        this means that the match must be either at the beginning, or after a
1785        newline. The bug was in the code for advancing after a failed match and
1786        checking that the new position followed a newline. It was not taking
1787        account of UTF-8 characters correctly.
1789    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1790        character classes. PCRE was not treating the sequence [:...:] as a
1791        character class unless the ... were all letters. Perl, however, seems to
1792        allow any characters between [: and :], though of course it rejects as
1793        unknown any "names" that contain non-letters, because all the known class
1794        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1795        for example, whereas PCRE did not - it did not recognize a POSIX character
1796        class. This seemed a bit dangerous, so the code has been changed to be
1797        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1798        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1799        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1800        Perl does, and where it didn't before.
1802    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1803        Windows environments %n is disabled by default.
1806    Version 7.4 21-Sep-07
1807    ---------------------
1809    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1810        means that a class such as [\s] counted as "explicit reference to CR or
1811        LF". That isn't really right - the whole point of the change was to try to
1812        help when there was an actual mention of one of the two characters. So now
1813        the change happens only if \r or \n (or a literal CR or LF) character is
1814        encountered.
1816    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1817        of both had grown to the point where there were only 3 bits left.
1818        Fortunately, there was spare space in the data structure, and so I have
1819        moved the internal flags into a new 16-bit field to free up more option
1820        bits.
1822    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1823        but did not set the internal JCHANGED flag - either of these is enough to
1824        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1825        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1826        start sets both bits.
1828    4.  Added options (at build time, compile time, exec time) to change \R from
1829        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1831    5.  doc/pcresyntax.html was missing from the distribution.
1833    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1834        compatibility, even though it is no longer used.
1836    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1837        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1838        windows.h file is present (where different names are used). [This was
1839        reversed later after testing - see 16 below.]
1841    8.  Changed all #include <config.h> to #include "config.h". There were also
1842        some further <pcre.h> cases that I changed to "pcre.h".
1844    9.  When pcregrep was used with the --colour option, it missed the line ending
1845        sequence off the lines that it output.
1847    10. It was pointed out to me that arrays of string pointers cause lots of
1848        relocations when a shared library is dynamically loaded. A technique of
1849        using a single long string with a table of offsets can drastically reduce
1850        these. I have refactored PCRE in four places to do this. The result is
1851        dramatic:
1853          Originally:                          290
1854          After changing UCP table:            187
1855          After changing error message table:   43
1856          After changing table of "verbs"       36
1857          After changing table of Posix names   22
1859        Thanks to the folks working on Gregex for glib for this insight.
1861    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1862        unicode-properties was also set.
1864    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1866    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1867        checked only for CRLF.
1869    14. Added casts to pcretest.c to avoid compiler warnings.
1871    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1873    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1874        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1875        entirely. This removes changes made in 7 above.
1877    17. The CMake files have been updated, and there is now more information about
1878        building with CMake in the NON-UNIX-USE document.
1881    Version 7.3 28-Aug-07
1882    ---------------------
1884     1. In the rejigging of the build system that eventually resulted in 7.1, the
1885        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1886        brackets there is not right, since it causes compilers to look for an
1887        installed pcre.h, not the version that is in the source that is being
1888        compiled (which of course may be different). I have changed it back to:
1890          #include "pcre.h"
1892        I have a vague recollection that the change was concerned with compiling in
1893        different directories, but in the new build system, that is taken care of
1894        by the VPATH setting the Makefile.
1896     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1897        when the subject happened to end in the byte 0x85 (e.g. if the last
1898        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1899        characters but of course it shouldn't be taken as a newline when it is part
1900        of another character. The bug was that, for an unlimited repeat of . in
1901        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1902        characters when looking for a newline.
1904     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1906     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1907        in debug output.
1909     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1910        long printing in the pcrecpp unittest when running under MinGW.
1912     6. ESC_K was left out of the EBCDIC table.
1914     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1915        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1916        limit also applies to "virtual nesting" when a pattern is recursive, and in
1917        this case 1000 isn't so big. I have been able to remove this limit at the
1918        expense of backing off one optimization in certain circumstances. Normally,
1919        when pcre_exec() would call its internal match() function recursively and
1920        immediately return the result unconditionally, it uses a "tail recursion"
1921        feature to save stack. However, when a subpattern that can match an empty
1922        string has an unlimited repetition quantifier, it no longer makes this
1923        optimization. That gives it a stack frame in which to save the data for
1924        checking that an empty string has been matched. Previously this was taken
1925        from the 1000-entry workspace that had been reserved. So now there is no
1926        explicit limit, but more stack is used.
1928     8. Applied Daniel's patches to solve problems with the import/export magic
1929        syntax that is required for Windows, and which was going wrong for the
1930        pcreposix and pcrecpp parts of the library. These were overlooked when this
1931        problem was solved for the main library.
1933     9. There were some crude static tests to avoid integer overflow when computing
1934        the size of patterns that contain repeated groups with explicit upper
1935        limits. As the maximum quantifier is 65535, the maximum group length was
1936        set at 30,000 so that the product of these two numbers did not overflow a
1937        32-bit integer. However, it turns out that people want to use groups that
1938        are longer than 30,000 bytes (though not repeat them that many times).
1939        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1940        made it possible to implement the integer overflow checks in a much more
1941        dynamic way, which I have now done. The artificial limitation on group
1942        length has been removed - we now have only the limit on the total length of
1943        the compiled pattern, which depends on the LINK_SIZE setting.
1945    10. Fixed a bug in the documentation for get/copy named substring when
1946        duplicate names are permitted. If none of the named substrings are set, the
1947        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1948        empty string.
1950    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1951        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1952        because the ] is interpreted as the first data character and the
1953        terminating ] is not found. PCRE has been made compatible with Perl in this
1954        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1955        cause memory overwriting.
1957    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1958        string has been matched (to stop an infinite loop). It was not recognizing
1959        a conditional subpattern that could match an empty string if that
1960        subpattern was within another subpattern. For example, it looped when
1961        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1962        condition was not nested. This bug has been fixed.
1964    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1965        past the start of the subject in the presence of bytes with the top bit
1966        set, for example "\x8aBCD".
1968    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1969        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1971    14. Optimized (?!) to (*FAIL).
1973    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1974        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1975        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1976        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1977        does: it's just the validity check that is more restrictive.
1979    16. Inserted checks for integer overflows during escape sequence (backslash)
1980        processing, and also fixed erroneous offset values for syntax errors during
1981        backslash processing.
1983    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1984        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1986    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1987        caused an overrun.
1989    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1990        something other than just ASCII characters) inside a group that had an
1991        unlimited repeat caused a loop at compile time (while checking to see
1992        whether the group could match an empty string).
1994    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1995        [\P{Any}] did so. (Error in the code for printing property names.)
1997    21. An orphan \E inside a character class could cause a crash.
1999    22. A repeated capturing bracket such as (A)? could cause a wild memory
2000        reference during compilation.
2002    23. There are several functions in pcre_compile() that scan along a compiled
2003        expression for various reasons (e.g. to see if it's fixed length for look
2004        behind). There were bugs in these functions when a repeated \p or \P was
2005        present in the pattern. These operators have additional parameters compared
2006        with \d, etc, and these were not being taken into account when moving along
2007        the compiled data. Specifically:
2009        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2010            length.
2012        (b) An item such as \pL+ within a repeated group could cause crashes or
2013            loops.
2015        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2016            "reference to non-existent subpattern" error.
2018        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2020    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2021        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2023    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2025    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2026        character were causing crashes (broken optimization).
2028    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2029        \p or \P) caused a compile-time loop.
2031    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2032        break. For example, the unstudied pattern [\r\n]A does not match the string
2033        "\r\nA" because change 7.0/46 below moves the current point on by two
2034        characters after failing to match at the start. However, the pattern \nA
2035        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2036        the same is true. There doesn't seem any very clean way out of this, but
2037        what I have chosen to do makes the common cases work: PCRE now takes note
2038        of whether there can be an explicit match for \r or \n anywhere in the
2039        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2040        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2041        pattern has explicit CR or LF references.
