ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 53 by nigel, Sat Feb 24 21:39:42 2007 UTC revision 980 by ph10, Mon Jun 18 16:47:14 2012 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.31 02-June-2012
5    -------------------------
7    1.  Fixing a wrong JIT test case and some compiler warnings.
9    2.  Removed a bashism from the RunTest script.
11    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
12        to unsigned type, result still unsigned" that was given by an MS compiler
13        on encountering the code "-sizeof(xxx)".
15    4.  Partial matching support is added to the JIT compiler.
17    5.  Fixed several bugs concerned with partial matching of items that consist
18        of more than one character:
20        (a) /^(..)\1/ did not partially match "aba" because checking references was
21            done on an "all or nothing" basis. This also applied to repeated
22            references.
24        (b) \R did not give a hard partial match if \r was found at the end of the
25            subject.
27        (c) \X did not give a hard partial match after matching one or more
28            characters at the end of the subject.
30        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
31            a partial match for the string "\r".
33        (e) When newline was set to CRLF, the metacharacter "." did not recognize
34            a partial match for a CR character at the end of the subject string.
36    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
37        running pcretest, the text "(JIT)" added to the output whenever JIT is
38        actually used to run the match.
40    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
41        or /S+[+] with a digit between 1 and 7.
43    8.  OP_NOT now supports any UTF character not just single-byte ones.
45    9.  (*MARK) control verb is now supported by the JIT compiler.
47    10. The command "./RunTest list" lists the available tests without actually
48        running any of them. (Because I keep forgetting what they all are.)
52    12. Applied a (slightly modified) user-supplied patch that improves performance
53        when the heap is used for recursion (compiled with --disable-stack-for-
54        recursion). Instead of malloc and free for each heap frame each time a
55        logical recursion happens, frames are retained on a chain and re-used where
56        possible. This sometimes gives as much as 30% improvement.
58    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
59        call.
61    14. As documented, (*COMMIT) is now confined to within a positive assertion.
63    15. It is now possible to link pcretest with libedit as an alternative to
64        libreadline.
66    16. (*COMMIT) control verb is now supported by the JIT compiler.
68    17. The Unicode data tables have been updated to Unicode 6.1.0.
70    18. Added --file-list option to pcregrep.
72    19. Added binary file support to pcregrep, including the -a, --binary-files,
73        -I, and --text options.
75    20. The madvise function is renamed for posix_madvise for QNX compatibility
76        reasons. Fixed by Giuseppe D'Angelo.
78    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
80    22. Changed the option for creating symbolic links for 16-bit man pages from
81        -s to -sf so that re-installing does not cause issues.
83    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
85    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
86        matches in certain environments (the workspace was not being correctly
87        retained). Also added to pcre_dfa_exec() a simple plausibility check on
88        some of the workspace data at the beginning of a restart.
90    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
91        was not doing so when it should - probably a typo introduced by SVN 528
92        (change 8.10/14).
94    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
95        \w+ when the character tables indicated that \x{c4} was a word character.
96        There were several related cases, all because the tests for doing a table
97        lookup were testing for characters less than 127 instead of 255.
99    27. If a pattern contains capturing parentheses that are not used in a match,
100        their slots in the ovector are set to -1. For those that are higher than
101        any matched groups, this happens at the end of processing. In the case when
102        there were back references that the ovector was too small to contain
103        (causing temporary malloc'd memory to be used during matching), and the
104        highest capturing number was not used, memory off the end of the ovector
105        was incorrectly being set to -1. (It was using the size of the temporary
106        memory instead of the true size.)
108    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
109        ovector size, it uses memory at the end of the block that it has got.
111    29. Check for an overlong MARK name and give an error at compile time. The
112        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
114    30. JIT compiler update.
116    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
117        Rill for the patch.
119    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
121    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
123    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
124        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
125        was enabled.
127    35. Improve JIT code generation for greedy plus quantifier.
129    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
130        1 to "aa" instead of to an empty string. The bug affected repeated groups
131        that could potentially match an empty string.
133    37. Optimizing single character iterators in JIT.
135    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
136        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
137        codepoints that are too big for the mode are faulted, and in a UTF mode,
138        disallowed codepoints are also faulted.
140    39. If PCRE was compiled with UTF support, in three places in the DFA
141        matcher there was code that should only have been obeyed in UTF mode, but
142        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
143        processing when bytes with values greater than 127 were present. In 16-bit
144        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
145        both cases the values are those that cannot be the first data item in a UTF
146        character. The three items that might have provoked this were recursions,
147        possessively repeated groups, and atomic groups.
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.
154    Version 8.30 04-February-2012
155    -----------------------------
157    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
158        name is defined in ctype.h.
160    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
161        only in quite long subpatterns.
163    3.  Removed the function pcre_info(), which has been obsolete and deprecated
164        since it was replaced by pcre_fullinfo() in February 2000.
166    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
167        match a (*MARK), and the match failed at the start of the subject, a
168        reference to memory before the start of the subject could occur. This bug
169        was introduced by fix 17 of release 8.21.
171    5.  A reference to an unset group with zero minimum repetition was giving
172        totally wrong answers (in non-JavaScript-compatibility mode). For example,
173        /(another)?(\1?)test/ matched against "hello world test". This bug was
174        introduced in release 8.13.
176    6.  Add support for 16-bit character strings (a large amount of work involving
177        many changes and refactorings).
179    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
180        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
181        from a file.
183    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
184        rounding is not applied in this particular case).
186    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
187        if they appear, or are escaped, in patterns.
189    10. Get rid of a number of -Wunused-but-set-variable warnings.
191    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
192        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
193        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
194        also returns the mark "x". This bug applied to capturing parentheses,
195        non-capturing parentheses, and atomic parentheses. It also applied to some
196        assertions.
198    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
199        information out of configure.ac instead of relying on pcre.h.generic, which
200        is not stored in the repository.
202    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
203        -lreadline.
205    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
207    15. Applied Graycode's patch to put the top-level frame on the stack rather
208        than the heap when not using the stack for recursion. This gives a
209        performance improvement in many cases when recursion is not deep.
211    16. Experimental code added to "pcretest -C" to output the stack frame size.
214    Version 8.21 12-Dec-2011
215    ------------------------
217    1.  Updating the JIT compiler.
219    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
220        are added as well.
222    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
223        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
224        calling _pcre_jit_exec. Some extra comments are added.
226    4.  (*MARK) settings inside atomic groups that do not contain any capturing
227        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
228        was introduced by change 18 for 8.20.
230    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
231        ECMA-262 standard.
233    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
234        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
235        This bug was probably introduced by change 9 of 8.13.
237    7.  While fixing 6 above, I noticed that a number of other items were being
238        incorrectly rejected as "not fixed length". This arose partly because newer
239        opcodes had not been added to the fixed-length checking code. I have (a)
240        corrected the bug and added tests for these items, and (b) arranged for an
241        error to occur if an unknown opcode is encountered while checking for fixed
242        length instead of just assuming "not fixed length". The items that were
243        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
244        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
245        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
247    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
248        being incorrectly compiled and would have given unpredicatble results.
250    9.  A possessively repeated subpattern with minimum repeat count greater than
251        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
252        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
253        the first (A) could occur when it should not.
255    10. Add a cast and remove a redundant test from the code.
257    11. JIT should use pcre_malloc/pcre_free for allocation.
259    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
260        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
261        is anything other than /usr, -L is still shown).
263    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
265    14. Perl does not support \N without a following name in a [] class; PCRE now
266        also gives an error.
268    15. If a forward reference was repeated with an upper limit of around 2000,
269        it caused the error "internal error: overran compiling workspace". The
270        maximum number of forward references (including repeats) was limited by the
271        internal workspace, and dependent on the LINK_SIZE. The code has been
272        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
273        the default depends on LINK_SIZE. There is a new upper limit (for safety)
274        of around 200,000 forward references. While doing this, I also speeded up
275        the filling in of repeated forward references.
277    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
278        incorrectly expecting the subject to contain another "a" after the start.
280    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
281        in the match, the SKIP should be ignored. This was not happening; instead
282        the SKIP was being treated as NOMATCH. For patterns such as
283        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
284        tested.
286    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
287        now much more compatible with Perl, in particular in cases where the result
288        is a non-match for a non-anchored pattern. For example, if
289        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
290        "m", where previously it did not return a name. A side effect of this
291        change is that for partial matches, the last encountered mark name is
292        returned, as for non matches. A number of tests that were previously not
293        Perl-compatible have been moved into the Perl-compatible test files. The
294        refactoring has had the pleasing side effect of removing one argument from
295        the match() function, thus reducing its stack requirements.
297    19. If the /S+ option was used in pcretest to study a pattern using JIT,
298        subsequent uses of /S (without +) incorrectly behaved like /S+.
300    21. Retrieve executable code size support for the JIT compiler and fixing
301        some warnings.
303    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
304        not work when the shorter character appeared right at the end of the
305        subject string.
307    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
308        systems.
310    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
311        output it when the /M option is used in pcretest.
313    25. The CheckMan script was not being included in the distribution. Also, added
314        an explicit "perl" to run Perl scripts from the PrepareRelease script
315        because this is reportedly needed in Windows.
317    26. If study data was being save in a file and studying had not found a set of
318        "starts with" bytes for the pattern, the data written to the file (though
319        never used) was taken from uninitialized memory and so caused valgrind to
320        complain.
322    27. Updated RunTest.bat as provided by Sheri Pierce.
324    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
326    29. Computation of memory usage for the table of capturing group names was
327        giving an unnecessarily large value.
330    Version 8.20 21-Oct-2011
331    ------------------------
333    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
334        a POSIX class. After further experiments with Perl, which convinced me that
335        Perl has bugs and confusions, a closing square bracket is no longer allowed
336        in a POSIX name. This bug also affected patterns with classes that started
337        with full stops.
339    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
340        captured substring, but while checking the failing first alternative,
341        substring 1 is temporarily captured. If the output vector supplied to
342        pcre_exec() was not big enough for this capture, the yield of the function
343        was still zero ("insufficient space for captured substrings"). This cannot
344        be totally fixed without adding another stack variable, which seems a lot
345        of expense for a edge case. However, I have improved the situation in cases
346        such as /(a)(b)x|abc/ matched against "abc", where the return code
347        indicates that fewer than the maximum number of slots in the ovector have
348        been set.
350    3.  Related to (2) above: when there are more back references in a pattern than
351        slots in the output vector, pcre_exec() uses temporary memory during
352        matching, and copies in the captures as far as possible afterwards. It was
353        using the entire output vector, but this conflicts with the specification
354        that only 2/3 is used for passing back captured substrings. Now it uses
355        only the first 2/3, for compatibility. This is, of course, another edge
356        case.
358    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
359        main code base, and can be used by building with --enable-jit. When this is
360        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
361        runtime --no-jit option is given.
363    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
364        ovector, the return from the function was zero, implying that there were
365        other matches that did not fit. The correct "exactly full" value is now
366        returned.
368    6.  If a subpattern that was called recursively or as a subroutine contained
369        (*PRUNE) or any other control that caused it to give a non-standard return,
370        invalid errors such as "Error -26 (nested recursion at the same subject
371        position)" or even infinite loops could occur.
373    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
374        computing the minimum length on reaching *ACCEPT, and so ended up with the
375        wrong value of 1 rather than 0. Further investigation indicates that
376        computing a minimum subject length in the presence of *ACCEPT is difficult
377        (think back references, subroutine calls), and so I have changed the code
378        so that no minimum is registered for a pattern that contains *ACCEPT.
380    8.  If (*THEN) was present in the first (true) branch of a conditional group,
381        it was not handled as intended. [But see 16 below.]
383    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
384        Sheri Pierce.
386    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
387        the first byte in a match must be "a".
389    11. Change 17 for 8.13 increased the recursion depth for patterns like
390        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
391        pattern contains any instances of (*THEN). If it does not, the old
392        optimizations are restored. It would be nice to do this on a per-group
393        basis, but at the moment that is not feasible.
395    12. In some environments, the output of pcretest -C is CRLF terminated. This
396        broke RunTest's code that checks for the link size. A single white space
397        character after the value is now allowed for.
399    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
400        For "fr", it uses the Windows-specific input and output files.
402    14. If (*THEN) appeared in a group that was called recursively or as a
403        subroutine, it did not work as intended. [But see next item.]
405    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
406        pattern fragments (but not containing any | characters). If A and B are
407        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
408        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
409        D. In other words, Perl considers parentheses that do not contain any |
410        characters to be part of a surrounding alternative, whereas PCRE was
411        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
412        differently. PCRE now behaves in the same way as Perl, except in the case
413        of subroutine/recursion calls such as (?1) which have in any case always
414        been different (but PCRE had them first :-).
416    16. Related to 15 above: Perl does not treat the | in a conditional group as
417        creating alternatives. Such a group is treated in the same way as an
418        ordinary group without any | characters when processing (*THEN). PCRE has
419        been changed to match Perl's behaviour.
421    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
422        RunGrepTest script failed.
424    18. Change 22 for version 13 caused atomic groups to use more stack. This is
425        inevitable for groups that contain captures, but it can lead to a lot of
426        stack use in large patterns. The old behaviour has been restored for atomic
427        groups that do not contain any capturing parentheses.
429    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
430        suppress the check for a minimum subject length at run time. (If it was
431        given to pcre_exec() or pcre_dfa_exec() it did work.)
433    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
434        fail to work when decoding hex characters in data strings in EBCDIC
435        environments.
437    21. It appears that in at least one Mac OS environment, the isxdigit() function
438        is implemented as a macro that evaluates to its argument more than once,
439        contravening the C 90 Standard (I haven't checked a later standard). There
440        was an instance in pcretest which caused it to go wrong when processing
441        \x{...} escapes in subject strings. The has been rewritten to avoid using
442        things like p++ in the argument of isxdigit().
445    Version 8.13 16-Aug-2011
446    ------------------------
448    1.  The Unicode data tables have been updated to Unicode 6.0.0.
450    2.  Two minor typos in pcre_internal.h have been fixed.
452    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
453        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
454        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
456    4.  There were a number of related bugs in the code for matching backrefences
457        caselessly in UTF-8 mode when codes for the characters concerned were
458        different numbers of bytes. For example, U+023A and U+2C65 are an upper
459        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
460        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
461        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
462        2-byte code at the end of the subject (it thought there wasn't enough data
463        left).
465    5.  Comprehensive information about what went wrong is now returned by
466        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
467        as the output vector has at least 2 elements. The offset of the start of
468        the failing character and a reason code are placed in the vector.
470    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
471        now returned is for the first byte of the failing character, instead of the
472        last byte inspected. This is an incompatible change, but I hope it is small
473        enough not to be a problem. It makes the returned offset consistent with
474        pcre_exec() and pcre_dfa_exec().
476    7.  pcretest now gives a text phrase as well as the error number when
477        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
478        failure, the offset and reason code are output.
480    8.  When \R was used with a maximizing quantifier it failed to skip backwards
481        over a \r\n pair if the subsequent match failed. Instead, it just skipped
482        back over a single character (\n). This seems wrong (because it treated the
483        two characters as a single entity when going forwards), conflicts with the
484        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
485        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
486        has been changed.
