ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 11 by nigel, Sat Feb 24 21:38:17 2007 UTC revision 969 by zherczeg, Sun May 27 04:33:38 2012 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.31
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.
122    Version 8.30 04-February-2012
123    -----------------------------
125    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
126        name is defined in ctype.h.
128    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
129        only in quite long subpatterns.
131    3.  Removed the function pcre_info(), which has been obsolete and deprecated
132        since it was replaced by pcre_fullinfo() in February 2000.
134    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
135        match a (*MARK), and the match failed at the start of the subject, a
136        reference to memory before the start of the subject could occur. This bug
137        was introduced by fix 17 of release 8.21.
139    5.  A reference to an unset group with zero minimum repetition was giving
140        totally wrong answers (in non-JavaScript-compatibility mode). For example,
141        /(another)?(\1?)test/ matched against "hello world test". This bug was
142        introduced in release 8.13.
144    6.  Add support for 16-bit character strings (a large amount of work involving
145        many changes and refactorings).
147    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
148        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
149        from a file.
151    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
152        rounding is not applied in this particular case).
154    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
155        if they appear, or are escaped, in patterns.
157    10. Get rid of a number of -Wunused-but-set-variable warnings.
159    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
160        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
161        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
162        also returns the mark "x". This bug applied to capturing parentheses,
163        non-capturing parentheses, and atomic parentheses. It also applied to some
164        assertions.
166    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
167        information out of configure.ac instead of relying on pcre.h.generic, which
168        is not stored in the repository.
170    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
171        -lreadline.
173    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
175    15. Applied Graycode's patch to put the top-level frame on the stack rather
176        than the heap when not using the stack for recursion. This gives a
177        performance improvement in many cases when recursion is not deep.
179    16. Experimental code added to "pcretest -C" to output the stack frame size.
182    Version 8.21 12-Dec-2011
183    ------------------------
185    1.  Updating the JIT compiler.
187    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
188        are added as well.
190    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
191        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
192        calling _pcre_jit_exec. Some extra comments are added.
194    4.  (*MARK) settings inside atomic groups that do not contain any capturing
195        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
196        was introduced by change 18 for 8.20.
198    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
199        ECMA-262 standard.
201    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
202        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
203        This bug was probably introduced by change 9 of 8.13.
205    7.  While fixing 6 above, I noticed that a number of other items were being
206        incorrectly rejected as "not fixed length". This arose partly because newer
207        opcodes had not been added to the fixed-length checking code. I have (a)
208        corrected the bug and added tests for these items, and (b) arranged for an
209        error to occur if an unknown opcode is encountered while checking for fixed
210        length instead of just assuming "not fixed length". The items that were
211        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
212        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
213        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
215    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
216        being incorrectly compiled and would have given unpredicatble results.
218    9.  A possessively repeated subpattern with minimum repeat count greater than
219        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
220        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
221        the first (A) could occur when it should not.
223    10. Add a cast and remove a redundant test from the code.
225    11. JIT should use pcre_malloc/pcre_free for allocation.
227    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
228        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
229        is anything other than /usr, -L is still shown).
231    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
233    14. Perl does not support \N without a following name in a [] class; PCRE now
234        also gives an error.
236    15. If a forward reference was repeated with an upper limit of around 2000,
237        it caused the error "internal error: overran compiling workspace". The
238        maximum number of forward references (including repeats) was limited by the
239        internal workspace, and dependent on the LINK_SIZE. The code has been
240        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
241        the default depends on LINK_SIZE. There is a new upper limit (for safety)
242        of around 200,000 forward references. While doing this, I also speeded up
243        the filling in of repeated forward references.
245    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
246        incorrectly expecting the subject to contain another "a" after the start.
248    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
249        in the match, the SKIP should be ignored. This was not happening; instead
250        the SKIP was being treated as NOMATCH. For patterns such as
251        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
252        tested.