2043    29. Added (*CR) etc for changing newline setting at start of pattern.
2046    Version 7.2 19-Jun-07
2047    ---------------------
2049     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
2050        which is apparently normally available under Windows.
2052     2. Re-jig the pcregrep tests with different newline settings in an attempt
2053        to make them independent of the local environment's newline setting.
2055     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
2057     4. Some of the "internals" tests were previously cut out when the link size
2058        was not 2, because the output contained actual offsets. The recent new
2059        "Z" feature of pcretest means that these can be cut out, making the tests
2060        usable with all link sizes.
2062     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
2063        stack recursion. This gives a massive performance boost under BSD, but just
2064        a small improvement under Linux. However, it saves one field in the frame
2065        in all cases.
2067     6. Added more features from the forthcoming Perl 5.10:
2069        (a) (?-n) (where n is a string of digits) is a relative subroutine or
2070            recursion call. It refers to the nth most recently opened parentheses.
2072        (b) (?+n) is also a relative subroutine call; it refers to the nth next
2073            to be opened parentheses.
2075        (c) Conditions that refer to capturing parentheses can be specified
2076            relatively, for example, (?(-2)... or (?(+3)...
2078        (d) \K resets the start of the current match so that everything before
2079            is not part of it.
2081        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2083        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2084            reference syntax.
2086        (g) (?| introduces a group in which the numbering of parentheses in each
2087            alternative starts with the same number.
2089        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2091     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2094     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2095        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2096        for detecting groups that can match an empty string.
2098     9. A pattern with a very large number of alternatives (more than several
2099        hundred) was running out of internal workspace during the pre-compile
2100        phase, where pcre_compile() figures out how much memory will be needed. A
2101        bit of new cunning has reduced the workspace needed for groups with
2102        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2103        workspace instead of running out of the 4096 that are available.
2105    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2107    11. Applied patch from Google to remove an optimization that didn't quite work.
2108        The report of the bug said:
2110          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2111          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2112          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2114    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2115        it matched the wrong number of bytes.
2118    Version 7.1 24-Apr-07
2119    ---------------------
2121     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
2122        that is more "standard", making use of automake and other Autotools. There
2123        is some re-arrangement of the files and adjustment of comments consequent
2124        on this.
2126     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
2127        for recursive directory scanning broke on some systems because the files
2128        are not scanned in any specific order and on different systems the order
2129        was different. A call to "sort" has been inserted into RunGrepTest for the
2130        approprate test as a short-term fix. In the longer term there may be an
2131        alternative.
2133     3. I had an email from Eric Raymond about problems translating some of PCRE's
2134        man pages to HTML (despite the fact that I distribute HTML pages, some
2135        people do their own conversions for various reasons). The problems
2136        concerned the use of low-level troff macros .br and .in. I have therefore
2137        removed all such uses from the man pages (some were redundant, some could
2138        be replaced by .nf/.fi pairs). The 132html script that I use to generate
2139        HTML has been updated to handle .nf/.fi and to complain if it encounters
2140        .br or .in.
2142     4. Updated comments in configure.ac that get placed in config.h.in and also
2143        arranged for config.h to be included in the distribution, with the name
2144        config.h.generic, for the benefit of those who have to compile without
2145        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
2147     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
2148        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
2149        makevp.bat for the current PCRE, using the additional files
2150        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
2152     6. A Windows user reported a minor discrepancy with test 2, which turned out
2153        to be caused by a trailing space on an input line that had got lost in his
2154        copy. The trailing space was an accident, so I've just removed it.
2156     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
2157        that is needed.
2159     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
2160        as "const" (a) because they are and (b) because it helps the PHP
2161        maintainers who have recently made a script to detect big data structures
2162        in the php code that should be moved to the .rodata section. I remembered
2163        to update Builducptable as well, so it won't revert if ucptable.h is ever
2164        re-created.
2166     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
2167        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
2168        order to be able to cut out the UTF-8 tables in the latter when UTF-8
2169        support is not required. This saves 1.5-2K of code, which is important in
2170        some applications.
2172        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
2173        so as not to refer to the tables, even though these functions will never be
2174        called when UTF-8 support is disabled. Otherwise there are problems with a
2175        shared library.
2177    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
2179        (a) It was defining its arguments as char * instead of void *.
2181        (b) It was assuming that all moves were upwards in memory; this was true
2182            a long time ago when I wrote it, but is no longer the case.
2184        The emulated memove() is provided for those environments that have neither
2185        memmove() nor bcopy(). I didn't think anyone used it these days, but that
2186        is clearly not the case, as these two bugs were recently reported.
2188    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
2189        and Detrail to create the HTML documentation, the .txt form of the man
2190        pages, and it removes trailing spaces from listed files. It also creates
2191        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
2192        case, it wraps all the #defines with #ifndefs. This script should be run
2193        before "make dist".
2195    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
2196        with Unicode property support.
2198        (a) For a maximizing quantifier, if the two different cases of the
2199            character were of different lengths in their UTF-8 codings (there are
2200            some cases like this - I found 11), and the matching function had to
2201            back up over a mixture of the two cases, it incorrectly assumed they
2202            were both the same length.
2204        (b) When PCRE was configured to use the heap rather than the stack for
2205            recursion during matching, it was not correctly preserving the data for
2206            the other case of a UTF-8 character when checking ahead for a match
2207            while processing a minimizing repeat. If the check also involved
2208            matching a wide character, but failed, corruption could cause an
2209            erroneous result when trying to check for a repeat of the original
2210            character.
2212    13. Some tidying changes to the testing mechanism:
2214        (a) The RunTest script now detects the internal link size and whether there
2215            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
2216            values substituted by "configure". (The RunGrepTest script already did
2217            this for UTF-8.) The configure.ac script no longer substitutes the
2218            relevant variables.
2220        (b) The debugging options /B and /D in pcretest show the compiled bytecode
2221            with length and offset values. This means that the output is different
2222            for different internal link sizes. Test 2 is skipped for link sizes
2223            other than 2 because of this, bypassing the problem. Unfortunately,
2224            there was also a test in test 3 (the locale tests) that used /B and
2225            failed for link sizes other than 2. Rather than cut the whole test out,
2226            I have added a new /Z option to pcretest that replaces the length and
2227            offset values with spaces. This is now used to make test 3 independent
2228            of link size. (Test 2 will be tidied up later.)
2230    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
2231        segmentation fault instead of returning the appropriate error message.
2233    15. In multiline mode when the newline sequence was set to "any", the pattern
2234        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
2235        This doesn't seem right; it now treats the CRLF combination as the line
2236        ending, and so does not match in that case. It's only a pattern such as ^$
2237        that would hit this one: something like ^ABC$ would have failed after \r
2238        and then tried again after \r\n.
2240    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
2241        in an attempt to make files that differ only in their line terminators
2242        compare equal. This works on Linux.
2244    17. Under certain error circumstances pcregrep might try to free random memory
2245        as it exited. This is now fixed, thanks to valgrind.
2247    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
2248        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
2249        was because its rules for how to advance for /g after matching an empty
2250        string at the end of a line did not allow for this case. They now check for
2251        it specially.
2253    20. pcretest is supposed to handle patterns and data of any length, by
2254        extending its buffers when necessary. It was getting this wrong when the
2255        buffer for a data line had to be extended.
2257    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
2258        CRLF as a newline sequence.
2260    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
2261        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
2262        I have nevertheless tidied it up.
2264    23. Added some casts to kill warnings from HP-UX ia64 compiler.
2266    24. Added a man page for pcre-config.
2269    Version 7.0 19-Dec-06
2270    ---------------------
2272     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
2273        moving to gcc 4.1.1.
2275     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
2276        sys/time.h, which is documented as needed for this function. It doesn't
2277        seem to matter on Linux, but it showed up on some releases of OS X.
2279     3. It seems that there are systems where bytes whose values are greater than
2280        127 match isprint() in the "C" locale. The "C" locale should be the
2281        default when a C program starts up. In most systems, only ASCII printing
2282        characters match isprint(). This difference caused the output from pcretest
2283        to vary, making some of the tests fail. I have changed pcretest so that:
2285        (a) When it is outputting text in the compiled version of a pattern, bytes
2286            other than 32-126 are always shown as hex escapes.