488    9.  Some internal refactoring has changed the processing so that the handling
489        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
490        time (the PCRE_DOTALL option was changed this way some time ago: version
491        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
492        which was always a bit of a fudge. It also means that there is one less
493        argument for the match() function, which reduces its stack requirements
494        slightly. This change also fixes an incompatibility with Perl: the pattern
495        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
497    10. More internal refactoring has drastically reduced the number of recursive
498        calls to match() for possessively repeated groups such as (abc)++ when
499        using pcre_exec().
501    11. While implementing 10, a number of bugs in the handling of groups were
502        discovered and fixed:
504        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
505        (a|)*(?1) gave a compile-time internal error.
506        ((a|)+)+  did not notice that the outer group could match an empty string.
507        (^a|^)+   was not marked as anchored.
508        (.*a|.*)+ was not marked as matching at start or after a newline.
510    12. Yet more internal refactoring has removed another argument from the match()
511        function. Special calls to this function are now indicated by setting a
512        value in a variable in the "match data" data block.
514    13. Be more explicit in pcre_study() instead of relying on "default" for
515        opcodes that mean there is no starting character; this means that when new
516        ones are added and accidentally left out of pcre_study(), testing should
517        pick them up.
519    14. The -s option of pcretest has been documented for ages as being an old
520        synonym of -m (show memory usage). I have changed it to mean "force study
521        for every regex", that is, assume /S for every regex. This is similar to -i
522        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
523        using it. It makes it easier to run collections of tests with and without
524        study enabled, and thereby test pcre_study() more easily. All the standard
525        tests are now run with and without -s (but some patterns can be marked as
526        "never study" - see 20 below).
528    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
529        restoration of the capturing data to the outer values was not happening
530        correctly.
532    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
533        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
534        pattern had matched an empty string, and so incorrectly returned a no
535        match.
537    17. There was optimizing code for the last branch of non-capturing parentheses,
538        and also for the obeyed branch of a conditional subexpression, which used
539        tail recursion to cut down on stack usage. Unfortunately, now that there is
540        the possibility of (*THEN) occurring in these branches, tail recursion is
541        no longer possible because the return has to be checked for (*THEN). These
542        two optimizations have therefore been removed. [But see 8.20/11 above.]
544    18. If a pattern containing \R was studied, it was assumed that \R always
545        matched two bytes, thus causing the minimum subject length to be
546        incorrectly computed because \R can also match just one byte.
548    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
549        was incorrectly computed.
551    20. If /S is present twice on a test pattern in pcretest input, it now
552        *disables* studying, thereby overriding the use of -s on the command line
553        (see 14 above). This is necessary for one or two tests to keep the output
554        identical in both cases.
556    21. When (*ACCEPT) was used in an assertion that matched an empty string and
557        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
559    22. When an atomic group that contained a capturing parenthesis was
560        successfully matched, but the branch in which it appeared failed, the
561        capturing was not being forgotten if a higher numbered group was later
562        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
563        group 1 to "a", when in fact it should be unset. This applied to multi-
564        branched capturing and non-capturing groups, repeated or not, and also to
565        positive assertions (capturing in negative assertions does not happen
566        in PCRE) and also to nested atomic groups.
568    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
569        subject after a captured substring, to make it easier to tell which of a
570        number of identical substrings has been captured.
572    24. The way atomic groups are processed by pcre_exec() has been changed so that
573        if they are repeated, backtracking one repetition now resets captured
574        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
575        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
576        "aaa". Previously, it would have been "a". As part of this code
577        refactoring, the way recursive calls are handled has also been changed.
579    25. If an assertion condition captured any substrings, they were not passed
580        back unless some other capturing happened later. For example, if
581        (?(?=(a))a) was matched against "a", no capturing was returned.
583    26. When studying a pattern that contained subroutine calls or assertions,
584        the code for finding the minimum length of a possible match was handling
585        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
586        group 1 called group 2 while simultaneously a separate group 2 called group
587        1). A stack overflow occurred in this case. I have fixed this by limiting
588        the recursion depth to 10.
590    27. Updated RunTest.bat in the distribution to the version supplied by Tom
591        Fortmann. This supports explicit test numbers on the command line, and has
592        argument validation and error reporting.
594    28. An instance of \X with an unlimited repeat could fail if at any point the
595        first character it looked at was a mark character.
597    29. Some minor code refactoring concerning Unicode properties and scripts
598        should reduce the stack requirement of match() slightly.
600    30. Added the '=' option to pcretest to check the setting of unused capturing
601        slots at the end of the pattern, which are documented as being -1, but are
602        not included in the return count.
604    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
605        compiled something random. Now it gives a compile-time error (as does
606        Perl).
608    32. A *MARK encountered during the processing of a positive assertion is now
609        recorded and passed back (compatible with Perl).
611    33. If --only-matching or --colour was set on a pcregrep call whose pattern
612        had alternative anchored branches, the search for a second match in a line
613        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
614        matched the line "0102" twice. The same bug affected patterns that started
615        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
616        twice.
618    34. Previously, PCRE did not allow quantification of assertions. However, Perl
619        does, and because of capturing effects, quantifying parenthesized
620        assertions may at times be useful. Quantifiers are now allowed for
621        parenthesized assertions.
623    35. A minor code tidy in pcre_compile() when checking options for \R usage.
625    36. \g was being checked for fancy things in a character class, when it should
626        just be a literal "g".
628    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
629        appearance of a nested POSIX class supersedes an apparent external class.
630        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
631        unescaped square brackets may also appear as part of class names. For
632        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
633        more like Perl. (But see 8.20/1 above.)
635    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
636        was because it thought it was \N{name}, which is not supported).
638    39. Add minix to OS list not supporting the -S option in pcretest.
640    40. PCRE tries to detect cases of infinite recursion at compile time, but it
641        cannot analyze patterns in sufficient detail to catch mutual recursions
642        such as ((?1))((?2)). There is now a runtime test that gives an error if a
643        subgroup is called recursively as a subpattern for a second time at the
644        same position in the subject string. In previous releases this might have
645        been caught by the recursion limit, or it might have run out of stack.
647    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
648        happen only once. PCRE was, however incorrectly giving a compile time error
649        "recursive call could loop indefinitely" because it cannot analyze the
650        pattern in sufficient detail. The compile time test no longer happens when
651        PCRE is compiling a conditional subpattern, but actual runaway loops are
652        now caught at runtime (see 40 above).
654    42. It seems that Perl allows any characters other than a closing parenthesis
655        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
656        has been changed to be the same.
658    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
659        as not to get warnings when autogen.sh is called. Also changed
660        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
662    44. To help people who use pcregrep to scan files containing exceedingly long
663        lines, the following changes have been made:
665        (a) The default value of the buffer size parameter has been increased from
666            8K to 20K. (The actual buffer used is three times this size.)
668        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
669            PCRE is built.
671        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
672            to be set at run time.
674        (d) Numerical values in pcregrep options can be followed by K or M, for
675            example --buffer-size=50K.
677        (e) If a line being scanned overflows pcregrep's buffer, an error is now
678            given and the return code is set to 2.
680    45. Add a pointer to the latest mark to the callout data block.
682    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
683        partial match of an empty string instead of no match. This was specific to
684        the use of ".".
686    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
687        complete match instead of a partial match. This bug was dependent on both
688        the PCRE_UTF8 and PCRE_DOTALL options being set.
690    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
691        starting byte set, because \b was not being ignored.
694    Version 8.12 15-Jan-2011
695    ------------------------
697    1.  Fixed some typos in the markup of the man pages, and wrote a script that
698        checks for such things as part of the documentation building process.
700    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
701        --match-limit and --recursion-limit options (added for 8.11). In
702        particular, this made one of the standard tests fail. (The integer value
703        went into the wrong half of a long int.)
705    3.  If the --colour option was given to pcregrep with -v (invert match), it
706        did strange things, either producing crazy output, or crashing. It should,
707        of course, ignore a request for colour when reporting lines that do not
708        match.
710    4.  Another pcregrep bug caused similar problems if --colour was specified with
711        -M (multiline) and the pattern match finished with a line ending.
713    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
714        matched in multiline mode, the following line was shown as part of the
715        match. This seems wrong, so I have changed it.
717    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
718        the check for further matches in the same line (so they could be coloured)
719        to overrun the end of the current line. If another match was found, it was
720        incorrectly shown (and then shown again when found in the next line).
722    7.  If pcregrep was compiled under Windows, there was a reference to the
723        function pcregrep_exit() before it was defined. I am assuming this was
724        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
725        reported by a user. I've moved the definition above the reference.
728    Version 8.11 10-Dec-2010
729    ------------------------
731    1.  (*THEN) was not working properly if there were untried alternatives prior
732        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
733        backtracked to try for "b" instead of moving to the next alternative branch
734        at the same level (in this case, to look for "c"). The Perl documentation
735        is clear that when (*THEN) is backtracked onto, it goes to the "next
736        alternative in the innermost enclosing group".
738    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
739        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
740        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
741        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
742        (*THEN).
744    3.  If \s appeared in a character class, it removed the VT character from
745        the class, even if it had been included by some previous item, for example
746        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
747        of \s, but is part of the POSIX "space" class.)
749    4.  A partial match never returns an empty string (because you can always
750        match an empty string at the end of the subject); however the checking for
751        an empty string was starting at the "start of match" point. This has been
752        changed to the "earliest inspected character" point, because the returned
753        data for a partial match starts at this character. This means that, for
754        example, /(?<=abc)def/ gives a partial match for the subject "abc"
755        (previously it gave "no match").
757    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
758        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
759        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
760        has an implication that the given string is incomplete (because a partial
761        match is preferred over a full match). For this reason, these items now
762        give a partial match in this situation. [Aside: previously, the one case
763        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
764        match rather than a full match, which was wrong by the old rules, but is
765        now correct.]
767    6.  There was a bug in the handling of #-introduced comments, recognized when
768        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
769        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
770        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
771        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
772        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
773        places in pcre_compile().
775    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
776        comments when looking ahead for named forward references to subpatterns,
777        the only newline sequence it recognized was NL. It now handles newlines
778        according to the set newline convention.
780    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
781        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
782        cater for a lack of strerror(). These oversights have been fixed.
784    9.  Added --match-limit and --recursion-limit to pcregrep.
786    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
788    11. When the -o option was used, pcregrep was setting a return code of 1, even
789        when matches were found, and --line-buffered was not being honoured.
791    12. Added an optional parentheses number to the -o and --only-matching options
792        of pcregrep.
794    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
795        can match an empty string. The code to do it in pcretest and pcredemo
796        needed fixing:
798        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
799            only one byte after an empty string match just before CRLF (this case
800            just got forgotten; "any" and "anycrlf" were OK).
802        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
803            mode when an empty string match preceded an ASCII character followed by
804            a non-ASCII character. (The code for advancing by one character rather
805            than one byte was nonsense.)
807        (c) The pcredemo.c sample program did not have any code at all to handle
808            the cases when CRLF is a valid newline sequence.
810    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
811        as a starting offset was within the subject string. There is now a new
812        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
813        negative or greater than the length of the string. In order to test this,
814        pcretest is extended to allow the setting of negative starting offsets.
816    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
817        starting offset points to the beginning of a UTF-8 character was
818        unnecessarily clumsy. I tidied it up.
820    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
821        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
823    17. Nobody had reported that the --include_dir option, which was added in
824        release 7.7 should have been called --include-dir (hyphen, not underscore)
825        for compatibility with GNU grep. I have changed it to --include-dir, but
826        left --include_dir as an undocumented synonym, and the same for
827        --exclude-dir, though that is not available in GNU grep, at least as of
828        release 2.5.4.
830    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
831        characters from a string of bytes) have been redefined so as not to use
832        loops, in order to improve performance in some environments. At the same
833        time, I abstracted some of the common code into auxiliary macros to save
834        repetition (this should not affect the compiled code).
836    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
837        compile-time error is now given if \c is not followed by an ASCII
838        character, that is, a byte less than 128. (In EBCDIC mode, the code is
839        different, and any byte value is allowed.)
841    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
842        START_OPTIMIZE option, which is now allowed at compile time - but just
843        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
844        to pcregrep and other applications that have no direct access to PCRE
845        options. The new /Y option in pcretest sets this option when calling
846        pcre_compile().
848    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
849        back references. Groups containing recursive back references were forced to
850        be atomic by that change, but in the case of named groups, the amount of
851        memory required was incorrectly computed, leading to "Failed: internal
852        error: code overflow". This has been fixed.
854    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
855        pcretest.c, to avoid build problems in some Borland environments.
858    Version 8.10 25-Jun-2010
859    ------------------------
861    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
862        THEN.
864    2.  (*ACCEPT) was not working when inside an atomic group.
866    3.  Inside a character class, \B is treated as a literal by default, but
867        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
868        causes the error). The code is unchanged, but I tidied the documentation.
870    4.  Inside a character class, PCRE always treated \R and \X as literals,
871        whereas Perl faults them if its -w option is set. I have changed PCRE so
872        that it faults them when PCRE_EXTRA is set.
874    5.  Added support for \N, which always matches any character other than
875        newline. (It is the same as "." when PCRE_DOTALL is not set.)
877    6.  When compiling pcregrep with newer versions of gcc which may have
878        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
879        declared with attribute warn_unused_result" were given. Just casting the
880        result to (void) does not stop the warnings; a more elaborate fudge is
881        needed. I've used a macro to implement this.
883    7.  Minor change to pcretest.c to avoid a compiler warning.
885    8.  Added four artifical Unicode properties to help with an option to make
886        \s etc use properties (see next item). The new properties are: Xan
887        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
889    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
890        use Unicode properties. (*UCP) at the start of a pattern can be used to set
891        this option. Modified pcretest to add /W to test this facility. Added
892        REG_UCP to make it available via the POSIX interface.
894    10. Added --line-buffered to pcregrep.
896    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
897        studied, and the match started with a letter with a code point greater than
898        127 whose first byte was different to the first byte of the other case of
899        the letter, the other case of this starting letter was not recognized
900        (#976).
902    12. If a pattern that was studied started with a repeated Unicode property
903        test, for example, \p{Nd}+, there was the theoretical possibility of
904        setting up an incorrect bitmap of starting bytes, but fortunately it could
905        not have actually happened in practice until change 8 above was made (it
906        added property types that matched character-matching opcodes).
908    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
909        possible starting bytes for non-anchored patterns.
911    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
912        \R, and also a number of cases that involve Unicode properties, both
913        explicit and implicit when PCRE_UCP is set.
915    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
916        input, it could crash or give wrong results if characters with values
917        greater than 0xc0 were present in the subject string. (Detail: it assumed
918        UTF-8 input when processing these items.)
920    16. Added a lot of (int) casts to avoid compiler warnings in systems where
921        size_t is 64-bit (#991).
923    17. Added a check for running out of memory when PCRE is compiled with
924        --disable-stack-for-recursion (#990).
926    18. If the last data line in a file for pcretest does not have a newline on
927        the end, a newline was missing in the output.