2288        (b) When it is outputting text that is a matched part of a subject string,
2289            it does the same, unless a different locale has been set for the match
2290            (using the /L modifier). In this case, it uses isprint() to decide.
2292     4. Fixed a major bug that caused incorrect computation of the amount of memory
2293        required for a compiled pattern when options that changed within the
2294        pattern affected the logic of the preliminary scan that determines the
2295        length. The relevant options are -x, and -i in UTF-8 mode. The result was
2296        that the computed length was too small. The symptoms of this bug were
2297        either the PCRE error "internal error: code overflow" from pcre_compile(),
2298        or a glibc crash with a message such as "pcretest: free(): invalid next
2299        size (fast)". Examples of patterns that provoked this bug (shown in
2300        pcretest format) are:
2302          /(?-x: )/x
2303          /(?x)(?-x: \s*#\s*)/
2304          /((?i)[\x{c0}])/8
2305          /(?i:[\x{c0}])/8
2307        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
2308        is now done differently.
2310     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
2311        wrapper classes; (b) implement a new function in the C++ scanner that is
2312        more efficient than the old way of doing things because it avoids levels of
2313        recursion in the regex matching; (c) add a paragraph to the documentation
2314        for the FullMatch() function.
2316     6. The escape sequence \n was being treated as whatever was defined as
2317        "newline". Not only was this contrary to the documentation, which states
2318        that \n is character 10 (hex 0A), but it also went horribly wrong when
2319        "newline" was defined as CRLF. This has been fixed.
2321     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
2322        was being set to -1 for the "end of line" case (supposedly a value that no
2323        character can have). Though this value is never used (the check for end of
2324        line is "zero bytes in current character"), it caused compiler complaints.
2325        I've changed it to 0xffffffff.
2327     8. In pcre_version.c, the version string was being built by a sequence of
2328        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
2329        string (as it is for production releases) called a macro with an empty
2330        argument. The C standard says the result of this is undefined. The gcc
2331        compiler treats it as an empty string (which was what was wanted) but it is
2332        reported that Visual C gives an error. The source has been hacked around to
2333        avoid this problem.
2335     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
2336        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
2337        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
2338        of them did).
2340    10. Originally, pcretest opened its input and output without "b"; then I was
2341        told that "b" was needed in some environments, so it was added for release
2342        5.0 to both the input and output. (It makes no difference on Unix-like
2343        systems.) Later I was told that it is wrong for the input on Windows. I've
2344        now abstracted the modes into two macros, to make it easier to fiddle with
2345        them, and removed "b" from the input mode under Windows.
2347    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
2349    12. Added -help and --help to pcretest as an official way of being reminded
2350        of the options.
2352    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
2353        and pcrecpp.cc because they annoy compilers at high warning levels.
2355    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
2357    15. Fixed an occurrence of == in configure.ac that should have been = (shell
2358        scripts are not C programs :-) and which was not noticed because it works
2359        on Linux.
2361    16. pcretest is supposed to handle any length of pattern and data line (as one
2362        line or as a continued sequence of lines) by extending its input buffer if
2363        necessary. This feature was broken for very long pattern lines, leading to
2364        a string of junk being passed to pcre_compile() if the pattern was longer
2365        than about 50K.
2367    17. I have done a major re-factoring of the way pcre_compile() computes the
2368        amount of memory needed for a compiled pattern. Previously, there was code
2369        that made a preliminary scan of the pattern in order to do this. That was
2370        OK when PCRE was new, but as the facilities have expanded, it has become
2371        harder and harder to keep it in step with the real compile phase, and there
2372        have been a number of bugs (see for example, 4 above). I have now found a
2373        cunning way of running the real compile function in a "fake" mode that
2374        enables it to compute how much memory it would need, while actually only
2375        ever using a few hundred bytes of working memory and without too many
2376        tests of the mode. This should make future maintenance and development
2377        easier. A side effect of this work is that the limit of 200 on the nesting
2378        depth of parentheses has been removed (though this was never a serious
2379        limitation, I suspect). However, there is a downside: pcre_compile() now
2380        runs more slowly than before (30% or more, depending on the pattern). I
2381        hope this isn't a big issue. There is no effect on runtime performance.
2383    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
2384        newline (only possible for the last line of a file) and it was a
2385        pattern that set a locale (followed by /Lsomething), pcretest crashed.
2387    19. Added additional timing features to pcretest. (1) The -tm option now times
2388        matching only, not compiling. (2) Both -t and -tm can be followed, as a
2389        separate command line item, by a number that specifies the number of
2390        repeats to use when timing. The default is 50000; this gives better
2391        precision, but takes uncomfortably long for very large patterns.
2393    20. Extended pcre_study() to be more clever in cases where a branch of a
2394        subpattern has no definite first character. For example, (a*|b*)[cd] would
2395        previously give no result from pcre_study(). Now it recognizes that the
2396        first character must be a, b, c, or d.
2398    21. There was an incorrect error "recursive call could loop indefinitely" if
2399        a subpattern (or the entire pattern) that was being tested for matching an
2400        empty string contained only one non-empty item after a nested subpattern.
2401        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
2402        incorrectly, because the \d was being skipped in the check.
2404    22. The pcretest program now has a new pattern option /B and a command line
2405        option -b, which is equivalent to adding /B to every pattern. This causes
2406        it to show the compiled bytecode, without the additional information that
2407        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
2408        is the same as /B/I).
2410    23. A new optimization is now able automatically to treat some sequences such
2411        as a*b as a*+b. More specifically, if something simple (such as a character
2412        or a simple class like \d) has an unlimited quantifier, and is followed by
2413        something that cannot possibly match the quantified thing, the quantifier
2414        is automatically "possessified".
2416    24. A recursive reference to a subpattern whose number was greater than 39
2417        went wrong under certain circumstances in UTF-8 mode. This bug could also
2418        have affected the operation of pcre_study().
2420    25. Realized that a little bit of performance could be had by replacing
2421        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
2423    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
2425    27. Possessive quantifiers such as a++ were previously implemented by turning
2426        them into atomic groups such as ($>a+). Now they have their own opcodes,
2427        which improves performance. This includes the automatically created ones
2428        from 23 above.
2430    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
2431        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
2432        the first matched character to be a colon. This applied both to named and
2433        numbered groups.
2435    29. The ucpinternal.h header file was missing its idempotency #ifdef.
2437    30. I was sent a "project" file called libpcre.a.dev which I understand makes
2438        building PCRE on Windows easier, so I have included it in the distribution.
2440    31. There is now a check in pcretest against a ridiculously large number being
2441        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
2442        loop, the loop is abandoned.
2444    32. Forward references to subpatterns in conditions such as (?(2)...) where
2445        subpattern 2 is defined later cause pcre_compile() to search forwards in
2446        the pattern for the relevant set of parentheses. This search went wrong
2447        when there were unescaped parentheses in a character class, parentheses
2448        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2450    33. "Subroutine" calls and backreferences were previously restricted to
2451        referencing subpatterns earlier in the regex. This restriction has now
2452        been removed.
2454    34. Added a number of extra features that are going to be in Perl 5.10. On the
2455        whole, these are just syntactic alternatives for features that PCRE had
2456        previously implemented using the Python syntax or my own invention. The
2457        other formats are all retained for compatibility.
2459        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2460            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2461            also .NET compatible.
2463        (b) A recursion or subroutine call to a named group can now be defined as
2464            (?&name) as well as (?P>name).
2466        (c) A backreference to a named group can now be defined as \k<name> or
2467            \k'name' as well as (?P=name). The new forms, as well as being in Perl
2468            5.10, are also .NET compatible.
2470        (d) A conditional reference to a named group can now use the syntax
2471            (?(<name>) or (?('name') as well as (?(name).
2473        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2474            groups (named and numbered) that are never evaluated inline, but can be
2475            called as "subroutines" from elsewhere. In effect, the DEFINE condition
2476            is always false. There may be only one alternative in such a group.