929    19. The default pcre_chartables.c file recognizes only ASCII characters (values
930        less than 128) in its various bitmaps. However, there is a facility for
931        generating tables according to the current locale when PCRE is compiled. It
932        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
933        characters, are recognized by isspace() and therefore were getting set in
934        these tables, and indeed these tables seem to approximate to ISO 8859. This
935        caused a problem in UTF-8 mode when pcre_study() was used to create a list
936        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
937        which of course cannot start UTF-8 characters. I have changed the code so
938        that only real ASCII characters (less than 128) and the correct starting
939        bytes for UTF-8 encodings are set for characters greater than 127 when in
940        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
941        altogether.)
943    20. Added the /T option to pcretest so as to be able to run tests with non-
944        standard character tables, thus making it possible to include the tests
945        used for 19 above in the standard set of tests.
947    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
948        reference to a subpattern the other side of a comment that contains an
949        opening parenthesis caused either an internal compiling error, or a
950        reference to the wrong subpattern.
953    Version 8.02 19-Mar-2010
954    ------------------------
956    1.  The Unicode data tables have been updated to Unicode 5.2.0.
958    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
959        configured.
961    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
962        original author of that file, following a query about its status.
964    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
965        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
967    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
968        quantifier applied to a forward-referencing subroutine call, could compile
969        incorrect code or give the error "internal error: previously-checked
970        referenced subpattern not found".
972    6.  Both MS Visual Studio and Symbian OS have problems with initializing
973        variables to point to external functions. For these systems, therefore,
974        pcre_malloc etc. are now initialized to local functions that call the
975        relevant global functions.
977    7.  There were two entries missing in the vectors called coptable and poptable
978        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
979        I've fixed the data, and added a kludgy way of testing at compile time that
980        the lengths are correct (equal to the number of opcodes).
982    8.  Following on from 7, I added a similar kludge to check the length of the
983        eint vector in pcreposix.c.
985    9.  Error texts for pcre_compile() are held as one long string to avoid too
986        much relocation at load time. To find a text, the string is searched,
987        counting zeros. There was no check for running off the end of the string,
988        which could happen if a new error number was added without updating the
989        string.
991    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
993    11. \K was not working if it appeared in an atomic group or in a group that
994        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
995        \K is "not well defined" if used in an assertion. PCRE now accepts it if
996        the assertion is positive, but not if it is negative.
998    12. Change 11 fortuitously reduced the size of the stack frame used in the
999        "match()" function of pcre_exec.c by one pointer. Forthcoming
1000        implementation of support for (*MARK) will need an extra pointer on the
1001        stack; I have reserved it now, so that the stack frame size does not
1002        decrease.
1004    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1005        item in branch that calls a recursion is a subroutine call - as in the
1006        second branch in the above example - was incorrectly given the compile-
1007        time error "recursive call could loop indefinitely" because pcre_compile()
1008        was not correctly checking the subroutine for matching a non-empty string.
1010    14. The checks for overrunning compiling workspace could trigger after an
1011        overrun had occurred. This is a "should never occur" error, but it can be
1012        triggered by pathological patterns such as hundreds of nested parentheses.
1013        The checks now trigger 100 bytes before the end of the workspace.
1015    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1018    Version 8.01 19-Jan-2010
1019    ------------------------
1021    1.  If a pattern contained a conditional subpattern with only one branch (in
1022        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1023        computed the wrong minimum data length (which is of course zero for such
1024        subpatterns). This could cause incorrect "no match" results.
1026    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1027        the pattern is reset in the first branch, pcre_compile() failed with
1028        "internal error: code overflow at offset...". This happened only when
1029        the reset was to the original external option setting. (An optimization
1030        abstracts leading options settings into an external setting, which was the
1031        cause of this.)
1033    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1034        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1035        assertion pattern did not match (meaning that the assertion was true), it
1036        was incorrectly treated as false if the SKIP had been reached during the
1037        matching. This also applied to assertions used as conditions.
1039    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1040        assertion subpattern, including such a pattern used as a condition,
1041        unpredictable results occurred, instead of the error return
1044    5.  The C++ GlobalReplace function was not working like Perl for the special
1045        situation when an empty string is matched. It now does the fancy magic
1046        stuff that is necessary.
1048    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1049        removed. (These were left over from very, very early versions of PCRE.)
1051    7.  Some cosmetic changes to the code to make life easier when compiling it
1052        as part of something else:
1054        (a) Change DEBUG to PCRE_DEBUG.
1056        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1057            called "current" as "current_branch", to prevent a collision with the
1058            Linux macro when compiled as a kernel module.
1060        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1061            prevent a collision with the Linux macro when compiled as a kernel
1062            module.
1064    8.  In pcre_compile() there are some checks for integer overflows that used to
1065        cast potentially large values to (double). This has been changed to that
1066        when building, a check for int64_t is made, and if it is found, it is used
1067        instead, thus avoiding the use of floating point arithmetic. (There is no
1068        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1069        double.
1071    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1072        2005 (difference between two addresses compared to an unsigned value).
1074    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1075        custom one, because of the following reported problem in Windows:
1077          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1078              under Win32.
1079          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1080              therefore missing the function definition.
1081          - The compiler thus generates a "C" signature for the test function.
1082          - The linker fails to find the "C" function.
1083          - PCRE fails to configure if asked to do so against libbz2.
1085    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1086        messages were output:
1088          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1089          rerunning libtoolize, to keep the correct libtool macros in-tree.
1090          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1092        I have done both of these things.
1094    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1095        most of the time, it *can* run out if it is given a pattern that contains a
1096        runaway infinite recursion. I updated the discussion in the pcrestack man
1097        page.
1099    13. Now that we have gone to the x.xx style of version numbers, the minor
1100        version may start with zero. Using 08 or 09 is a bad idea because users
1101        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1102        interpreted as invalid octal numbers. I've updated the previous comment in
1103        configure.ac, and also added a check that gives an error if 08 or 09 are
1104        used.
1106    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1107        causing partial matching to fail when the end of the subject matched \W
1108        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1110    15. There were some discrepancies between the declarations in pcre_internal.h
1111        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1112        their definitions. The declarations used "const uschar *" and the
1113        definitions used USPTR. Even though USPTR is normally defined as "const
1114        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1115        reported that: "This difference in casting confuses some C++ compilers, for
1116        example, SunCC recognizes above declarations as different functions and
1117        generates broken code for hbpcre." I have changed the declarations to use
1118        USPTR.
1120    16. GNU libtool is named differently on some systems. The autogen.sh script now
1121        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1122        (FreeBSD).
1124    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1125        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1126        comment: "Figure out how to create a longlong from a string: strtoll and
1127        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1128        instance, but it only takes 2 args instead of 3!"
1130    18. A subtle bug concerned with back references has been fixed by a change of
1131        specification, with a corresponding code fix. A pattern such as
1132        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1133        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1134        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1135        same bug. Such groups have to be quantified to be useful, or contained
1136        inside another quantified group. (If there's no repetition, the reference
1137        can never match.) The problem arises because, having left the group and
1138        moved on to the rest of the pattern, a later failure that backtracks into
1139        the group uses the captured value from the final iteration of the group
1140        rather than the correct earlier one. I have fixed this in PCRE by forcing
1141        any group that contains a reference to itself to be an atomic group; that
1142        is, there cannot be any backtracking into it once it has completed. This is
1143        similar to recursive and subroutine calls.
1146    Version 8.00 19-Oct-09
1147    ----------------------
1149    1.  The table for translating pcre_compile() error codes into POSIX error codes
1150        was out-of-date, and there was no check on the pcre_compile() error code
1151        being within the table. This could lead to an OK return being given in
1152        error.
1154    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1155        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1156        in a Windows environment.
1158    3.  The pcregrep --count option prints the count for each file even when it is
1159        zero, as does GNU grep. However, pcregrep was also printing all files when
1160        --files-with-matches was added. Now, when both options are given, it prints
1161        counts only for those files that have at least one match. (GNU grep just
1162        prints the file name in this circumstance, but including the count seems
1163        more useful - otherwise, why use --count?) Also ensured that the
1164        combination -clh just lists non-zero counts, with no names.
1166    4.  The long form of the pcregrep -F option was incorrectly implemented as
1167        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1168        but it seems right to fix it, and I didn't think it was worth preserving
1169        the old behaviour.
1171    5.  The command line items --regex=pattern and --regexp=pattern were not
1172        recognized by pcregrep, which required --regex pattern or --regexp pattern
1173        (with a space rather than an '='). The man page documented the '=' forms,
1174        which are compatible with GNU grep; these now work.
1176    6.  No libpcreposix.pc file was created for pkg-config; there was just
1177        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1179    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1180        when UCP support is not needed, by modifying the Python script that
1181        generates it from Unicode data files. This should not matter if the module
1182        is correctly used as a library, but I received one complaint about 50K of
1183        unwanted data. My guess is that the person linked everything into his
1184        program rather than using a library. Anyway, it does no harm.
1186    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1187        was a minimum greater than 1 for a wide character in a possessive
1188        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1189        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1190        character. Chaos in the form of incorrect output or a compiling loop could
1191        result.
1193    9.  The restrictions on what a pattern can contain when partial matching is
1194        requested for pcre_exec() have been removed. All patterns can now be
1195        partially matched by this function. In addition, if there are at least two
1196        slots in the offset vector, the offset of the earliest inspected character
1197        for the match and the offset of the end of the subject are set in them when
1198        PCRE_ERROR_PARTIAL is returned.
1200    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1201        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1202        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1203        and may be more useful for multi-segment matching.
1205    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1206        used to be given if ever the end of the subject was reached; now it is
1207        given only if matching could not proceed because another character was
1208        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1209        string "Z", which now yields "no match" instead of "partial match". In the
1210        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1211        final character ended with (*FAIL).
1213    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1214        if the pattern had a "must contain" character that was already found in the
1215        earlier partial match, unless partial matching was again requested. For
1216        example, with the pattern /dog.(body)?/, the "must contain" character is
1217        "g". If the first part-match was for the string "dog", restarting with
1218        "sbody" failed. This bug has been fixed.
1220    13. The string returned by pcre_dfa_exec() after a partial match has been
1221        changed so that it starts at the first inspected character rather than the
1222        first character of the match. This makes a difference only if the pattern
1223        starts with a lookbehind assertion or \b or \B (\K is not supported by
1224        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1225        matching functions compatible, and I think it's the right thing to do.
1227    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1228        so that the demonstration program is easily available in environments where
1229        PCRE has not been installed from source.
1231    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1232        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1233        library.
1235    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1236        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1237        is not the first non-POSIX option to be added. Clearly some people find
1238        these options useful.
1240    17. If a caller to the POSIX matching function regexec() passes a non-zero
1241        value for nmatch with a NULL value for pmatch, the value of
1242        nmatch is forced to zero.
1244    18. RunGrepTest did not have a test for the availability of the -u option of
1245        the diff command, as RunTest does. It now checks in the same way as
1246        RunTest, and also checks for the -b option.
1248    19. If an odd number of negated classes containing just a single character
1249        interposed, within parentheses, between a forward reference to a named
1250        subpattern and the definition of the subpattern, compilation crashed with
1251        an internal error, complaining that it could not find the referenced
1252        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1253        [The bug was that it was starting one character too far in when skipping
1254        over the character class, thus treating the ] as data rather than
1255        terminating the class. This meant it could skip too much.]
1257    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1258        /g option in pcretest when the pattern contains \K, which makes it possible
1259        to have an empty string match not at the start, even when the pattern is
1260        anchored. Updated pcretest and pcredemo to use this option.
1262    21. If the maximum number of capturing subpatterns in a recursion was greater
1263        than the maximum at the outer level, the higher number was returned, but
1264        with unset values at the outer level. The correct (outer level) value is
1265        now given.
1267    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1268        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1269        make it do so. The string so far is captured, making this feature
1270        compatible with Perl.
1272    23. The tests have been re-organized, adding tests 11 and 12, to make it
1273        possible to check the Perl 5.10 features against Perl 5.10.
1275    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1276        pattern matches a fixed length string. PCRE did not allow this; now it
1277        does. Neither allows recursion.
1279    25. I finally figured out how to implement a request to provide the minimum
1280        length of subject string that was needed in order to match a given pattern.
1281        (It was back references and recursion that I had previously got hung up
1282        on.) This code has now been added to pcre_study(); it finds a lower bound
1283        to the length of subject needed. It is not necessarily the greatest lower
1284        bound, but using it to avoid searching strings that are too short does give
1285        some useful speed-ups. The value is available to calling programs via
1286        pcre_fullinfo().
1288    26. While implementing 25, I discovered to my embarrassment that pcretest had
1289        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1290        study optimizations had never been tested with that matching function.
1291        Oops. What is worse, even when it was passed study data, there was a bug in
1292        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1293        were also very few tests of studied patterns with pcre_dfa_exec().
1295    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1296        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1297        on the other side of the coin, they are no longer allowed to have different
1298        names, because these cannot be distinguished in PCRE, and this has caused
1299        confusion. (This is a difference from Perl.)
1301    28. When duplicate subpattern names are present (necessarily with different
1302        numbers, as required by 27 above), and a test is made by name in a
1303        conditional pattern, either for a subpattern having been matched, or for
1304        recursion in such a pattern, all the associated numbered subpatterns are
1305        tested, and the overall condition is true if the condition is true for any
1306        one of them. This is the way Perl works, and is also more like the way
1307        testing by number works.
1310    Version 7.9 11-Apr-09
1311    ---------------------
1313    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1314        (pcretest), all targets were linked against these libraries. This included
1315        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1316        libraries. This caused unwanted dependencies to be created. This problem
1317        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1318        pcretest is linked with readline.
1320    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1321        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1322        moved outside it again, because FALSE and TRUE are already defined in AIX,
1323        but BOOL is not.
1325    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1326        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1328    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1329        hyphens) following file names and line numbers when outputting matching
1330        lines. This is not true; no spaces are inserted. I have also clarified the
1331        wording for the --colour (or --color) option.
1333    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1334        was not coloured; this is different to GNU grep, so I have changed it to be
1335        the same.
1337    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1338        each matching line was coloured. Now it goes on to look for further matches
1339        of any of the test patterns, which is the same behaviour as GNU grep.
1341    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1342        doesn't make sense to accept an empty string match in pcregrep, so I have
1343        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1344        seems to be how GNU grep behaves.
1346    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1347        start or after a newline", because the conditional assertion was not being
1348        correctly handled. The rule now is that both the assertion and what follows
1349        in the first alternative must satisfy the test.
1351    9.  If auto-callout was enabled in a pattern with a conditional group whose
1352        condition was an assertion, PCRE could crash during matching, both with
1353        pcre_exec() and pcre_dfa_exec().
1355    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1356        used for matching.
1358    11. Unicode property support in character classes was not working for
1359        characters (bytes) greater than 127 when not in UTF-8 mode.
1361    12. Added the -M command line option to pcretest.
1363    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1365    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1367    16. Added comments and documentation about mis-use of no_arg in the C++
1368        wrapper.
1370    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1371        from Martin Jerabek that uses macro names for all relevant character and
1372        string constants.
1374    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1375        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1376        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1377        these, but not everybody uses configure.