2478        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2479            as the simple (?(R). The condition is true only if the most recent
2480            recursion is that of the given number or name. It does not search out
2481            through the entire recursion stack.
2483        (g) The escape \gN or \g{N} has been added, where N is a positive or
2484            negative number, specifying an absolute or relative reference.
2486    35. Tidied to get rid of some further signed/unsigned compiler warnings and
2487        some "unreachable code" warnings.
2489    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2490        things, this adds five new scripts.
2492    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2493        There were also incompatibilities regarding the handling of \Q..\E inside
2494        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2495        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2497    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2498        matches an empty string, and forcibly breaks the loop. There were bugs in
2499        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2500        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2501        separate and independent bugs (that affected different cases) have been
2502        fixed.
2504    39. Refactored the code to abolish the use of different opcodes for small
2505        capturing bracket numbers. This is a tidy that I avoided doing when I
2506        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2507        The new approach is not only tidier, it makes it possible to reduce the
2508        memory needed to fix the previous bug (38).
2510    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2511        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2512        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2513        mode.
2515    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2516        report.
2518    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2519        copy construction and assignment in the C++ wrapper.
2521    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2522        couple of bugs that could have given wrong results in the "--newline=crlf"
2523        case.
2525    44. Added a number of casts and did some reorganization of signed/unsigned int
2526        variables following suggestions from Dair Grant. Also renamed the variable
2527        "this" as "item" because it is a C++ keyword.
2529    45. Arranged for dftables to add
2531          #include "pcre_internal.h"
2533        to pcre_chartables.c because without it, gcc 4.x may remove the array
2534        definition from the final binary if PCRE is built into a static library and
2535        dead code stripping is activated.
2537    46. For an unanchored pattern, if a match attempt fails at the start of a
2538        newline sequence, and the newline setting is CRLF or ANY, and the next two
2539        characters are CRLF, advance by two characters instead of one.
2542    Version 6.7 04-Jul-06
2543    ---------------------
2545     1. In order to handle tests when input lines are enormously long, pcretest has
2546        been re-factored so that it automatically extends its buffers when
2547        necessary. The code is crude, but this _is_ just a test program. The
2548        default size has been increased from 32K to 50K.
2550     2. The code in pcre_study() was using the value of the re argument before
2551        testing it for NULL. (Of course, in any sensible call of the function, it
2552        won't be NULL.)
2554     3. The memmove() emulation function in pcre_internal.h, which is used on
2555        systems that lack both memmove() and bcopy() - that is, hardly ever -
2556        was missing a "static" storage class specifier.
2558     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2559        containing an extended class (one that cannot be represented by a bitmap
2560        because it contains high-valued characters or Unicode property items, e.g.
2561        [\pZ]). Almost always one would set UTF-8 mode when processing such a
2562        pattern, but PCRE should not loop if you do not (it no longer does).
2563        [Detail: two cases were found: (a) a repeated subpattern containing an
2564        extended class; (b) a recursive reference to a subpattern that followed a
2565        previous extended class. It wasn't skipping over the extended class
2566        correctly when UTF-8 mode was not set.]
2568     5. A negated single-character class was not being recognized as fixed-length
2569        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2570        compile error "lookbehind assertion is not fixed length".
2572     6. The RunPerlTest auxiliary script was showing an unexpected difference
2573        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2574        write a Perl script that can interpret lines of an input file either as
2575        byte characters or as UTF-8, which is what "perltest" was being required to
2576        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2577        can't do is switch easily at run time between having the "use utf8;" pragma
2578        or not. In the end, I fudged it by using the RunPerlTest script to insert
2579        "use utf8;" explicitly for the UTF-8 tests.
2581     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2582        the end of the subject string, contrary to the documentation and to what
2583        Perl does. This was true of both matching functions. Now it matches only at
2584        the start of the subject and immediately after *internal* newlines.
2586     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2587        a pointer to an int instead of a pointer to an unsigned long int. This
2588        caused problems on 64-bit systems.
2590     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2591        instance of the 'standard' template library not being so standard".
2593    10. There was no check on the number of named subpatterns nor the maximum
2594        length of a subpattern name. The product of these values is used to compute
2595        the size of the memory block for a compiled pattern. By supplying a very
2596        long subpattern name and a large number of named subpatterns, the size
2597        computation could be caused to overflow. This is now prevented by limiting
2598        the length of names to 32 characters, and the number of named subpatterns
2599        to 10,000.
2601    11. Subpatterns that are repeated with specific counts have to be replicated in
2602        the compiled pattern. The size of memory for this was computed from the
2603        length of the subpattern and the repeat count. The latter is limited to
2604        65535, but there was no limit on the former, meaning that integer overflow
2605        could in principle occur. The compiled length of a repeated subpattern is
2606        now limited to 30,000 bytes in order to prevent this.
2608    12. Added the optional facility to have named substrings with the same name.
2610    13. Added the ability to use a named substring as a condition, using the
2611        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2612        are numbers (not recommended). Forward references are permitted.
2614    14. Added forward references in named backreferences (if you see what I mean).
2616    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2617        pattern could run off the end of the subject. For example, the pattern
2618        "(?s)(.{1,5})"8 did this with the subject "ab".
2620    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2621        PCRE_CASELESS was set when matching characters that were quantified with ?
2622        or *.
2624    17. A character class other than a single negated character that had a minimum
2625        but no maximum quantifier - for example [ab]{6,} - was not handled
2626        correctly by pce_dfa_exec(). It would match only one character.
2628    18. A valid (though odd) pattern that looked like a POSIX character
2629        class but used an invalid character after [ (for example [[,abc,]]) caused
2630        pcre_compile() to give the error "Failed: internal error: code overflow" or
2631        in some cases to crash with a glibc free() error. This could even happen if
2632        the pattern terminated after [[ but there just happened to be a sequence of
2633        letters, a binary zero, and a closing ] in the memory that followed.
2635    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2636        over the years. Originally (before any Unicode support), just the bottom 8
2637        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2638        output from "man perlunicode" includes this:
2640          The regular expression compiler produces polymorphic opcodes.  That
2641          is, the pattern adapts to the data and automatically switches to
2642          the Unicode character scheme when presented with Unicode data--or
2643          instead uses a traditional byte scheme when presented with byte
2644          data.
2646        Sadly, a wide octal escape does not cause a switch, and in a string with
2647        no other multibyte characters, these octal escapes are treated as before.
2648        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2649        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2650        Unicode string.
2652        I have not perpetrated such confusion in PCRE. Up till now, it took just
2653        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2654        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2655        translate to the appropriate multibyte character.
2657    29. Applied some refactoring to reduce the number of warnings from Microsoft
2658        and Borland compilers. This has included removing the fudge introduced
2659        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2660        a warning about an unused variable.
2662    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2663        characters since release 4.0, because Perl (from release 5.004) does not.
2664        [Or at least, is documented not to: some releases seem to be in conflict
2665        with the documentation.] However, when a pattern was studied with
2666        pcre_study() and all its branches started with \s, PCRE still included VT
2667        as a possible starting character. Of course, this did no harm; it just
2668        caused an unnecessary match attempt.
2670    22. Removed a now-redundant internal flag bit that recorded the fact that case
2671        dependency changed within the pattern. This was once needed for "required
2672        byte" processing, but is no longer used. This recovers a now-scarce options
2673        bit. Also moved the least significant internal flag bit to the most-
2674        significant bit of the word, which was not previously used (hangover from
2675        the days when it was an int rather than a uint) to free up another bit for
2676        the future.
2678    23. Added support for CRLF line endings as well as CR and LF. As well as the
2679        default being selectable at build time, it can now be changed at runtime
2680        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2681        specify that it is scanning data with non-default line endings.
2683    24. Changed the definition of CXXLINK to make it agree with the definition of
2684        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2686    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2687        recursions. This makes a big different to stack usage for some patterns.