1379    19. A conditional group that had only one branch was not being correctly
1380        recognized as an item that could match an empty string. This meant that an
1381        enclosing group might also not be so recognized, causing infinite looping
1382        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1383        with the subject "ab", where knowledge that the repeated group can match
1384        nothing is needed in order to break the loop.
1386    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1387        exec(), but without supplying a callout function, matching went wrong.
1389    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1390        leak if the size of the offset vector was greater than 30. When the vector
1391        is smaller, the saved offsets during recursion go onto a local stack
1392        vector, but for larger vectors malloc() is used. It was failing to free
1393        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1394        error, in fact).
1396    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1397        heapframe that is used only when UTF-8 support is enabled. This caused no
1398        problem, but was untidy.
1400    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1401        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1402        included within another project.
1404    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1405        slightly modified by me:
1407          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1408              not building pcregrep.
1410          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1411              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1413    25. Forward references, both numeric and by name, in patterns that made use of
1414        duplicate group numbers, could behave incorrectly or give incorrect errors,
1415        because when scanning forward to find the reference group, PCRE was not
1416        taking into account the duplicate group numbers. A pattern such as
1417        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1419    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1420        the feature of a custom pointer more persuasive (as requested by a user).
1422    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1423        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1424        pre-defined.
1426    28. Added support for (*UTF8) at the start of a pattern.
1428    29. Arrange for flags added by the "release type" setting in CMake to be shown
1429        in the configuration summary.
1432    Version 7.8 05-Sep-08
1433    ---------------------
1435    1.  Replaced UCP searching code with optimized version as implemented for Ad
1436        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1437        stage table and inline lookup instead of a function, giving speed ups of 2
1438        to 5 times on some simple patterns that I tested. Permission was given to
1439        distribute the MultiStage2.py script that generates the tables (it's not in
1440        the tarball, but is in the Subversion repository).
1442    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1443        scripts.
1445    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1446        a group with a zero qualifier. The result of the study could be incorrect,
1447        or the function might crash, depending on the pattern.
1449    4.  Caseless matching was not working for non-ASCII characters in back
1450        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1451        It now works when Unicode Property Support is available.
1453    5.  In pcretest, an escape such as \x{de} in the data was always generating
1454        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1455        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1456        truncation.
1458    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1460    7.  Added two (int) casts to pcregrep when printing the difference of two
1461        pointers, in case they are 64-bit values.
1463    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1464        test 2 if it fails.
1466    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1467        and a #define of that name to empty if it is not externally set. This is to
1468        allow users of MSVC to set it if necessary.
1470    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1471        the convenience functions in the pcre_get.c source file.
1473    11. An option change at the start of a pattern that had top-level alternatives
1474        could cause overwriting and/or a crash. This command provoked a crash in
1475        some environments:
1477          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1479        This potential security problem was recorded as CVE-2008-2371.
1481    12. For a pattern where the match had to start at the beginning or immediately
1482        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1483        pcre_dfa_exec() could read past the end of the passed subject if there was
1484        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1485        pcretest so that it places the subject at the end of its malloc-ed buffer.
1487    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1488        exec() might read past the end of the data buffer in UTF-8 mode.
1490    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1491        the data contained the byte 0x85 as part of a UTF-8 character within its
1492        first line. This applied both to normal and DFA matching.
1494    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1495        /^[^d]*?$/8 failed to match "abc".
1497    16. Added a missing copyright notice to pcrecpp_internal.h.
1499    17. Make it more clear in the documentation that values returned from
1500        pcre_exec() in ovector are byte offsets, not character counts.
1502    18. Tidied a few places to stop certain compilers from issuing warnings.
1504    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1505        supplied by Stefan Weber. I made a further small update for 7.8 because
1506        there is a change of source arrangements: the pcre_searchfuncs.c module is
1507        replaced by pcre_ucd.c.
1510    Version 7.7 07-May-08
1511    ---------------------
1513    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1514        a string to a long long, pretend we don't even have a long long." This is
1515        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1517    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1518        pre-7.6 versions, which defined a global no_arg variable instead of putting
1519        it in the RE class. (See also #8 below.)
1521    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1522        Lopes.
1524    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1526        (1) The include/exclude patterns were being applied to the whole pathnames
1527            of files, instead of just to the final components.
1529        (2) If there was more than one level of directory, the subdirectories were
1530            skipped unless they satisfied the include/exclude conditions. This is
1531            inconsistent with GNU grep (and could even be seen as contrary to the
1532            pcregrep specification - which I improved to make it absolutely clear).
1533            The action now is always to scan all levels of directory, and just
1534            apply the include/exclude patterns to regular files.
1536    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1537        --exclude_dir in the tests to avoid scanning .svn directories.
1539    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1540        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1541        doesn't support NULs in patterns.
1543    7.  Added some missing "const"s to declarations of static tables in
1544        pcre_compile.c and pcre_dfa_exec.c.
1546    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1547        caused by fix #2  above. (Subsequently also a second patch to fix the
1548        first patch. And a third patch - this was a messy problem.)
1550    9.  Applied Craig's patch to remove the use of push_back().
1552    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1553        matching function regexec().
1555    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1556        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1557        references. PCRE supports relative numbers with this syntax (I don't think
1558        Oniguruma does).
1560    12. Previously, a group with a zero repeat such as (...){0} was completely
1561        omitted from the compiled regex. However, this means that if the group
1562        was called as a subroutine from elsewhere in the pattern, things went wrong
1563        (an internal error was given). Such groups are now left in the compiled
1564        pattern, with a new opcode that causes them to be skipped at execution
1565        time.
1567    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1568        to the way PCRE behaves:
1570        (a) A lone ] character is dis-allowed (Perl treats it as data).
1572        (b) A back reference to an unmatched subpattern matches an empty string
1573            (Perl fails the current match path).
1575        (c) A data ] in a character class must be notated as \] because if the
1576            first data character in a class is ], it defines an empty class. (In
1577            Perl it is not possible to have an empty class.) The empty class []
1578            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1579            The negative empty class [^] matches any one character, independently
1580            of the DOTALL setting.
1582    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1583        non-existent subpattern following a character class starting with ']' and
1584        containing () gave an internal compiling error instead of "reference to
1585        non-existent subpattern". Fortunately, when the pattern did exist, the
1586        compiled code was correct. (When scanning forwards to check for the
1587        existencd of the subpattern, it was treating the data ']' as terminating
1588        the class, so got the count wrong. When actually compiling, the reference
1589        was subsequently set up correctly.)
1591    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1592        it was being rejected as not supported by pcre_dfa_exec(), even though
1593        other assertions are supported. I have made pcre_dfa_exec() support
1594        (*FAIL).
1596    16. The implementation of 13c above involved the invention of a new opcode,
1597        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1598        cannot be changed at match time, I realized I could make a small
1599        improvement to matching performance by compiling OP_ALLANY instead of
1600        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1601        on the OP_ANY path.
1603    17. Compiling pcretest on Windows with readline support failed without the
1604        following two fixes: (1) Make the unistd.h include conditional on
1605        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1607    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1608        ncurses library to be included for pcretest when ReadLine support is
1609        requested, but also to allow for it to be overridden. This patch came from
1610        Daniel Bergström.
1612    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1613        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1614        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1615        spotting this.
1618    Version 7.6 28-Jan-08
1619    ---------------------
1621    1.  A character class containing a very large number of characters with
1622        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1623        overflow.
1625    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1626        HAVE_LONG_LONG is not defined.
1628    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1629        bring them up to date and include new features. This patch includes:
1631        - Fixed PH's badly added libz and libbz2 support.
1632        - Fixed a problem with static linking.
1633        - Added pcredemo. [But later removed - see 7 below.]
1634        - Fixed dftables problem and added an option.
1635        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1636            HAVE_LONG_LONG.
1637        - Added readline support for pcretest.
1638        - Added an listing of the option settings after cmake has run.
1640    4.  A user submitted a patch to Makefile that makes it easy to create
1641        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1642        Makefile.am that cause it to include this special target, without
1643        affecting anything else. Note that the same mingw target plus all
1644        the other distribution libraries and programs are now supported
1645        when configuring with CMake (see 6 below) instead of with
1646        Configure/Make.
1648    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1649        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1650        exported in the Windows port". It has not yet been confirmed that the patch
1651        solves the problem, but it does no harm.
1653    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1654        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1655        with CMake, and also correct the comment about stack recursion.
1657    7.  Remove the automatic building of pcredemo from the ./configure system and
1658        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1659        of a program that users should build themselves after PCRE is installed, so
1660        building it automatically is not really right. What is more, it gave
1661        trouble in some build environments.
1663    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1666    Version 7.5 10-Jan-08
1667    ---------------------
1669    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1670        values in parens when parsing an RE using the C++ wrapper."
1672    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1673        Characters greater than 255 were excluded from the class instead of being
1674        included.
1676    3.  The same bug as (2) above applied to negated POSIX classes such as
1677        [:^space:].
1679    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1680        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1681        I have changed it.
1683    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1684        first named subpattern) and a construct such as (?&a) would reference the
1685        first named subpattern whose name started with "a" (in other words, the
1686        length check was missing). Both these problems are fixed. "Subpattern name
1687        expected" is now given for (?&) (a zero-length name), and this patch also
1688        makes it give the same error for \k'' (previously it complained that that
1689        was a reference to a non-existent subpattern).
1691    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1692        this is right because (?- can be followed by option settings as well as by
1693        digits. I have, however, made the messages clearer.
1695    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1696        than the number used in the conditional) now cause a compile-time error.
1697        This is actually not compatible with Perl, which accepts such patterns, but
1698        treats the conditional as always being FALSE (as PCRE used to), but it
1699        seems to me that giving a diagnostic is better.
1701    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1702        and messages.
1704    9.  Fix two occurrences of "backslash" in comments that should have been
1705        "backspace".
1707    10. Remove two redundant lines of code that can never be obeyed (their function
1708        was moved elsewhere).
1710    11. The program that makes PCRE's Unicode character property table had a bug
1711        which caused it to generate incorrect table entries for sequences of
1712        characters that have the same character type, but are in different scripts.
1713        It amalgamated them into a single range, with the script of the first of
1714        them. In other words, some characters were in the wrong script. There were
1715        thirteen such cases, affecting characters in the following ranges:
1717          U+002b0 - U+002c1
1718          U+0060c - U+0060d
1719          U+0061e - U+00612
1720          U+0064b - U+0065e
1721          U+0074d - U+0076d
1722          U+01800 - U+01805
1723          U+01d00 - U+01d77
1724          U+01d9b - U+01dbf
1725          U+0200b - U+0200f
1726          U+030fc - U+030fe
1727          U+03260 - U+0327f
1728          U+0fb46 - U+0fbb1
1729          U+10450 - U+1049d
1731    12. The -o option (show only the matching part of a line) for pcregrep was not
1732        compatible with GNU grep in that, if there was more than one match in a
1733        line, it showed only the first of them. It now behaves in the same way as
1734        GNU grep.
1736    13. If the -o and -v options were combined for pcregrep, it printed a blank
1737        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1738        does the same. The return code can be used to tell if there were any
1739        non-matching lines.
1741    14. Added --file-offsets and --line-offsets to pcregrep.
1743    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1744        infinitely looping recursion. The bug was that positive lookaheads were not
1745        being skipped when checking for a possible empty match (negative lookaheads
1746        and both kinds of lookbehind were skipped).
1748    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1749        inclusion of <windows.h> to before rather than after the definition of
1750        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1752    17. Specifying a possessive quantifier with a specific limit for a Unicode
1753        character property caused pcre_compile() to compile bad code, which led at
1754        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1755        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1756        caused the error; without that there was no problem.
1758    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1760    19. Added --enable-pcretest-libreadline.
1762    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1763        RE::GlobalReplace(). As a result, the number of replacements returned was
1764        double what it should be. I removed one of the increments, but Craig sent a
1765        later patch that removed the other one (the right fix) and added unit tests
1766        that check the return values (which was not done before).
1768    21. Several CMake things:
1770        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1771            the names libpcre and libpcreposix, not just pcre and pcreposix.
1773        (2) The above change means that pcretest and pcregrep are now correctly
1774            linked with the newly-built libraries, not previously installed ones.
1778    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1779        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1780        UTF-8 newline character). The key issue is that the pattern starts .*;
1781        this means that the match must be either at the beginning, or after a
1782        newline. The bug was in the code for advancing after a failed match and
1783        checking that the new position followed a newline. It was not taking
1784        account of UTF-8 characters correctly.
1786    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1787        character classes. PCRE was not treating the sequence [:...:] as a
1788        character class unless the ... were all letters. Perl, however, seems to
1789        allow any characters between [: and :], though of course it rejects as
1790        unknown any "names" that contain non-letters, because all the known class
1791        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1792        for example, whereas PCRE did not - it did not recognize a POSIX character
1793        class. This seemed a bit dangerous, so the code has been changed to be
1794        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1795        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1796        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1797        Perl does, and where it didn't before.
1799    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1800        Windows environments %n is disabled by default.
1803    Version 7.4 21-Sep-07
1804    ---------------------
1806    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1807        means that a class such as [\s] counted as "explicit reference to CR or
1808        LF". That isn't really right - the whole point of the change was to try to
1809        help when there was an actual mention of one of the two characters. So now
1810        the change happens only if \r or \n (or a literal CR or LF) character is
1811        encountered.
1813    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1814        of both had grown to the point where there were only 3 bits left.
1815        Fortunately, there was spare space in the data structure, and so I have
1816        moved the internal flags into a new 16-bit field to free up more option
1817        bits.
1819    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1820        but did not set the internal JCHANGED flag - either of these is enough to
1821        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1822        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1823        start sets both bits.
1825    4.  Added options (at build time, compile time, exec time) to change \R from
1826        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1828    5.  doc/pcresyntax.html was missing from the distribution.
1830    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1831        compatibility, even though it is no longer used.
1833    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1834        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1835        windows.h file is present (where different names are used). [This was
1836        reversed later after testing - see 16 below.]
1838    8.  Changed all #include <config.h> to #include "config.h". There were also
1839        some further <pcre.h> cases that I changed to "pcre.h".
1841    9.  When pcregrep was used with the --colour option, it missed the line ending
1842        sequence off the lines that it output.
1844    10. It was pointed out to me that arrays of string pointers cause lots of
1845        relocations when a shared library is dynamically loaded. A technique of
1846        using a single long string with a table of offsets can drastically reduce
1847        these. I have refactored PCRE in four places to do this. The result is
1848        dramatic:
1850          Originally:                          290
1851          After changing UCP table:            187
1852          After changing error message table:   43
1853          After changing table of "verbs"       36
1854          After changing table of Posix names   22
1856        Thanks to the folks working on Gregex for glib for this insight.
1858    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1859        unicode-properties was also set.
1861    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1863    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1864        checked only for CRLF.
1866    14. Added casts to pcretest.c to avoid compiler warnings.
1868    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1870    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1871        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1872        entirely. This removes changes made in 7 above.
1874    17. The CMake files have been updated, and there is now more information about
1875        building with CMake in the NON-UNIX-USE document.
1878    Version 7.3 28-Aug-07
1879    ---------------------
1881     1. In the rejigging of the build system that eventually resulted in 7.1, the
1882        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1883        brackets there is not right, since it causes compilers to look for an
1884        installed pcre.h, not the version that is in the source that is being
1885        compiled (which of course may be different). I have changed it back to:
1887          #include "pcre.h"
1889        I have a vague recollection that the change was concerned with compiling in
1890        different directories, but in the new build system, that is taken care of
1891        by the VPATH setting the Makefile.
1893     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1894        when the subject happened to end in the byte 0x85 (e.g. if the last
1895        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1896        characters but of course it shouldn't be taken as a newline when it is part
1897        of another character. The bug was that, for an unlimited repeat of . in
1898        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1899        characters when looking for a newline.
1901     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1903     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1904        in debug output.
1906     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1907        long printing in the pcrecpp unittest when running under MinGW.
1909     6. ESC_K was left out of the EBCDIC table.
1911     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1912        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1913        limit also applies to "virtual nesting" when a pattern is recursive, and in
1914        this case 1000 isn't so big. I have been able to remove this limit at the
1915        expense of backing off one optimization in certain circumstances. Normally,
1916        when pcre_exec() would call its internal match() function recursively and
1917        immediately return the result unconditionally, it uses a "tail recursion"
1918        feature to save stack. However, when a subpattern that can match an empty
1919        string has an unlimited repetition quantifier, it no longer makes this
1920        optimization. That gives it a stack frame in which to save the data for
1921        checking that an empty string has been matched. Previously this was taken
1922        from the 1000-entry workspace that had been reserved. So now there is no
1923        explicit limit, but more stack is used.
1925     8. Applied Daniel's patches to solve problems with the import/export magic
1926        syntax that is required for Windows, and which was going wrong for the
1927        pcreposix and pcrecpp parts of the library. These were overlooked when this
1928        problem was solved for the main library.
1930     9. There were some crude static tests to avoid integer overflow when computing
1931        the size of patterns that contain repeated groups with explicit upper
1932        limits. As the maximum quantifier is 65535, the maximum group length was
1933        set at 30,000 so that the product of these two numbers did not overflow a
1934        32-bit integer. However, it turns out that people want to use groups that
1935        are longer than 30,000 bytes (though not repeat them that many times).
1936        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1937        made it possible to implement the integer overflow checks in a much more
1938        dynamic way, which I have now done. The artificial limitation on group
1939        length has been removed - we now have only the limit on the total length of
1940        the compiled pattern, which depends on the LINK_SIZE setting.
1942    10. Fixed a bug in the documentation for get/copy named substring when
1943        duplicate names are permitted. If none of the named substrings are set, the
1944        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1945        empty string.
1947    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1948        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1949        because the ] is interpreted as the first data character and the
1950        terminating ] is not found. PCRE has been made compatible with Perl in this
1951        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1952        cause memory overwriting.
1954    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1955        string has been matched (to stop an infinite loop). It was not recognizing
1956        a conditional subpattern that could match an empty string if that
1957        subpattern was within another subpattern. For example, it looped when
1958        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1959        condition was not nested. This bug has been fixed.
1961    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1962        past the start of the subject in the presence of bytes with the top bit
1963        set, for example "\x8aBCD".
1965    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1966        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1968    14. Optimized (?!) to (*FAIL).
1970    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1971        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1972        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1973        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1974        does: it's just the validity check that is more restrictive.
1976    16. Inserted checks for integer overflows during escape sequence (backslash)
1977        processing, and also fixed erroneous offset values for syntax errors during
1978        backslash processing.
1980    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1981        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1983    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1984        caused an overrun.
1986    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1987        something other than just ASCII characters) inside a group that had an
1988        unlimited repeat caused a loop at compile time (while checking to see
1989        whether the group could match an empty string).
1991    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1992        [\P{Any}] did so. (Error in the code for printing property names.)
1994    21. An orphan \E inside a character class could cause a crash.
1996    22. A repeated capturing bracket such as (A)? could cause a wild memory
1997        reference during compilation.
1999    23. There are several functions in pcre_compile() that scan along a compiled
2000        expression for various reasons (e.g. to see if it's fixed length for look
2001        behind). There were bugs in these functions when a repeated \p or \P was
2002        present in the pattern. These operators have additional parameters compared
2003        with \d, etc, and these were not being taken into account when moving along
2004        the compiled data. Specifically:
2006        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2007            length.
2009        (b) An item such as \pL+ within a repeated group could cause crashes or
2010            loops.
2012        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2013            "reference to non-existent subpattern" error.
2015        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2017    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2018        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2020    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2022    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2023        character were causing crashes (broken optimization).
2025    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2026        \p or \P) caused a compile-time loop.
2028    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2029        break. For example, the unstudied pattern [\r\n]A does not match the string
2030        "\r\nA" because change 7.0/46 below moves the current point on by two
2031        characters after failing to match at the start. However, the pattern \nA
2032        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2033        the same is true. There doesn't seem any very clean way out of this, but
2034        what I have chosen to do makes the common cases work: PCRE now takes note
2035        of whether there can be an explicit match for \r or \n anywhere in the
2036        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2037        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2038        pattern has explicit CR or LF references.
2040    29. Added (*CR) etc for changing newline setting at start of pattern.
2043    Version 7.2 19-Jun-07
2044    ---------------------
2046     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
2047        which is apparently normally available under Windows.
2049     2. Re-jig the pcregrep tests with different newline settings in an attempt
2050        to make them independent of the local environment's newline setting.
2052     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
2054     4. Some of the "internals" tests were previously cut out when the link size
2055        was not 2, because the output contained actual offsets. The recent new
2056        "Z" feature of pcretest means that these can be cut out, making the tests
2057        usable with all link sizes.
2059     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
2060        stack recursion. This gives a massive performance boost under BSD, but just
2061        a small improvement under Linux. However, it saves one field in the frame
2062        in all cases.
2064     6. Added more features from the forthcoming Perl 5.10:
2066        (a) (?-n) (where n is a string of digits) is a relative subroutine or
2067            recursion call. It refers to the nth most recently opened parentheses.
2069        (b) (?+n) is also a relative subroutine call; it refers to the nth next
2070            to be opened parentheses.
2072        (c) Conditions that refer to capturing parentheses can be specified
2073            relatively, for example, (?(-2)... or (?(+3)...
2075        (d) \K resets the start of the current match so that everything before
2076            is not part of it.
2078        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2080        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2081            reference syntax.
2083        (g) (?| introduces a group in which the numbering of parentheses in each
2084            alternative starts with the same number.
2086        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2088     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2091     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2092        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2093        for detecting groups that can match an empty string.
2095     9. A pattern with a very large number of alternatives (more than several
2096        hundred) was running out of internal workspace during the pre-compile
2097        phase, where pcre_compile() figures out how much memory will be needed. A
2098        bit of new cunning has reduced the workspace needed for groups with
2099        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2100        workspace instead of running out of the 4096 that are available.
2102    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2104    11. Applied patch from Google to remove an optimization that didn't quite work.
2105        The report of the bug said:
2107          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2108          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2109          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2111    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2112        it matched the wrong number of bytes.
2115    Version 7.1 24-Apr-07
2116    ---------------------
2118     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
2119        that is more "standard", making use of automake and other Autotools. There
2120        is some re-arrangement of the files and adjustment of comments consequent
2121        on this.
2123     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
2124        for recursive directory scanning broke on some systems because the files
2125        are not scanned in any specific order and on different systems the order
2126        was different. A call to "sort" has been inserted into RunGrepTest for the
2127        approprate test as a short-term fix. In the longer term there may be an
2128        alternative.
2130     3. I had an email from Eric Raymond about problems translating some of PCRE's
2131        man pages to HTML (despite the fact that I distribute HTML pages, some
2132        people do their own conversions for various reasons). The problems
2133        concerned the use of low-level troff macros .br and .in. I have therefore
2134        removed all such uses from the man pages (some were redundant, some could
2135        be replaced by .nf/.fi pairs). The 132html script that I use to generate
2136        HTML has been updated to handle .nf/.fi and to complain if it encounters
2137        .br or .in.
2139     4. Updated comments in configure.ac that get placed in config.h.in and also
2140        arranged for config.h to be included in the distribution, with the name
2141        config.h.generic, for the benefit of those who have to compile without
2142        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
2144     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
2145        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
2146        makevp.bat for the current PCRE, using the additional files
2147        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
2149     6. A Windows user reported a minor discrepancy with test 2, which turned out
2150        to be caused by a trailing space on an input line that had got lost in his
2151        copy. The trailing space was an accident, so I've just removed it.
2153     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
2154        that is needed.
2156     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
2157        as "const" (a) because they are and (b) because it helps the PHP
2158        maintainers who have recently made a script to detect big data structures
2159        in the php code that should be moved to the .rodata section. I remembered
2160        to update Builducptable as well, so it won't revert if ucptable.h is ever
2161        re-created.
2163     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
2164        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
2165        order to be able to cut out the UTF-8 tables in the latter when UTF-8
2166        support is not required. This saves 1.5-2K of code, which is important in
2167        some applications.
2169        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
2170        so as not to refer to the tables, even though these functions will never be
2171        called when UTF-8 support is disabled. Otherwise there are problems with a
2172        shared library.
2174    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
2176        (a) It was defining its arguments as char * instead of void *.
2178        (b) It was assuming that all moves were upwards in memory; this was true
2179            a long time ago when I wrote it, but is no longer the case.
2181        The emulated memove() is provided for those environments that have neither
2182        memmove() nor bcopy(). I didn't think anyone used it these days, but that
2183        is clearly not the case, as these two bugs were recently reported.
2185    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
2186        and Detrail to create the HTML documentation, the .txt form of the man
2187        pages, and it removes trailing spaces from listed files. It also creates
2188        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
2189        case, it wraps all the #defines with #ifndefs. This script should be run
2190        before "make dist".
2192    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
2193        with Unicode property support.
2195        (a) For a maximizing quantifier, if the two different cases of the
2196            character were of different lengths in their UTF-8 codings (there are
2197            some cases like this - I found 11), and the matching function had to
2198            back up over a mixture of the two cases, it incorrectly assumed they
2199            were both the same length.
2201        (b) When PCRE was configured to use the heap rather than the stack for
2202            recursion during matching, it was not correctly preserving the data for
2203            the other case of a UTF-8 character when checking ahead for a match
2204            while processing a minimizing repeat. If the check also involved
2205            matching a wide character, but failed, corruption could cause an
2206            erroneous result when trying to check for a repeat of the original
2207            character.
2209    13. Some tidying changes to the testing mechanism:
2211        (a) The RunTest script now detects the internal link size and whether there
2212            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
2213            values substituted by "configure". (The RunGrepTest script already did
2214            this for UTF-8.) The configure.ac script no longer substitutes the
2215            relevant variables.
2217        (b) The debugging options /B and /D in pcretest show the compiled bytecode
2218            with length and offset values. This means that the output is different
2219            for different internal link sizes. Test 2 is skipped for link sizes
2220            other than 2 because of this, bypassing the problem. Unfortunately,
2221            there was also a test in test 3 (the locale tests) that used /B and
2222            failed for link sizes other than 2. Rather than cut the whole test out,
2223            I have added a new /Z option to pcretest that replaces the length and
2224            offset values with spaces. This is now used to make test 3 independent
2225            of link size. (Test 2 will be tidied up later.)
2227    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
2228        segmentation fault instead of returning the appropriate error message.
2230    15. In multiline mode when the newline sequence was set to "any", the pattern
2231        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
2232        This doesn't seem right; it now treats the CRLF combination as the line
2233        ending, and so does not match in that case. It's only a pattern such as ^$
2234        that would hit this one: something like ^ABC$ would have failed after \r
2235        and then tried again after \r\n.
2237    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
2238        in an attempt to make files that differ only in their line terminators
2239        compare equal. This works on Linux.
2241    17. Under certain error circumstances pcregrep might try to free random memory
2242        as it exited. This is now fixed, thanks to valgrind.
2244    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
2245        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
2246        was because its rules for how to advance for /g after matching an empty
2247        string at the end of a line did not allow for this case. They now check for
2248        it specially.
2250    20. pcretest is supposed to handle patterns and data of any length, by
2251        extending its buffers when necessary. It was getting this wrong when the
2252        buffer for a data line had to be extended.
2254    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
2255        CRLF as a newline sequence.
2257    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
2258        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
2259        I have nevertheless tidied it up.
2261    23. Added some casts to kill warnings from HP-UX ia64 compiler.
2263    24. Added a man page for pcre-config.
2266    Version 7.0 19-Dec-06
2267    ---------------------
2269     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
2270        moving to gcc 4.1.1.
2272     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
2273        sys/time.h, which is documented as needed for this function. It doesn't
2274        seem to matter on Linux, but it showed up on some releases of OS X.
2276     3. It seems that there are systems where bytes whose values are greater than
2277        127 match isprint() in the "C" locale. The "C" locale should be the
2278        default when a C program starts up. In most systems, only ASCII printing
2279        characters match isprint(). This difference caused the output from pcretest
2280        to vary, making some of the tests fail. I have changed pcretest so that:
2282        (a) When it is outputting text in the compiled version of a pattern, bytes
2283            other than 32-126 are always shown as hex escapes.
2285        (b) When it is outputting text that is a matched part of a subject string,
2286            it does the same, unless a different locale has been set for the match
2287            (using the /L modifier). In this case, it uses isprint() to decide.
2289     4. Fixed a major bug that caused incorrect computation of the amount of memory
2290        required for a compiled pattern when options that changed within the
2291        pattern affected the logic of the preliminary scan that determines the
2292        length. The relevant options are -x, and -i in UTF-8 mode. The result was
2293        that the computed length was too small. The symptoms of this bug were
2294        either the PCRE error "internal error: code overflow" from pcre_compile(),
2295        or a glibc crash with a message such as "pcretest: free(): invalid next
2296        size (fast)". Examples of patterns that provoked this bug (shown in
2297        pcretest format) are:
2299          /(?-x: )/x
2300          /(?x)(?-x: \s*#\s*)/
2301          /((?i)[\x{c0}])/8
2302          /(?i:[\x{c0}])/8
2304        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
2305        is now done differently.
2307     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
2308        wrapper classes; (b) implement a new function in the C++ scanner that is
2309        more efficient than the old way of doing things because it avoids levels of
2310        recursion in the regex matching; (c) add a paragraph to the documentation
2311        for the FullMatch() function.
2313     6. The escape sequence \n was being treated as whatever was defined as
2314        "newline". Not only was this contrary to the documentation, which states
2315        that \n is character 10 (hex 0A), but it also went horribly wrong when
2316        "newline" was defined as CRLF. This has been fixed.
2318     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
2319        was being set to -1 for the "end of line" case (supposedly a value that no
2320        character can have). Though this value is never used (the check for end of
2321        line is "zero bytes in current character"), it caused compiler complaints.
2322        I've changed it to 0xffffffff.
2324     8. In pcre_version.c, the version string was being built by a sequence of
2325        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
2326        string (as it is for production releases) called a macro with an empty
2327        argument. The C standard says the result of this is undefined. The gcc
2328        compiler treats it as an empty string (which was what was wanted) but it is
2329        reported that Visual C gives an error. The source has been hacked around to
2330        avoid this problem.
2332     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
2333        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
2334        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
2335        of them did).