2689    26. If a subpattern containing a named recursion or subroutine reference such
2690        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2691        the space required for the compiled pattern went wrong and gave too small a
2692        value. Depending on the environment, this could lead to "Failed: internal
2693        error: code overflow at offset 49" or "glibc detected double free or
2694        corruption" errors.
2696    27. Applied patches from Google (a) to support the new newline modes and (b) to
2697        advance over multibyte UTF-8 characters in GlobalReplace.
2699    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2700        difference for some implementation of PCRE in some Windows version.
2702    29. Added some extra testing facilities to pcretest:
2704        \q<number>   in a data line sets the "match limit" value
2705        \Q<number>   in a data line sets the "match recursion limt" value
2706        -S <number>  sets the stack size, where <number> is in megabytes
2708        The -S option isn't available for Windows.
2711    Version 6.6 06-Feb-06
2712    ---------------------
2714     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2715        in pcreposix.h. I have copied the definition from pcre.h.
2717     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2718        because pcre.h is no longer a built file.
2720     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2721        not normally included in the compiled code.
2724    Version 6.5 01-Feb-06
2725    ---------------------
2727     1. When using the partial match feature with pcre_dfa_exec(), it was not
2728        anchoring the second and subsequent partial matches at the new starting
2729        point. This could lead to incorrect results. For example, with the pattern
2730        /1234/, partially matching against "123" and then "a4" gave a match.
2732     2. Changes to pcregrep:
2734        (a) All non-match returns from pcre_exec() were being treated as failures
2735            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2736            error message is output. Some extra information is given for the
2737            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2738            probably the only errors that are likely to be caused by users (by
2739            specifying a regex that has nested indefinite repeats, for instance).
2740            If there are more than 20 of these errors, pcregrep is abandoned.
2742        (b) A binary zero was treated as data while matching, but terminated the
2743            output line if it was written out. This has been fixed: binary zeroes
2744            are now no different to any other data bytes.
2746        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2747            used to set a locale for matching. The --locale=xxxx long option has
2748            been added (no short equivalent) to specify a locale explicitly on the
2749            pcregrep command, overriding the environment variables.
2751        (d) When -B was used with -n, some line numbers in the output were one less
2752            than they should have been.
2754        (e) Added the -o (--only-matching) option.
2756        (f) If -A or -C was used with -c (count only), some lines of context were
2757            accidentally printed for the final match.
2759        (g) Added the -H (--with-filename) option.
2761        (h) The combination of options -rh failed to suppress file names for files
2762            that were found from directory arguments.
2764        (i) Added the -D (--devices) and -d (--directories) options.
2766        (j) Added the -F (--fixed-strings) option.
2768        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2770        (l) Added the --colo(u)r option.
2772        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2773            is not present by default.
2775     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2776        items such as (?R) or (?1), when the recursion could match a number of
2777        alternatives. If it matched one of the alternatives, but subsequently,
2778        outside the recursion, there was a failure, the code tried to back up into
2779        the recursion. However, because of the way PCRE is implemented, this is not
2780        possible, and the result was an incorrect result from the match.
2782        In order to prevent this happening, the specification of recursion has
2783        been changed so that all such subpatterns are automatically treated as
2784        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2786     4. I had overlooked the fact that, in some locales, there are characters for
2787        which isalpha() is true but neither isupper() nor islower() are true. In
2788        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2789        and ordfeminine) are like this. This affected the treatment of \w and \W
2790        when they appeared in character classes, but not when they appeared outside
2791        a character class. The bit map for "word" characters is now created
2792        separately from the results of isalnum() instead of just taking it from the
2793        upper, lower, and digit maps. (Plus the underscore character, of course.)
2795     5. The above bug also affected the handling of POSIX character classes such as
2796        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2797        permanent tables. Instead, the bit maps for such a class were previously
2798        created as the appropriate unions of the upper, lower, and digit bitmaps.
2799        Now they are created by subtraction from the [[:word:]] class, which has
2800        its own bitmap.
2802     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2803        It is created by subtracting the vertical space characters (\x09, \x0a,
2804        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2805        subtraction was done in the overall bitmap for a character class, meaning
2806        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2807        be recognized. This bug has been fixed.
2809     7. Patches from the folks at Google:
2811          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2812          real life, but is still worth protecting against".
2814          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2815          regular expressions".
2817          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2818          have it.
2820          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2821          "configure" and the latter not, in order to fix a problem somebody had
2822          with compiling the Arg class on HP-UX.
2824          (e) Improve the error-handling of the C++ wrapper a little bit.
2826          (f) New tests for checking recursion limiting.
2828     8. The pcre_memmove() function, which is used only if the environment does not
2829        have a standard memmove() function (and is therefore rarely compiled),
2830        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2831        returning a result (though PCRE never actually uses the result).
2833     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2834        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2835        returned instead of calling malloc() with an overflowing number that would
2836        most likely cause subsequent chaos.
2838    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2840    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2841        with this option is matched, the nmatch and pmatch options of regexec() are
2842        ignored.
2844    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2845        provided in case anyone wants to the the POSIX interface with UTF-8
2846        strings.
2848    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2849        C++ linking (needed for some HP-UX environments).
2851    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2852        (unused parameter) and in the pcre_printint() function (omitted "default"
2853        switch label when the default is to do nothing).
2855    15. Added some code to make it possible, when PCRE is compiled as a C++
2856        library, to replace subject pointers for pcre_exec() with a smart pointer
2857        class, thus making it possible to process discontinuous strings.
2859    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2860        much the same function. They were added by different people who were trying
2861        to make PCRE easy to compile on non-Unix systems. It has been suggested
2862        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2863        for compiling on Windows systems. I have therefore replaced it with
2864        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2865        defaults to "extern" for C or "extern C" for C++, which works fine on
2866        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2867        SCOPE with something explicit in config.h. In addition:
2869        (a) pcreposix.h still had just "extern" instead of either of these macros;
2870            I have replaced it with PCRE_DATA_SCOPE.
2872        (b) Functions such as _pcre_xclass(), which are internal to the library,
2873            but external in the C sense, all had PCRE_EXPORT in their definitions.
2874            This is apparently wrong for the Windows case, so I have removed it.
2875            (It makes no difference on Unix-like systems.)
2877    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2878        of recursive calls to match(). This is different to MATCH_LIMIT because
2879        that limits the total number of calls to match(), not all of which increase
2880        the depth of recursion. Limiting the recursion depth limits the amount of
2881        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2882        when PCRE is compiled, and changed at run time. A patch from Google adds
2883        this functionality to the C++ interface.
2885    18. Changes to the handling of Unicode character properties:
2887        (a) Updated the table to Unicode 4.1.0.
2889        (b) Recognize characters that are not in the table as "Cn" (undefined).
2891        (c) I revised the way the table is implemented to a much improved format
2892            which includes recognition of ranges. It now supports the ranges that
2893            are defined in UnicodeData.txt, and it also amalgamates other
2894            characters into ranges. This has reduced the number of entries in the
2895            table from around 16,000 to around 3,000, thus reducing its size
2896            considerably. I realized I did not need to use a tree structure after
2897            all - a binary chop search is just as efficient. Having reduced the
2898            number of entries, I extended their size from 6 bytes to 8 bytes to
2899            allow for more data.
2901        (d) Added support for Unicode script names via properties such as \p{Han}.
2903    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2904        matching that character.
2906    20. When matching a repeated Unicode property with a minimum greater than zero,
2907        (for example \pL{2,}), PCRE could look past the end of the subject if it
2908        reached it while seeking the minimum number of characters. This could
2909        happen only if some of the characters were more than one byte long, because
2910        there is a check for at least the minimum number of bytes.
2912    21. Refactored the implementation of \p and \P so as to be more general, to
2913        allow for more different types of property in future. This has changed the
2914        compiled form incompatibly. Anybody with saved compiled patterns that use
2915        \p or \P will have to recompile them.
2917    22. Added "Any" and "L&" to the supported property types.
2919    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2920        but give a compile time error if the value is greater than 0xff.
2922    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2923        accidentally not being installed or uninstalled.