2337    10. Originally, pcretest opened its input and output without "b"; then I was
2338        told that "b" was needed in some environments, so it was added for release
2339        5.0 to both the input and output. (It makes no difference on Unix-like
2340        systems.) Later I was told that it is wrong for the input on Windows. I've
2341        now abstracted the modes into two macros, to make it easier to fiddle with
2342        them, and removed "b" from the input mode under Windows.
2344    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
2346    12. Added -help and --help to pcretest as an official way of being reminded
2347        of the options.
2349    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
2350        and pcrecpp.cc because they annoy compilers at high warning levels.
2352    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
2354    15. Fixed an occurrence of == in configure.ac that should have been = (shell
2355        scripts are not C programs :-) and which was not noticed because it works
2356        on Linux.
2358    16. pcretest is supposed to handle any length of pattern and data line (as one
2359        line or as a continued sequence of lines) by extending its input buffer if
2360        necessary. This feature was broken for very long pattern lines, leading to
2361        a string of junk being passed to pcre_compile() if the pattern was longer
2362        than about 50K.
2364    17. I have done a major re-factoring of the way pcre_compile() computes the
2365        amount of memory needed for a compiled pattern. Previously, there was code
2366        that made a preliminary scan of the pattern in order to do this. That was
2367        OK when PCRE was new, but as the facilities have expanded, it has become
2368        harder and harder to keep it in step with the real compile phase, and there
2369        have been a number of bugs (see for example, 4 above). I have now found a
2370        cunning way of running the real compile function in a "fake" mode that
2371        enables it to compute how much memory it would need, while actually only
2372        ever using a few hundred bytes of working memory and without too many
2373        tests of the mode. This should make future maintenance and development
2374        easier. A side effect of this work is that the limit of 200 on the nesting
2375        depth of parentheses has been removed (though this was never a serious
2376        limitation, I suspect). However, there is a downside: pcre_compile() now
2377        runs more slowly than before (30% or more, depending on the pattern). I
2378        hope this isn't a big issue. There is no effect on runtime performance.
2380    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
2381        newline (only possible for the last line of a file) and it was a
2382        pattern that set a locale (followed by /Lsomething), pcretest crashed.
2384    19. Added additional timing features to pcretest. (1) The -tm option now times
2385        matching only, not compiling. (2) Both -t and -tm can be followed, as a
2386        separate command line item, by a number that specifies the number of
2387        repeats to use when timing. The default is 50000; this gives better
2388        precision, but takes uncomfortably long for very large patterns.
2390    20. Extended pcre_study() to be more clever in cases where a branch of a
2391        subpattern has no definite first character. For example, (a*|b*)[cd] would
2392        previously give no result from pcre_study(). Now it recognizes that the
2393        first character must be a, b, c, or d.
2395    21. There was an incorrect error "recursive call could loop indefinitely" if
2396        a subpattern (or the entire pattern) that was being tested for matching an
2397        empty string contained only one non-empty item after a nested subpattern.
2398        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
2399        incorrectly, because the \d was being skipped in the check.
2401    22. The pcretest program now has a new pattern option /B and a command line
2402        option -b, which is equivalent to adding /B to every pattern. This causes
2403        it to show the compiled bytecode, without the additional information that
2404        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
2405        is the same as /B/I).
2407    23. A new optimization is now able automatically to treat some sequences such
2408        as a*b as a*+b. More specifically, if something simple (such as a character
2409        or a simple class like \d) has an unlimited quantifier, and is followed by
2410        something that cannot possibly match the quantified thing, the quantifier
2411        is automatically "possessified".
2413    24. A recursive reference to a subpattern whose number was greater than 39
2414        went wrong under certain circumstances in UTF-8 mode. This bug could also
2415        have affected the operation of pcre_study().
2417    25. Realized that a little bit of performance could be had by replacing
2418        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
2420    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
2422    27. Possessive quantifiers such as a++ were previously implemented by turning
2423        them into atomic groups such as ($>a+). Now they have their own opcodes,
2424        which improves performance. This includes the automatically created ones
2425        from 23 above.
2427    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
2428        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
2429        the first matched character to be a colon. This applied both to named and
2430        numbered groups.
2432    29. The ucpinternal.h header file was missing its idempotency #ifdef.
2434    30. I was sent a "project" file called libpcre.a.dev which I understand makes
2435        building PCRE on Windows easier, so I have included it in the distribution.
2437    31. There is now a check in pcretest against a ridiculously large number being
2438        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
2439        loop, the loop is abandoned.
2441    32. Forward references to subpatterns in conditions such as (?(2)...) where
2442        subpattern 2 is defined later cause pcre_compile() to search forwards in
2443        the pattern for the relevant set of parentheses. This search went wrong
2444        when there were unescaped parentheses in a character class, parentheses
2445        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2447    33. "Subroutine" calls and backreferences were previously restricted to
2448        referencing subpatterns earlier in the regex. This restriction has now
2449        been removed.
2451    34. Added a number of extra features that are going to be in Perl 5.10. On the
2452        whole, these are just syntactic alternatives for features that PCRE had
2453        previously implemented using the Python syntax or my own invention. The
2454        other formats are all retained for compatibility.
2456        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2457            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2458            also .NET compatible.
2460        (b) A recursion or subroutine call to a named group can now be defined as
2461            (?&name) as well as (?P>name).
2463        (c) A backreference to a named group can now be defined as \k<name> or
2464            \k'name' as well as (?P=name). The new forms, as well as being in Perl
2465            5.10, are also .NET compatible.
2467        (d) A conditional reference to a named group can now use the syntax
2468            (?(<name>) or (?('name') as well as (?(name).
2470        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2471            groups (named and numbered) that are never evaluated inline, but can be
2472            called as "subroutines" from elsewhere. In effect, the DEFINE condition
2473            is always false. There may be only one alternative in such a group.
2475        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2476            as the simple (?(R). The condition is true only if the most recent
2477            recursion is that of the given number or name. It does not search out
2478            through the entire recursion stack.
2480        (g) The escape \gN or \g{N} has been added, where N is a positive or
2481            negative number, specifying an absolute or relative reference.
2483    35. Tidied to get rid of some further signed/unsigned compiler warnings and
2484        some "unreachable code" warnings.
2486    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2487        things, this adds five new scripts.
2489    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2490        There were also incompatibilities regarding the handling of \Q..\E inside
2491        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2492        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2494    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2495        matches an empty string, and forcibly breaks the loop. There were bugs in
2496        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2497        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2498        separate and independent bugs (that affected different cases) have been
2499        fixed.
2501    39. Refactored the code to abolish the use of different opcodes for small
2502        capturing bracket numbers. This is a tidy that I avoided doing when I
2503        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2504        The new approach is not only tidier, it makes it possible to reduce the
2505        memory needed to fix the previous bug (38).
2507    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2508        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2509        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2510        mode.
2512    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2513        report.
2515    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2516        copy construction and assignment in the C++ wrapper.
2518    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2519        couple of bugs that could have given wrong results in the "--newline=crlf"
2520        case.
2522    44. Added a number of casts and did some reorganization of signed/unsigned int
2523        variables following suggestions from Dair Grant. Also renamed the variable
2524        "this" as "item" because it is a C++ keyword.
2526    45. Arranged for dftables to add
2528          #include "pcre_internal.h"
2530        to pcre_chartables.c because without it, gcc 4.x may remove the array
2531        definition from the final binary if PCRE is built into a static library and
2532        dead code stripping is activated.
2534    46. For an unanchored pattern, if a match attempt fails at the start of a
2535        newline sequence, and the newline setting is CRLF or ANY, and the next two
2536        characters are CRLF, advance by two characters instead of one.
2539    Version 6.7 04-Jul-06
2540    ---------------------
2542     1. In order to handle tests when input lines are enormously long, pcretest has
2543        been re-factored so that it automatically extends its buffers when
2544        necessary. The code is crude, but this _is_ just a test program. The
2545        default size has been increased from 32K to 50K.
2547     2. The code in pcre_study() was using the value of the re argument before
2548        testing it for NULL. (Of course, in any sensible call of the function, it
2549        won't be NULL.)
2551     3. The memmove() emulation function in pcre_internal.h, which is used on
2552        systems that lack both memmove() and bcopy() - that is, hardly ever -
2553        was missing a "static" storage class specifier.
2555     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2556        containing an extended class (one that cannot be represented by a bitmap
2557        because it contains high-valued characters or Unicode property items, e.g.
2558        [\pZ]). Almost always one would set UTF-8 mode when processing such a
2559        pattern, but PCRE should not loop if you do not (it no longer does).
2560        [Detail: two cases were found: (a) a repeated subpattern containing an
2561        extended class; (b) a recursive reference to a subpattern that followed a
2562        previous extended class. It wasn't skipping over the extended class
2563        correctly when UTF-8 mode was not set.]
2565     5. A negated single-character class was not being recognized as fixed-length
2566        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2567        compile error "lookbehind assertion is not fixed length".
2569     6. The RunPerlTest auxiliary script was showing an unexpected difference
2570        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2571        write a Perl script that can interpret lines of an input file either as
2572        byte characters or as UTF-8, which is what "perltest" was being required to
2573        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2574        can't do is switch easily at run time between having the "use utf8;" pragma
2575        or not. In the end, I fudged it by using the RunPerlTest script to insert
2576        "use utf8;" explicitly for the UTF-8 tests.
2578     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2579        the end of the subject string, contrary to the documentation and to what
2580        Perl does. This was true of both matching functions. Now it matches only at
2581        the start of the subject and immediately after *internal* newlines.
2583     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2584        a pointer to an int instead of a pointer to an unsigned long int. This
2585        caused problems on 64-bit systems.
2587     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2588        instance of the 'standard' template library not being so standard".
2590    10. There was no check on the number of named subpatterns nor the maximum
2591        length of a subpattern name. The product of these values is used to compute
2592        the size of the memory block for a compiled pattern. By supplying a very
2593        long subpattern name and a large number of named subpatterns, the size
2594        computation could be caused to overflow. This is now prevented by limiting
2595        the length of names to 32 characters, and the number of named subpatterns
2596        to 10,000.
2598    11. Subpatterns that are repeated with specific counts have to be replicated in
2599        the compiled pattern. The size of memory for this was computed from the
2600        length of the subpattern and the repeat count. The latter is limited to
2601        65535, but there was no limit on the former, meaning that integer overflow
2602        could in principle occur. The compiled length of a repeated subpattern is
2603        now limited to 30,000 bytes in order to prevent this.
2605    12. Added the optional facility to have named substrings with the same name.
2607    13. Added the ability to use a named substring as a condition, using the
2608        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2609        are numbers (not recommended). Forward references are permitted.
2611    14. Added forward references in named backreferences (if you see what I mean).
2613    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2614        pattern could run off the end of the subject. For example, the pattern
2615        "(?s)(.{1,5})"8 did this with the subject "ab".
2617    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2618        PCRE_CASELESS was set when matching characters that were quantified with ?
2619        or *.
2621    17. A character class other than a single negated character that had a minimum
2622        but no maximum quantifier - for example [ab]{6,} - was not handled
2623        correctly by pce_dfa_exec(). It would match only one character.
2625    18. A valid (though odd) pattern that looked like a POSIX character
2626        class but used an invalid character after [ (for example [[,abc,]]) caused
2627        pcre_compile() to give the error "Failed: internal error: code overflow" or
2628        in some cases to crash with a glibc free() error. This could even happen if
2629        the pattern terminated after [[ but there just happened to be a sequence of
2630        letters, a binary zero, and a closing ] in the memory that followed.
2632    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2633        over the years. Originally (before any Unicode support), just the bottom 8
2634        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2635        output from "man perlunicode" includes this:
2637          The regular expression compiler produces polymorphic opcodes.  That
2638          is, the pattern adapts to the data and automatically switches to
2639          the Unicode character scheme when presented with Unicode data--or
2640          instead uses a traditional byte scheme when presented with byte
2641          data.
2643        Sadly, a wide octal escape does not cause a switch, and in a string with
2644        no other multibyte characters, these octal escapes are treated as before.
2645        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2646        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2647        Unicode string.
2649        I have not perpetrated such confusion in PCRE. Up till now, it took just
2650        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2651        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2652        translate to the appropriate multibyte character.
2654    29. Applied some refactoring to reduce the number of warnings from Microsoft
2655        and Borland compilers. This has included removing the fudge introduced
2656        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2657        a warning about an unused variable.
2659    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2660        characters since release 4.0, because Perl (from release 5.004) does not.
2661        [Or at least, is documented not to: some releases seem to be in conflict
2662        with the documentation.] However, when a pattern was studied with
2663        pcre_study() and all its branches started with \s, PCRE still included VT
2664        as a possible starting character. Of course, this did no harm; it just
2665        caused an unnecessary match attempt.
2667    22. Removed a now-redundant internal flag bit that recorded the fact that case
2668        dependency changed within the pattern. This was once needed for "required
2669        byte" processing, but is no longer used. This recovers a now-scarce options
2670        bit. Also moved the least significant internal flag bit to the most-
2671        significant bit of the word, which was not previously used (hangover from
2672        the days when it was an int rather than a uint) to free up another bit for
2673        the future.
2675    23. Added support for CRLF line endings as well as CR and LF. As well as the
2676        default being selectable at build time, it can now be changed at runtime
2677        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2678        specify that it is scanning data with non-default line endings.
2680    24. Changed the definition of CXXLINK to make it agree with the definition of
2681        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2683    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2684        recursions. This makes a big different to stack usage for some patterns.
2686    26. If a subpattern containing a named recursion or subroutine reference such
2687        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2688        the space required for the compiled pattern went wrong and gave too small a
2689        value. Depending on the environment, this could lead to "Failed: internal
2690        error: code overflow at offset 49" or "glibc detected double free or
2691        corruption" errors.
2693    27. Applied patches from Google (a) to support the new newline modes and (b) to
2694        advance over multibyte UTF-8 characters in GlobalReplace.
2696    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2697        difference for some implementation of PCRE in some Windows version.
2699    29. Added some extra testing facilities to pcretest:
2701        \q<number>   in a data line sets the "match limit" value
2702        \Q<number>   in a data line sets the "match recursion limt" value
2703        -S <number>  sets the stack size, where <number> is in megabytes
2705        The -S option isn't available for Windows.
2708    Version 6.6 06-Feb-06
2709    ---------------------
2711     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2712        in pcreposix.h. I have copied the definition from pcre.h.
2714     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2715        because pcre.h is no longer a built file.
2717     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2718        not normally included in the compiled code.
2721    Version 6.5 01-Feb-06
2722    ---------------------
2724     1. When using the partial match feature with pcre_dfa_exec(), it was not
2725        anchoring the second and subsequent partial matches at the new starting
2726        point. This could lead to incorrect results. For example, with the pattern
2727        /1234/, partially matching against "123" and then "a4" gave a match.
2729     2. Changes to pcregrep:
2731        (a) All non-match returns from pcre_exec() were being treated as failures
2732            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2733            error message is output. Some extra information is given for the
2734            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2735            probably the only errors that are likely to be caused by users (by
2736            specifying a regex that has nested indefinite repeats, for instance).