2925    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2926        made were to insert the current release number. This seemed silly, because
2927        it made things harder for people building PCRE on systems that don't run
2928        "configure". I have turned pcre.h into a distributed file, no longer built
2929        by "configure", with the version identification directly included. There is
2930        no longer a pcre.h.in file.
2932        However, this change necessitated a change to the pcre-config script as
2933        well. It is built from pcre-config.in, and one of the substitutions was the
2934        release number. I have updated configure.ac so that ./configure now finds
2935        the release number by grepping pcre.h.
2937    26. Added the ability to run the tests under valgrind.
2940    Version 6.4 05-Sep-05
2941    ---------------------
2943     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2944        "--" to be printed when multiple files were scanned, even when none of the
2945        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2946        consider it to be a bug, and have restored the previous behaviour.
2948     2. A couple of code tidies to get rid of compiler warnings.
2950     3. The pcretest program used to cheat by referring to symbols in the library
2951        whose names begin with _pcre_. These are internal symbols that are not
2952        really supposed to be visible externally, and in some environments it is
2953        possible to suppress them. The cheating is now confined to including
2954        certain files from the library's source, which is a bit cleaner.
2956     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2957        file's purpose clearer.
2959     5. Reorganized pcre_ucp_findchar().
2962    Version 6.3 15-Aug-05
2963    ---------------------
2965     1. The file libpcre.pc.in did not have general read permission in the tarball.
2967     2. There were some problems when building without C++ support:
2969        (a) If C++ support was not built, "make install" and "make test" still
2970            tried to test it.
2972        (b) There were problems when the value of CXX was explicitly set. Some
2973            changes have been made to try to fix these, and ...
2975        (c) --disable-cpp can now be used to explicitly disable C++ support.
2977        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2978            backslash in a target when C++ was disabled. This confuses some
2979            versions of "make", apparently. Using an intermediate variable solves
2980            this. (Same for CPP_LOBJ.)
2982     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2983        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2984        necessary on certain architectures.
2986     4. Added a setting of -export-symbols-regex to the link command to remove
2987        those symbols that are exported in the C sense, but actually are local
2988        within the library, and not documented. Their names all begin with
2989        "_pcre_". This is not a perfect job, because (a) we have to except some
2990        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2991        available (and never for static libraries). I have made a note to try to
2992        find a way round (a) in the future.
2995    Version 6.2 01-Aug-05
2996    ---------------------
2998     1. There was no test for integer overflow of quantifier values. A construction
2999        such as {1111111111111111} would give undefined results. What is worse, if
3000        a minimum quantifier for a parenthesized subpattern overflowed and became
3001        negative, the calculation of the memory size went wrong. This could have
3002        led to memory overwriting.
3004     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
3006     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
3007        operating environments where this matters.
3009     4. Applied Giuseppe Maxia's patch to add additional features for controlling
3010        PCRE options from within the C++ wrapper.
3012     5. Named capturing subpatterns were not being correctly counted when a pattern
3013        was compiled. This caused two problems: (a) If there were more than 100
3014        such subpatterns, the calculation of the memory needed for the whole
3015        compiled pattern went wrong, leading to an overflow error. (b) Numerical
3016        back references of the form \12, where the number was greater than 9, were
3017        not recognized as back references, even though there were sufficient
3018        previous subpatterns.
3020     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
3021        versions of gcc, e.g. 2.95.4.
3024    Version 6.1 21-Jun-05
3025    ---------------------
3027     1. There was one reference to the variable "posix" in pcretest.c that was not
3028        surrounded by "#if !defined NOPOSIX".
3030     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
3031        the cross-check on the old pcre_info() function, for the benefit of the
3032        cut-down version of PCRE that is currently imported into Exim.
3034     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
3035        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
3036        patterns that start (?i) but I don't think that matters. The use of (?i) is
3037        just an example; this all applies to the other options as well.
3039     4. Since libtool seems to echo the compile commands it is issuing, the output
3040        from "make" can be reduced a bit by putting "@" in front of each libtool
3041        compile command.
3043     5. Patch from the folks at Google for configure.in to be a bit more thorough
3044        in checking for a suitable C++ installation before trying to compile the
3045        C++ stuff. This should fix a reported problem when a compiler was present,
3046        but no suitable headers.
3048     6. The man pages all had just "PCRE" as their title. I have changed them to
3049        be the relevant file name. I have also arranged that these names are
3050        retained in the file doc/pcre.txt, which is a concatenation in text format
3051        of all the man pages except the little individual ones for each function.
3053     7. The NON-UNIX-USE file had not been updated for the different set of source
3054        files that come with release 6. I also added a few comments about the C++
3055        wrapper.
3058    Version 6.0 07-Jun-05
3059    ---------------------
3061     1. Some minor internal re-organization to help with my DFA experiments.
3063     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
3064        didn't matter for the library itself when fully configured, but did matter
3065        when compiling without UCP support, or within Exim, where the ucp files are
3066        not imported.
3068     3. Refactoring of the library code to split up the various functions into
3069        different source modules. The addition of the new DFA matching code (see
3070        below) to a single monolithic source would have made it really too
3071        unwieldy, quite apart from causing all the code to be include in a
3072        statically linked application, when only some functions are used. This is
3073        relevant even without the DFA addition now that patterns can be compiled in
3074        one application and matched in another.
3076        The downside of splitting up is that there have to be some external
3077        functions and data tables that are used internally in different modules of
3078        the library but which are not part of the API. These have all had their
3079        names changed to start with "_pcre_" so that they are unlikely to clash
3080        with other external names.
3082     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
3083        a different (DFA) algorithm. Although it is slower than the original
3084        function, it does have some advantages for certain types of matching
3085        problem.
3087     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
3088        including restarting after a partial match.
3090     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
3091        defined when compiling for Windows was sent to me. I have put it into the
3092        code, though I have no means of testing or verifying it.
3094     7. Added the pcre_refcount() auxiliary function.
3096     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
3097        match before or at the first newline in the subject string. In pcretest,
3098        the /f option on a pattern can be used to set this.
3100     9. A repeated \w when used in UTF-8 mode with characters greater than 256
3101        would behave wrongly. This has been present in PCRE since release 4.0.
3103    10. A number of changes to the pcregrep command:
3105        (a) Refactored how -x works; insert ^(...)$ instead of setting
3106            PCRE_ANCHORED and checking the length, in preparation for adding
3107            something similar for -w.
3109        (b) Added the -w (match as a word) option.
3111        (c) Refactored the way lines are read and buffered so as to have more
3112            than one at a time available.
3114        (d) Implemented a pcregrep test script.
3116        (e) Added the -M (multiline match) option. This allows patterns to match
3117            over several lines of the subject. The buffering ensures that at least
3118            8K, or the rest of the document (whichever is the shorter) is available
3119            for matching (and similarly the previous 8K for lookbehind assertions).
3121        (f) Changed the --help output so that it now says
3123              -w, --word-regex(p)
3125            instead of two lines, one with "regex" and the other with "regexp"
3126            because that confused at least one person since the short forms are the
3127            same. (This required a bit of code, as the output is generated
3128            automatically from a table. It wasn't just a text change.)
3130        (g) -- can be used to terminate pcregrep options if the next thing isn't an
3131            option but starts with a hyphen. Could be a pattern or a path name
3132            starting with a hyphen, for instance.
3134        (h) "-" can be given as a file name to represent stdin.
3136        (i) When file names are being printed, "(standard input)" is used for
3137            the standard input, for compatibility with GNU grep. Previously
3138            "<stdin>" was used.
3140        (j) The option --label=xxx can be used to supply a name to be used for
3141            stdin when file names are being printed. There is no short form.
3143        (k) Re-factored the options decoding logic because we are going to add
3144            two more options that take data. Such options can now be given in four
3145            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
3147        (l) Added the -A, -B, and -C options for requesting that lines of context
3148            around matches be printed.
3150        (m) Added the -L option to print the names of files that do not contain
3151            any matching lines, that is, the complement of -l.
3153        (n) The return code is 2 if any file cannot be opened, but pcregrep does
3154            continue to scan other files.