2737            If there are more than 20 of these errors, pcregrep is abandoned.
2739        (b) A binary zero was treated as data while matching, but terminated the
2740            output line if it was written out. This has been fixed: binary zeroes
2741            are now no different to any other data bytes.
2743        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2744            used to set a locale for matching. The --locale=xxxx long option has
2745            been added (no short equivalent) to specify a locale explicitly on the
2746            pcregrep command, overriding the environment variables.
2748        (d) When -B was used with -n, some line numbers in the output were one less
2749            than they should have been.
2751        (e) Added the -o (--only-matching) option.
2753        (f) If -A or -C was used with -c (count only), some lines of context were
2754            accidentally printed for the final match.
2756        (g) Added the -H (--with-filename) option.
2758        (h) The combination of options -rh failed to suppress file names for files
2759            that were found from directory arguments.
2761        (i) Added the -D (--devices) and -d (--directories) options.
2763        (j) Added the -F (--fixed-strings) option.
2765        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2767        (l) Added the --colo(u)r option.
2769        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2770            is not present by default.
2772     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2773        items such as (?R) or (?1), when the recursion could match a number of
2774        alternatives. If it matched one of the alternatives, but subsequently,
2775        outside the recursion, there was a failure, the code tried to back up into
2776        the recursion. However, because of the way PCRE is implemented, this is not
2777        possible, and the result was an incorrect result from the match.
2779        In order to prevent this happening, the specification of recursion has
2780        been changed so that all such subpatterns are automatically treated as
2781        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2783     4. I had overlooked the fact that, in some locales, there are characters for
2784        which isalpha() is true but neither isupper() nor islower() are true. In
2785        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2786        and ordfeminine) are like this. This affected the treatment of \w and \W
2787        when they appeared in character classes, but not when they appeared outside
2788        a character class. The bit map for "word" characters is now created
2789        separately from the results of isalnum() instead of just taking it from the
2790        upper, lower, and digit maps. (Plus the underscore character, of course.)
2792     5. The above bug also affected the handling of POSIX character classes such as
2793        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2794        permanent tables. Instead, the bit maps for such a class were previously
2795        created as the appropriate unions of the upper, lower, and digit bitmaps.
2796        Now they are created by subtraction from the [[:word:]] class, which has
2797        its own bitmap.
2799     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2800        It is created by subtracting the vertical space characters (\x09, \x0a,
2801        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2802        subtraction was done in the overall bitmap for a character class, meaning
2803        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2804        be recognized. This bug has been fixed.
2806     7. Patches from the folks at Google:
2808          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2809          real life, but is still worth protecting against".
2811          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2812          regular expressions".
2814          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2815          have it.
2817          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2818          "configure" and the latter not, in order to fix a problem somebody had
2819          with compiling the Arg class on HP-UX.
2821          (e) Improve the error-handling of the C++ wrapper a little bit.
2823          (f) New tests for checking recursion limiting.
2825     8. The pcre_memmove() function, which is used only if the environment does not
2826        have a standard memmove() function (and is therefore rarely compiled),
2827        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2828        returning a result (though PCRE never actually uses the result).
2830     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2831        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2832        returned instead of calling malloc() with an overflowing number that would
2833        most likely cause subsequent chaos.
2835    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2837    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2838        with this option is matched, the nmatch and pmatch options of regexec() are
2839        ignored.
2841    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2842        provided in case anyone wants to the the POSIX interface with UTF-8
2843        strings.
2845    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2846        C++ linking (needed for some HP-UX environments).
2848    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2849        (unused parameter) and in the pcre_printint() function (omitted "default"
2850        switch label when the default is to do nothing).
2852    15. Added some code to make it possible, when PCRE is compiled as a C++
2853        library, to replace subject pointers for pcre_exec() with a smart pointer
2854        class, thus making it possible to process discontinuous strings.
2856    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2857        much the same function. They were added by different people who were trying
2858        to make PCRE easy to compile on non-Unix systems. It has been suggested
2859        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2860        for compiling on Windows systems. I have therefore replaced it with
2861        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2862        defaults to "extern" for C or "extern C" for C++, which works fine on
2863        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2864        SCOPE with something explicit in config.h. In addition:
2866        (a) pcreposix.h still had just "extern" instead of either of these macros;
2867            I have replaced it with PCRE_DATA_SCOPE.
2869        (b) Functions such as _pcre_xclass(), which are internal to the library,
2870            but external in the C sense, all had PCRE_EXPORT in their definitions.
2871            This is apparently wrong for the Windows case, so I have removed it.
2872            (It makes no difference on Unix-like systems.)
2874    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2875        of recursive calls to match(). This is different to MATCH_LIMIT because
2876        that limits the total number of calls to match(), not all of which increase
2877        the depth of recursion. Limiting the recursion depth limits the amount of
2878        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2879        when PCRE is compiled, and changed at run time. A patch from Google adds
2880        this functionality to the C++ interface.
2882    18. Changes to the handling of Unicode character properties:
2884        (a) Updated the table to Unicode 4.1.0.
2886        (b) Recognize characters that are not in the table as "Cn" (undefined).
2888        (c) I revised the way the table is implemented to a much improved format
2889            which includes recognition of ranges. It now supports the ranges that
2890            are defined in UnicodeData.txt, and it also amalgamates other
2891            characters into ranges. This has reduced the number of entries in the
2892            table from around 16,000 to around 3,000, thus reducing its size
2893            considerably. I realized I did not need to use a tree structure after
2894            all - a binary chop search is just as efficient. Having reduced the
2895            number of entries, I extended their size from 6 bytes to 8 bytes to
2896            allow for more data.
2898        (d) Added support for Unicode script names via properties such as \p{Han}.
2900    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2901        matching that character.
2903    20. When matching a repeated Unicode property with a minimum greater than zero,
2904        (for example \pL{2,}), PCRE could look past the end of the subject if it
2905        reached it while seeking the minimum number of characters. This could
2906        happen only if some of the characters were more than one byte long, because
2907        there is a check for at least the minimum number of bytes.
2909    21. Refactored the implementation of \p and \P so as to be more general, to
2910        allow for more different types of property in future. This has changed the
2911        compiled form incompatibly. Anybody with saved compiled patterns that use
2912        \p or \P will have to recompile them.
2914    22. Added "Any" and "L&" to the supported property types.
2916    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2917        but give a compile time error if the value is greater than 0xff.
2919    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2920        accidentally not being installed or uninstalled.
2922    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2923        made were to insert the current release number. This seemed silly, because
2924        it made things harder for people building PCRE on systems that don't run
2925        "configure". I have turned pcre.h into a distributed file, no longer built
2926        by "configure", with the version identification directly included. There is
2927        no longer a pcre.h.in file.
2929        However, this change necessitated a change to the pcre-config script as
2930        well. It is built from pcre-config.in, and one of the substitutions was the
2931        release number. I have updated configure.ac so that ./configure now finds
2932        the release number by grepping pcre.h.
2934    26. Added the ability to run the tests under valgrind.
2937    Version 6.4 05-Sep-05
2938    ---------------------
2940     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2941        "--" to be printed when multiple files were scanned, even when none of the
2942        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2943        consider it to be a bug, and have restored the previous behaviour.
2945     2. A couple of code tidies to get rid of compiler warnings.
2947     3. The pcretest program used to cheat by referring to symbols in the library
2948        whose names begin with _pcre_. These are internal symbols that are not
2949        really supposed to be visible externally, and in some environments it is
2950        possible to suppress them. The cheating is now confined to including
2951        certain files from the library's source, which is a bit cleaner.
2953     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2954        file's purpose clearer.
2956     5. Reorganized pcre_ucp_findchar().
2959    Version 6.3 15-Aug-05
2960    ---------------------
2962     1. The file libpcre.pc.in did not have general read permission in the tarball.
2964     2. There were some problems when building without C++ support:
2966        (a) If C++ support was not built, "make install" and "make test" still
2967            tried to test it.
2969        (b) There were problems when the value of CXX was explicitly set. Some
2970            changes have been made to try to fix these, and ...
2972        (c) --disable-cpp can now be used to explicitly disable C++ support.
2974        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2975            backslash in a target when C++ was disabled. This confuses some
2976            versions of "make", apparently. Using an intermediate variable solves
2977            this. (Same for CPP_LOBJ.)
2979     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2980        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2981        necessary on certain architectures.
2983     4. Added a setting of -export-symbols-regex to the link command to remove
2984        those symbols that are exported in the C sense, but actually are local
2985        within the library, and not documented. Their names all begin with
2986        "_pcre_". This is not a perfect job, because (a) we have to except some
2987        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2988        available (and never for static libraries). I have made a note to try to
2989        find a way round (a) in the future.
2992    Version 6.2 01-Aug-05
2993    ---------------------
2995     1. There was no test for integer overflow of quantifier values. A construction
2996        such as {1111111111111111} would give undefined results. What is worse, if
2997        a minimum quantifier for a parenthesized subpattern overflowed and became
2998        negative, the calculation of the memory size went wrong. This could have
2999        led to memory overwriting.
3001     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
3003     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
3004        operating environments where this matters.
3006     4. Applied Giuseppe Maxia's patch to add additional features for controlling
3007        PCRE options from within the C++ wrapper.
3009     5. Named capturing subpatterns were not being correctly counted when a pattern
3010        was compiled. This caused two problems: (a) If there were more than 100
3011        such subpatterns, the calculation of the memory needed for the whole
3012        compiled pattern went wrong, leading to an overflow error. (b) Numerical
3013        back references of the form \12, where the number was greater than 9, were
3014        not recognized as back references, even though there were sufficient
3015        previous subpatterns.
3017     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
3018        versions of gcc, e.g. 2.95.4.
3021    Version 6.1 21-Jun-05
3022    ---------------------
3024     1. There was one reference to the variable "posix" in pcretest.c that was not
3025        surrounded by "#if !defined NOPOSIX".
3027     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
3028        the cross-check on the old pcre_info() function, for the benefit of the
3029        cut-down version of PCRE that is currently imported into Exim.
3031     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
3032        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
3033        patterns that start (?i) but I don't think that matters. The use of (?i) is
3034        just an example; this all applies to the other options as well.
3036     4. Since libtool seems to echo the compile commands it is issuing, the output
3037        from "make" can be reduced a bit by putting "@" in front of each libtool
3038        compile command.
3040     5. Patch from the folks at Google for configure.in to be a bit more thorough
3041        in checking for a suitable C++ installation before trying to compile the
3042        C++ stuff. This should fix a reported problem when a compiler was present,
3043        but no suitable headers.
3045     6. The man pages all had just "PCRE" as their title. I have changed them to
3046        be the relevant file name. I have also arranged that these names are
3047        retained in the file doc/pcre.txt, which is a concatenation in text format
3048        of all the man pages except the little individual ones for each function.
3050     7. The NON-UNIX-USE file had not been updated for the different set of source
3051        files that come with release 6. I also added a few comments about the C++
3052        wrapper.
3055    Version 6.0 07-Jun-05
3056    ---------------------
3058     1. Some minor internal re-organization to help with my DFA experiments.
3060     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
3061        didn't matter for the library itself when fully configured, but did matter
3062        when compiling without UCP support, or within Exim, where the ucp files are
3063        not imported.
3065     3. Refactoring of the library code to split up the various functions into
3066        different source modules. The addition of the new DFA matching code (see
3067        below) to a single monolithic source would have made it really too
3068        unwieldy, quite apart from causing all the code to be include in a
3069        statically linked application, when only some functions are used. This is
3070        relevant even without the DFA addition now that patterns can be compiled in
3071        one application and matched in another.
3073        The downside of splitting up is that there have to be some external
3074        functions and data tables that are used internally in different modules of
3075        the library but which are not part of the API. These have all had their
3076        names changed to start with "_pcre_" so that they are unlikely to clash
3077        with other external names.
3079     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
3080        a different (DFA) algorithm. Although it is slower than the original
3081        function, it does have some advantages for certain types of matching
3082        problem.
3084     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
3085        including restarting after a partial match.
3087     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
3088        defined when compiling for Windows was sent to me. I have put it into the
3089        code, though I have no means of testing or verifying it.
3091     7. Added the pcre_refcount() auxiliary function.
3093     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
3094        match before or at the first newline in the subject string. In pcretest,
3095        the /f option on a pattern can be used to set this.
3097     9. A repeated \w when used in UTF-8 mode with characters greater than 256
3098        would behave wrongly. This has been present in PCRE since release 4.0.
3100    10. A number of changes to the pcregrep command:
3102        (a) Refactored how -x works; insert ^(...)$ instead of setting
3103            PCRE_ANCHORED and checking the length, in preparation for adding
3104            something similar for -w.
3106        (b) Added the -w (match as a word) option.
3108        (c) Refactored the way lines are read and buffered so as to have more
3109            than one at a time available.
3111        (d) Implemented a pcregrep test script.
3113        (e) Added the -M (multiline match) option. This allows patterns to match
3114            over several lines of the subject. The buffering ensures that at least
3115            8K, or the rest of the document (whichever is the shorter) is available
3116            for matching (and similarly the previous 8K for lookbehind assertions).
3118        (f) Changed the --help output so that it now says
3120              -w, --word-regex(p)
3122            instead of two lines, one with "regex" and the other with "regexp"
3123            because that confused at least one person since the short forms are the
3124            same. (This required a bit of code, as the output is generated
3125            automatically from a table. It wasn't just a text change.)
3127        (g) -- can be used to terminate pcregrep options if the next thing isn't an
3128            option but starts with a hyphen. Could be a pattern or a path name
3129            starting with a hyphen, for instance.
3131        (h) "-" can be given as a file name to represent stdin.
3133        (i) When file names are being printed, "(standard input)" is used for
3134            the standard input, for compatibility with GNU grep. Previously
3135            "<stdin>" was used.
3137        (j) The option --label=xxx can be used to supply a name to be used for
3138            stdin when file names are being printed. There is no short form.
3140        (k) Re-factored the options decoding logic because we are going to add
3141            two more options that take data. Such options can now be given in four
3142            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
3144        (l) Added the -A, -B, and -C options for requesting that lines of context
3145            around matches be printed.
3147        (m) Added the -L option to print the names of files that do not contain
3148            any matching lines, that is, the complement of -l.
3150        (n) The return code is 2 if any file cannot be opened, but pcregrep does
3151            continue to scan other files.
3153        (o) The -s option was incorrectly implemented. For compatibility with other
3154            greps, it now suppresses the error message for a non-existent or non-
3155            accessible file (but not the return code). There is a new option called
3156            -q that suppresses the output of matching lines, which was what -s was
3157            previously doing.
3159        (p) Added --include and --exclude options to specify files for inclusion
3160            and exclusion when recursing.
3162    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
3163        Hopefully, it now does.
3165    12. Missing cast in pcre_study().
3167    13. Added an "uninstall" target to the makefile.
3169    14. Replaced "extern" in the function prototypes in Makefile.in with
3170        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
3171        world, but is set differently for Windows.