3156        (o) The -s option was incorrectly implemented. For compatibility with other
3157            greps, it now suppresses the error message for a non-existent or non-
3158            accessible file (but not the return code). There is a new option called
3159            -q that suppresses the output of matching lines, which was what -s was
3160            previously doing.
3162        (p) Added --include and --exclude options to specify files for inclusion
3163            and exclusion when recursing.
3165    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
3166        Hopefully, it now does.
3168    12. Missing cast in pcre_study().
3170    13. Added an "uninstall" target to the makefile.
3172    14. Replaced "extern" in the function prototypes in Makefile.in with
3173        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
3174        world, but is set differently for Windows.
3176    15. Added a second compiling function called pcre_compile2(). The only
3177        difference is that it has an extra argument, which is a pointer to an
3178        integer error code. When there is a compile-time failure, this is set
3179        non-zero, in addition to the error test pointer being set to point to an
3180        error message. The new argument may be NULL if no error number is required
3181        (but then you may as well call pcre_compile(), which is now just a
3182        wrapper). This facility is provided because some applications need a
3183        numeric error indication, but it has also enabled me to tidy up the way
3184        compile-time errors are handled in the POSIX wrapper.
3186    16. Added VPATH=.libs to the makefile; this should help when building with one
3187        prefix path and installing with another. (Or so I'm told by someone who
3188        knows more about this stuff than I do.)
3190    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
3191        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
3192        match everything, including newlines. This is not POSIX-compatible, but
3193        somebody wanted the feature. From pcretest it can be activated by using
3194        both the P and the s flags.
3196    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
3198    19. libpcre.pc was being incorrectly installed as executable.
3200    20. A couple of places in pcretest check for end-of-line by looking for '\n';
3201        it now also looks for '\r' so that it will work unmodified on Windows.
3203    21. Added Google's contributed C++ wrapper to the distribution.
3205    22. Added some untidy missing memory free() calls in pcretest, to keep
3206        Electric Fence happy when testing.
3210    Version 5.0 13-Sep-04
3211    ---------------------
3213     1. Internal change: literal characters are no longer packed up into items
3214        containing multiple characters in a single byte-string. Each character
3215        is now matched using a separate opcode. However, there may be more than one
3216        byte in the character in UTF-8 mode.
3218     2. The pcre_callout_block structure has two new fields: pattern_position and
3219        next_item_length. These contain the offset in the pattern to the next match
3220        item, and its length, respectively.
3222     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
3223        insertion of callouts before each pattern item. Added the /C option to
3224        pcretest to make use of this.
3226     4. On the advice of a Windows user, the lines
3228          #if defined(_WIN32) || defined(WIN32)
3229          _setmode( _fileno( stdout ), 0x8000 );
3230          #endif  /* defined(_WIN32) || defined(WIN32) */
3232        have been added to the source of pcretest. This apparently does useful
3233        magic in relation to line terminators.
3235     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
3236        for the benefit of those environments where the "b" makes a difference.
3238     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
3239        to know about it. I have put a hack into configure.in that adds in code
3240        to set GCC=yes if CC=icc. This seems to end up at a point in the
3241        generated configure script that is early enough to affect the setting of
3242        compiler options, which is what is needed, but I have no means of testing
3243        whether it really works. (The user who reported this had patched the
3244        generated configure script, which of course I cannot do.)
3246        LATER: After change 22 below (new libtool files), the configure script
3247        seems to know about icc (and also ecc). Therefore, I have commented out
3248        this hack in configure.in.
3250     7. Added support for pkg-config (2 patches were sent in).
3252     8. Negated POSIX character classes that used a combination of internal tables
3253        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
3254        [[:^ascii]]. Typically, they would match almost any characters. The other
3255        POSIX classes were not broken in this way.
3257     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
3258        to find the match, as PCRE was deluded into thinking that the match had to
3259        start at the start point or following a newline. The same bug applied to
3260        patterns with negative forward assertions or any backward assertions
3261        preceding ".*" at the start, unless the pattern required a fixed first
3262        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
3264    10. In UTF-8 mode, when moving forwards in the subject after a failed match
3265        starting at the last subject character, bytes beyond the end of the subject
3266        string were read.
3268    11. Renamed the variable "class" as "classbits" to make life easier for C++
3269        users. (Previously there was a macro definition, but it apparently wasn't
3270        enough.)
3272    12. Added the new field "tables" to the extra data so that tables can be passed
3273        in at exec time, or the internal tables can be re-selected. This allows
3274        a compiled regex to be saved and re-used at a later time by a different
3275        program that might have everything at different addresses.
3277    13. Modified the pcre-config script so that, when run on Solaris, it shows a
3278        -R library as well as a -L library.
3280    14. The debugging options of pcretest (-d on the command line or D on a
3281        pattern) showed incorrect output for anything following an extended class
3282        that contained multibyte characters and which was followed by a quantifier.
3284    15. Added optional support for general category Unicode character properties
3285        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
3286        support. It adds about 90K to the size of the library. The meanings of the
3287        inbuilt class escapes such as \d and \s have NOT been changed.
3289    16. Updated pcredemo.c to include calls to free() to release the memory for the
3290        compiled pattern.
3292    17. The generated file chartables.c was being created in the source directory
3293        instead of in the building directory. This caused the build to fail if the
3294        source directory was different from the building directory, and was
3295        read-only.
3297    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
3298        file. No doubt somebody will tell me if they don't make sense... Also added
3299        Dan Mooney's comments about building on OpenVMS.
3301    19. Added support for partial matching via the PCRE_PARTIAL option for
3302        pcre_exec() and the \P data escape in pcretest.
3304    20. Extended pcretest with 3 new pattern features:
3306        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
3307              write the compiled pattern to the file whose name is "rest-of-line".
3308              This is a straight binary dump of the data, with the saved pointer to
3309              the character tables forced to be NULL. The study data, if any, is
3310              written too. After writing, pcretest reads a new pattern.
3312        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
3313              compiled pattern from the given file. There must not be any
3314              occurrences of "<" in the file name (pretty unlikely); if there are,
3315              pcretest will instead treat the initial "<" as a pattern delimiter.
3316              After reading in the pattern, pcretest goes on to read data lines as
3317              usual.
3319        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
3320              and 16-bit fields in a compiled pattern, to simulate a pattern that
3321              was compiled on a host of opposite endianness.
3323    21. The pcre-exec() function can now cope with patterns that were compiled on
3324        hosts of opposite endianness, with this restriction:
3326          As for any compiled expression that is saved and used later, the tables
3327          pointer field cannot be preserved; the extra_data field in the arguments
3328          to pcre_exec() should be used to pass in a tables address if a value
3329          other than the default internal tables were used at compile time.
3331    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
3332        now diagnosed as an error. Previously, most of the time, a negative number
3333        would have been treated as zero, but if in addition "ovector" was passed as
3334        NULL, a crash could occur.
3336    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
3337        new versions from the libtool 1.5 distribution (the last one is a copy of
3338        a file called libtool.m4). This seems to have fixed the need to patch
3339        "configure" to support Darwin 1.3 (which I used to do). However, I still
3340        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
3341        workstation).
3343    24. Changed the PCRE licence to be the more standard "BSD" licence.
3346    Version 4.5 01-Dec-03
3347    ---------------------
3349     1. There has been some re-arrangement of the code for the match() function so
3350        that it can be compiled in a version that does not call itself recursively.
3351        Instead, it keeps those local variables that need separate instances for
3352        each "recursion" in a frame on the heap, and gets/frees frames whenever it
3353        needs to "recurse". Keeping track of where control must go is done by means
3354        of setjmp/longjmp. The whole thing is implemented by a set of macros that
3355        hide most of the details from the main code, and operates only if
3356        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
3357        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
3358        operating.
3360        To make it easier for callers to provide specially tailored get/free
3361        functions for this usage, two new functions, pcre_stack_malloc, and
3362        pcre_stack_free, are used. They are always called in strict stacking order,
3363        and the size of block requested is always the same.
3365        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
3366        PCRE has been compiled to use the stack or the heap for recursion. The
3367        -C option of pcretest uses this to show which version is compiled.
3369        A new data escape \S, is added to pcretest; it causes the amounts of store
3370        obtained and freed by both kinds of malloc/free at match time to be added
3371        to the output.
3373     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
3374        what's available on my current Linux desktop machine.
3376     3. When matching a UTF-8 string, the test for a valid string at the start has
3377        been extended. If start_offset is not zero, PCRE now checks that it points
3378        to a byte that is the start of a UTF-8 character. If not, it returns
3379        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
3380        this is necessary because there may be backward assertions in the pattern.
3381        When matching the same subject several times, it may save resources to use
3382        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
3384     4. The code for checking the validity of UTF-8 strings has been tightened so
3385        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
3386        containing "overlong sequences".
3388     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
3389        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
3390        should have been "&", but it just so happened that all the cases this let
3391        through by mistake were picked up later in the function.
3393     6. I had used a variable called "isblank" - this is a C99 function, causing
3394        some compilers to warn. To avoid this, I renamed it (as "blankclass").
3396     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
3397        prompting; (b) run "./pcretest /dev/null" at the start of the test script
3398        so the version is shown; (c) stop "make test" echoing "./RunTest".
3400     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
3402     9. The prototype for memmove() for systems that don't have it was using
3403        size_t, but the inclusion of the header that defines size_t was later. I've
3404        moved the #includes for the C headers earlier to avoid this.
3406    10. Added some adjustments to the code to make it easier to compiler on certain
3407        special systems:
3409          (a) Some "const" qualifiers were missing.
3410          (b) Added the macro EXPORT before all exported functions; by default this
3411              is defined to be empty.
3412          (c) Changed the dftables auxiliary program (that builds chartables.c) so
3413              that it reads its output file name as an argument instead of writing
3414              to the standard output and assuming this can be redirected.
3416    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
3417        class containing characters with values greater than 255, PCRE compilation
3418        went into a loop.
3420    12. A recursive reference to a subpattern that was within another subpattern
3421        that had a minimum quantifier of zero caused PCRE to crash. For example,
3422        (x(y(?2))z)? provoked this bug with a subject that got as far as the
3423        recursion. If the recursively-called subpattern itself had a zero repeat,
3424        that was OK.
3426    13. In pcretest, the buffer for reading a data line was set at 30K, but the
3427        buffer into which it was copied (for escape processing) was still set at
3428        1024, so long lines caused crashes.
3430    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
3431        "internal error: code overflow...". This applied to any character class
3432        that was followed by a possessive quantifier.
3434    15. Modified the Makefile to add libpcre.la as a prerequisite for
3435        libpcreposix.la because I was told this is needed for a parallel build to
3436        work.
3438    16. If a pattern that contained .* following optional items at the start was
3439        studied, the wrong optimizing data was generated, leading to matching
3440        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
3441        matching string must start with a or b or c. The correct conclusion for
3442        this pattern is that a match can start with any character.
3445    Version 4.4 13-Aug-03
3446    ---------------------
3448     1. In UTF-8 mode, a character class containing characters with values between
3449        127 and 255 was not handled correctly if the compiled pattern was studied.
3450        In fixing this, I have also improved the studying algorithm for such
3451        classes (slightly).
3453     2. Three internal functions had redundant arguments passed to them. Removal
3454        might give a very teeny performance improvement.
3456     3. Documentation bug: the value of the capture_top field in a callout is *one
3457        more than* the number of the hightest numbered captured substring.
3459     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
3460        in incorrectly linking with a previously installed version. They now link
3461        explicitly with libpcre.la.
3463     5. configure.in no longer needs to recognize Cygwin specially.
3465     6. A problem in pcre.in for Windows platforms is fixed.
3467     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
3468        pcretest, it used to include the size of the study block as part of its
3469        output. Unfortunately, the structure contains a field that has a different
3470        size on different hardware architectures. This meant that the tests that
3471        showed this size failed. As the block is currently always of a fixed size,
3472        this information isn't actually particularly useful in pcretest output, so
3473        I have just removed it.
3475     8. Three pre-processor statements accidentally did not start in column 1.
3476        Sadly, there are *still* compilers around that complain, even though
3477        standard C has not required this for well over a decade. Sigh.
3479     9. In pcretest, the code for checking callouts passed small integers in the
3480        callout_data field, which is a void * field. However, some picky compilers
3481        complained about the casts involved for this on 64-bit systems. Now
3482        pcretest passes the address of the small integer instead, which should get
3483        rid of the warnings.
3485    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
3486        both compile and run time, and gives an error if an invalid UTF-8 sequence
3487        is found. There is a option for disabling this check in cases where the
3488        string is known to be correct and/or the maximum performance is wanted.
3490    11. In response to a bug report, I changed one line in Makefile.in from
3492            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
3493        to
3494            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
3496        to look similar to other lines, but I have no way of telling whether this
3497        is the right thing to do, as I do not use Windows. No doubt I'll get told
3498        if it's wrong...
3501    Version 4.3 21-May-03
3502    ---------------------
3504    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
3505       Makefile.
3507    2. Some refactoring to improve the quality of the code:
3509       (i)   The utf8_table... variables are now declared "const".
3511       (ii)  The code for \cx, which used the "case flipping" table to upper case
3512             lower case letters, now just substracts 32. This is ASCII-specific,
3513             but the whole concept of \cx is ASCII-specific, so it seems
3514             reasonable.
3516       (iii) PCRE was using its character types table to recognize decimal and
3517             hexadecimal digits in the pattern. This is silly, because it handles
3518             only 0-9, a-f, and A-F, but the character types table is locale-
3519             specific, which means strange things might happen. A private
3520             table is now used for this - though it costs 256 bytes, a table is
3521             much faster than multiple explicit tests. Of course, the standard
3522             character types table is still used for matching digits in subject
3523             strings against \d.
3525       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
3526             ending in _t are). So I've renamed it as ESC_tee.
3528    3. The first argument for regexec() in the POSIX wrapper should have been
3529       defined as "const".
3531    4. Changed pcretest to use malloc() for its buffers so that they can be
3532       Electric Fenced for debugging.
3534    5. There were several places in the code where, in UTF-8 mode, PCRE would try
3535       to read one or more bytes before the start of the subject string. Often this
3536       had no effect on PCRE's behaviour, but in some circumstances it could
3537       provoke a segmentation fault.
3539    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
3540       to try to read one or more bytes before the start of the subject string.
3542    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3543       UTF-8 support could misbehave in various ways if the subject string
3544       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3545       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3546       back over UTF-8 characters.)
3549    Version 4.2 14-Apr-03
3550    ---------------------
3552    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
3554    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
3555         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
3556         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
3557         [WIN_PREFIX]: new variable, "cyg" for Cygwin
3558         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
3559           and BUILD_EXEEXT
3560         Note: automatic setting of the BUILD variables is not yet working
3561         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
3562           compile-time but not at link-time
3563         [LINK]: use for linking executables only
3564         make different versions for Windows and non-Windows
3565         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
3566           libraries
3567         [LINK_FOR_BUILD]: new variable
3568         [OBJEXT]: use throughout
3569         [EXEEXT]: use throughout
3570         <winshared>: new target
3571         <wininstall>: new target
3572         <dftables.o>: use native compiler
3573         <dftables>: use native linker
3574         <install>: handle Windows platform correctly
3575         <clean>: ditto
3576         <check>: ditto
3577         copy DLL to top builddir before testing
3579       As part of these changes, -no-undefined was removed again. This was reported
3580       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3581       in any case.
3583    3. Some tidies to get rid of compiler warnings:
3585       . In the match_data structure, match_limit was an unsigned long int, whereas
3586         match_call_count was an int. I've made them both unsigned long ints.
3588       . In pcretest the fact that a const uschar * doesn't automatically cast to
3589         a void * provoked a warning.
3591       . Turning on some more compiler warnings threw up some "shadow" variables
3592         and a few more missing casts.
3594    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3595       option, a class that contained a single character with a value between 128
3596       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3598    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3599       option, a class that contained several characters, but with at least one
3600       whose value was between 128 and 255 caused PCRE to crash.