3173    15. Added a second compiling function called pcre_compile2(). The only
3174        difference is that it has an extra argument, which is a pointer to an
3175        integer error code. When there is a compile-time failure, this is set
3176        non-zero, in addition to the error test pointer being set to point to an
3177        error message. The new argument may be NULL if no error number is required
3178        (but then you may as well call pcre_compile(), which is now just a
3179        wrapper). This facility is provided because some applications need a
3180        numeric error indication, but it has also enabled me to tidy up the way
3181        compile-time errors are handled in the POSIX wrapper.
3183    16. Added VPATH=.libs to the makefile; this should help when building with one
3184        prefix path and installing with another. (Or so I'm told by someone who
3185        knows more about this stuff than I do.)
3187    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
3188        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
3189        match everything, including newlines. This is not POSIX-compatible, but
3190        somebody wanted the feature. From pcretest it can be activated by using
3191        both the P and the s flags.
3193    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
3195    19. libpcre.pc was being incorrectly installed as executable.
3197    20. A couple of places in pcretest check for end-of-line by looking for '\n';
3198        it now also looks for '\r' so that it will work unmodified on Windows.
3200    21. Added Google's contributed C++ wrapper to the distribution.
3202    22. Added some untidy missing memory free() calls in pcretest, to keep
3203        Electric Fence happy when testing.
3207    Version 5.0 13-Sep-04
3208    ---------------------
3210     1. Internal change: literal characters are no longer packed up into items
3211        containing multiple characters in a single byte-string. Each character
3212        is now matched using a separate opcode. However, there may be more than one
3213        byte in the character in UTF-8 mode.
3215     2. The pcre_callout_block structure has two new fields: pattern_position and
3216        next_item_length. These contain the offset in the pattern to the next match
3217        item, and its length, respectively.
3219     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
3220        insertion of callouts before each pattern item. Added the /C option to
3221        pcretest to make use of this.
3223     4. On the advice of a Windows user, the lines
3225          #if defined(_WIN32) || defined(WIN32)
3226          _setmode( _fileno( stdout ), 0x8000 );
3227          #endif  /* defined(_WIN32) || defined(WIN32) */
3229        have been added to the source of pcretest. This apparently does useful
3230        magic in relation to line terminators.
3232     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
3233        for the benefit of those environments where the "b" makes a difference.
3235     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
3236        to know about it. I have put a hack into configure.in that adds in code
3237        to set GCC=yes if CC=icc. This seems to end up at a point in the
3238        generated configure script that is early enough to affect the setting of
3239        compiler options, which is what is needed, but I have no means of testing
3240        whether it really works. (The user who reported this had patched the
3241        generated configure script, which of course I cannot do.)
3243        LATER: After change 22 below (new libtool files), the configure script
3244        seems to know about icc (and also ecc). Therefore, I have commented out
3245        this hack in configure.in.
3247     7. Added support for pkg-config (2 patches were sent in).
3249     8. Negated POSIX character classes that used a combination of internal tables
3250        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
3251        [[:^ascii]]. Typically, they would match almost any characters. The other
3252        POSIX classes were not broken in this way.
3254     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
3255        to find the match, as PCRE was deluded into thinking that the match had to
3256        start at the start point or following a newline. The same bug applied to
3257        patterns with negative forward assertions or any backward assertions
3258        preceding ".*" at the start, unless the pattern required a fixed first
3259        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
3261    10. In UTF-8 mode, when moving forwards in the subject after a failed match
3262        starting at the last subject character, bytes beyond the end of the subject
3263        string were read.
3265    11. Renamed the variable "class" as "classbits" to make life easier for C++
3266        users. (Previously there was a macro definition, but it apparently wasn't
3267        enough.)
3269    12. Added the new field "tables" to the extra data so that tables can be passed
3270        in at exec time, or the internal tables can be re-selected. This allows
3271        a compiled regex to be saved and re-used at a later time by a different
3272        program that might have everything at different addresses.
3274    13. Modified the pcre-config script so that, when run on Solaris, it shows a
3275        -R library as well as a -L library.
3277    14. The debugging options of pcretest (-d on the command line or D on a
3278        pattern) showed incorrect output for anything following an extended class
3279        that contained multibyte characters and which was followed by a quantifier.
3281    15. Added optional support for general category Unicode character properties
3282        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
3283        support. It adds about 90K to the size of the library. The meanings of the
3284        inbuilt class escapes such as \d and \s have NOT been changed.
3286    16. Updated pcredemo.c to include calls to free() to release the memory for the
3287        compiled pattern.
3289    17. The generated file chartables.c was being created in the source directory
3290        instead of in the building directory. This caused the build to fail if the
3291        source directory was different from the building directory, and was
3292        read-only.
3294    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
3295        file. No doubt somebody will tell me if they don't make sense... Also added
3296        Dan Mooney's comments about building on OpenVMS.
3298    19. Added support for partial matching via the PCRE_PARTIAL option for
3299        pcre_exec() and the \P data escape in pcretest.
3301    20. Extended pcretest with 3 new pattern features:
3303        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
3304              write the compiled pattern to the file whose name is "rest-of-line".
3305              This is a straight binary dump of the data, with the saved pointer to
3306              the character tables forced to be NULL. The study data, if any, is
3307              written too. After writing, pcretest reads a new pattern.
3309        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
3310              compiled pattern from the given file. There must not be any
3311              occurrences of "<" in the file name (pretty unlikely); if there are,
3312              pcretest will instead treat the initial "<" as a pattern delimiter.
3313              After reading in the pattern, pcretest goes on to read data lines as
3314              usual.
3316        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
3317              and 16-bit fields in a compiled pattern, to simulate a pattern that
3318              was compiled on a host of opposite endianness.
3320    21. The pcre-exec() function can now cope with patterns that were compiled on
3321        hosts of opposite endianness, with this restriction:
3323          As for any compiled expression that is saved and used later, the tables
3324          pointer field cannot be preserved; the extra_data field in the arguments
3325          to pcre_exec() should be used to pass in a tables address if a value
3326          other than the default internal tables were used at compile time.
3328    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
3329        now diagnosed as an error. Previously, most of the time, a negative number
3330        would have been treated as zero, but if in addition "ovector" was passed as
3331        NULL, a crash could occur.
3333    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
3334        new versions from the libtool 1.5 distribution (the last one is a copy of
3335        a file called libtool.m4). This seems to have fixed the need to patch
3336        "configure" to support Darwin 1.3 (which I used to do). However, I still
3337        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
3338        workstation).
3340    24. Changed the PCRE licence to be the more standard "BSD" licence.
3343    Version 4.5 01-Dec-03
3344    ---------------------
3346     1. There has been some re-arrangement of the code for the match() function so
3347        that it can be compiled in a version that does not call itself recursively.
3348        Instead, it keeps those local variables that need separate instances for
3349        each "recursion" in a frame on the heap, and gets/frees frames whenever it
3350        needs to "recurse". Keeping track of where control must go is done by means
3351        of setjmp/longjmp. The whole thing is implemented by a set of macros that
3352        hide most of the details from the main code, and operates only if
3353        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
3354        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
3355        operating.
3357        To make it easier for callers to provide specially tailored get/free
3358        functions for this usage, two new functions, pcre_stack_malloc, and
3359        pcre_stack_free, are used. They are always called in strict stacking order,
3360        and the size of block requested is always the same.
3362        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
3363        PCRE has been compiled to use the stack or the heap for recursion. The
3364        -C option of pcretest uses this to show which version is compiled.
3366        A new data escape \S, is added to pcretest; it causes the amounts of store
3367        obtained and freed by both kinds of malloc/free at match time to be added
3368        to the output.
3370     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
3371        what's available on my current Linux desktop machine.
3373     3. When matching a UTF-8 string, the test for a valid string at the start has
3374        been extended. If start_offset is not zero, PCRE now checks that it points
3375        to a byte that is the start of a UTF-8 character. If not, it returns
3376        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
3377        this is necessary because there may be backward assertions in the pattern.
3378        When matching the same subject several times, it may save resources to use
3379        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
3381     4. The code for checking the validity of UTF-8 strings has been tightened so
3382        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
3383        containing "overlong sequences".
3385     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
3386        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
3387        should have been "&", but it just so happened that all the cases this let
3388        through by mistake were picked up later in the function.
3390     6. I had used a variable called "isblank" - this is a C99 function, causing
3391        some compilers to warn. To avoid this, I renamed it (as "blankclass").
3393     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
3394        prompting; (b) run "./pcretest /dev/null" at the start of the test script
3395        so the version is shown; (c) stop "make test" echoing "./RunTest".
3397     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
3399     9. The prototype for memmove() for systems that don't have it was using
3400        size_t, but the inclusion of the header that defines size_t was later. I've
3401        moved the #includes for the C headers earlier to avoid this.
3403    10. Added some adjustments to the code to make it easier to compiler on certain
3404        special systems:
3406          (a) Some "const" qualifiers were missing.
3407          (b) Added the macro EXPORT before all exported functions; by default this
3408              is defined to be empty.
3409          (c) Changed the dftables auxiliary program (that builds chartables.c) so
3410              that it reads its output file name as an argument instead of writing
3411              to the standard output and assuming this can be redirected.
3413    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
3414        class containing characters with values greater than 255, PCRE compilation
3415        went into a loop.
3417    12. A recursive reference to a subpattern that was within another subpattern
3418        that had a minimum quantifier of zero caused PCRE to crash. For example,
3419        (x(y(?2))z)? provoked this bug with a subject that got as far as the
3420        recursion. If the recursively-called subpattern itself had a zero repeat,
3421        that was OK.
3423    13. In pcretest, the buffer for reading a data line was set at 30K, but the
3424        buffer into which it was copied (for escape processing) was still set at
3425        1024, so long lines caused crashes.
3427    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
3428        "internal error: code overflow...". This applied to any character class
3429        that was followed by a possessive quantifier.
3431    15. Modified the Makefile to add libpcre.la as a prerequisite for
3432        libpcreposix.la because I was told this is needed for a parallel build to
3433        work.
3435    16. If a pattern that contained .* following optional items at the start was
3436        studied, the wrong optimizing data was generated, leading to matching
3437        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
3438        matching string must start with a or b or c. The correct conclusion for
3439        this pattern is that a match can start with any character.
3442    Version 4.4 13-Aug-03
3443    ---------------------
3445     1. In UTF-8 mode, a character class containing characters with values between
3446        127 and 255 was not handled correctly if the compiled pattern was studied.
3447        In fixing this, I have also improved the studying algorithm for such
3448        classes (slightly).
3450     2. Three internal functions had redundant arguments passed to them. Removal
3451        might give a very teeny performance improvement.
3453     3. Documentation bug: the value of the capture_top field in a callout is *one
3454        more than* the number of the hightest numbered captured substring.
3456     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
3457        in incorrectly linking with a previously installed version. They now link
3458        explicitly with libpcre.la.
3460     5. configure.in no longer needs to recognize Cygwin specially.
3462     6. A problem in pcre.in for Windows platforms is fixed.
3464     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
3465        pcretest, it used to include the size of the study block as part of its
3466        output. Unfortunately, the structure contains a field that has a different
3467        size on different hardware architectures. This meant that the tests that
3468        showed this size failed. As the block is currently always of a fixed size,
3469        this information isn't actually particularly useful in pcretest output, so
3470        I have just removed it.
3472     8. Three pre-processor statements accidentally did not start in column 1.
3473        Sadly, there are *still* compilers around that complain, even though
3474        standard C has not required this for well over a decade. Sigh.
3476     9. In pcretest, the code for checking callouts passed small integers in the
3477        callout_data field, which is a void * field. However, some picky compilers
3478        complained about the casts involved for this on 64-bit systems. Now
3479        pcretest passes the address of the small integer instead, which should get
3480        rid of the warnings.
3482    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
3483        both compile and run time, and gives an error if an invalid UTF-8 sequence
3484        is found. There is a option for disabling this check in cases where the
3485        string is known to be correct and/or the maximum performance is wanted.
3487    11. In response to a bug report, I changed one line in Makefile.in from
3489            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
3490        to
3491            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
3493        to look similar to other lines, but I have no way of telling whether this
3494        is the right thing to do, as I do not use Windows. No doubt I'll get told
3495        if it's wrong...
3498    Version 4.3 21-May-03
3499    ---------------------
3501    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
3502       Makefile.
3504    2. Some refactoring to improve the quality of the code:
3506       (i)   The utf8_table... variables are now declared "const".
3508       (ii)  The code for \cx, which used the "case flipping" table to upper case
3509             lower case letters, now just substracts 32. This is ASCII-specific,
3510             but the whole concept of \cx is ASCII-specific, so it seems
3511             reasonable.
3513       (iii) PCRE was using its character types table to recognize decimal and
3514             hexadecimal digits in the pattern. This is silly, because it handles
3515             only 0-9, a-f, and A-F, but the character types table is locale-
3516             specific, which means strange things might happen. A private
3517             table is now used for this - though it costs 256 bytes, a table is
3518             much faster than multiple explicit tests. Of course, the standard
3519             character types table is still used for matching digits in subject
3520             strings against \d.
3522       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
3523             ending in _t are). So I've renamed it as ESC_tee.
3525    3. The first argument for regexec() in the POSIX wrapper should have been
3526       defined as "const".
3528    4. Changed pcretest to use malloc() for its buffers so that they can be
3529       Electric Fenced for debugging.
3531    5. There were several places in the code where, in UTF-8 mode, PCRE would try
3532       to read one or more bytes before the start of the subject string. Often this
3533       had no effect on PCRE's behaviour, but in some circumstances it could
3534       provoke a segmentation fault.
3536    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
3537       to try to read one or more bytes before the start of the subject string.
3539    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3540       UTF-8 support could misbehave in various ways if the subject string
3541       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3542       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3543       back over UTF-8 characters.)
3546    Version 4.2 14-Apr-03
3547    ---------------------
3549    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
3551    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
3552         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
3553         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
3554         [WIN_PREFIX]: new variable, "cyg" for Cygwin
3555         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
3556           and BUILD_EXEEXT
3557         Note: automatic setting of the BUILD variables is not yet working
3558         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
3559           compile-time but not at link-time
3560         [LINK]: use for linking executables only
3561         make different versions for Windows and non-Windows
3562         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
3563           libraries
3564         [LINK_FOR_BUILD]: new variable
3565         [OBJEXT]: use throughout
3566         [EXEEXT]: use throughout
3567         <winshared>: new target
3568         <wininstall>: new target
3569         <dftables.o>: use native compiler
3570         <dftables>: use native linker
3571         <install>: handle Windows platform correctly
3572         <clean>: ditto
3573         <check>: ditto
3574         copy DLL to top builddir before testing
3576       As part of these changes, -no-undefined was removed again. This was reported
3577       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3578       in any case.
3580    3. Some tidies to get rid of compiler warnings:
3582       . In the match_data structure, match_limit was an unsigned long int, whereas
3583         match_call_count was an int. I've made them both unsigned long ints.
3585       . In pcretest the fact that a const uschar * doesn't automatically cast to
3586         a void * provoked a warning.
3588       . Turning on some more compiler warnings threw up some "shadow" variables
3589         and a few more missing casts.
3591    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3592       option, a class that contained a single character with a value between 128
3593       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3595    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3596       option, a class that contained several characters, but with at least one
3597       whose value was between 128 and 255 caused PCRE to crash.
3600    Version 4.1 12-Mar-03