254    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
255        now much more compatible with Perl, in particular in cases where the result
256        is a non-match for a non-anchored pattern. For example, if
257        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
258        "m", where previously it did not return a name. A side effect of this
259        change is that for partial matches, the last encountered mark name is
260        returned, as for non matches. A number of tests that were previously not
261        Perl-compatible have been moved into the Perl-compatible test files. The
262        refactoring has had the pleasing side effect of removing one argument from
263        the match() function, thus reducing its stack requirements.
265    19. If the /S+ option was used in pcretest to study a pattern using JIT,
266        subsequent uses of /S (without +) incorrectly behaved like /S+.
268    21. Retrieve executable code size support for the JIT compiler and fixing
269        some warnings.
271    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
272        not work when the shorter character appeared right at the end of the
273        subject string.
275    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
276        systems.
278    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
279        output it when the /M option is used in pcretest.
281    25. The CheckMan script was not being included in the distribution. Also, added
282        an explicit "perl" to run Perl scripts from the PrepareRelease script
283        because this is reportedly needed in Windows.
285    26. If study data was being save in a file and studying had not found a set of
286        "starts with" bytes for the pattern, the data written to the file (though
287        never used) was taken from uninitialized memory and so caused valgrind to
288        complain.
290    27. Updated RunTest.bat as provided by Sheri Pierce.
292    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
294    29. Computation of memory usage for the table of capturing group names was
295        giving an unnecessarily large value.
298    Version 8.20 21-Oct-2011
299    ------------------------
301    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
302        a POSIX class. After further experiments with Perl, which convinced me that
303        Perl has bugs and confusions, a closing square bracket is no longer allowed
304        in a POSIX name. This bug also affected patterns with classes that started
305        with full stops.
307    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
308        captured substring, but while checking the failing first alternative,
309        substring 1 is temporarily captured. If the output vector supplied to
310        pcre_exec() was not big enough for this capture, the yield of the function
311        was still zero ("insufficient space for captured substrings"). This cannot
312        be totally fixed without adding another stack variable, which seems a lot
313        of expense for a edge case. However, I have improved the situation in cases
314        such as /(a)(b)x|abc/ matched against "abc", where the return code
315        indicates that fewer than the maximum number of slots in the ovector have
316        been set.
318    3.  Related to (2) above: when there are more back references in a pattern than
319        slots in the output vector, pcre_exec() uses temporary memory during
320        matching, and copies in the captures as far as possible afterwards. It was
321        using the entire output vector, but this conflicts with the specification
322        that only 2/3 is used for passing back captured substrings. Now it uses
323        only the first 2/3, for compatibility. This is, of course, another edge
324        case.
326    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
327        main code base, and can be used by building with --enable-jit. When this is
328        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
329        runtime --no-jit option is given.
331    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
332        ovector, the return from the function was zero, implying that there were
333        other matches that did not fit. The correct "exactly full" value is now
334        returned.
336    6.  If a subpattern that was called recursively or as a subroutine contained
337        (*PRUNE) or any other control that caused it to give a non-standard return,
338        invalid errors such as "Error -26 (nested recursion at the same subject
339        position)" or even infinite loops could occur.
341    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
342        computing the minimum length on reaching *ACCEPT, and so ended up with the
343        wrong value of 1 rather than 0. Further investigation indicates that
344        computing a minimum subject length in the presence of *ACCEPT is difficult
345        (think back references, subroutine calls), and so I have changed the code
346        so that no minimum is registered for a pattern that contains *ACCEPT.
348    8.  If (*THEN) was present in the first (true) branch of a conditional group,
349        it was not handled as intended. [But see 16 below.]
351    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
352        Sheri Pierce.
354    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
355        the first byte in a match must be "a".
357    11. Change 17 for 8.13 increased the recursion depth for patterns like
358        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
359        pattern contains any instances of (*THEN). If it does not, the old
360        optimizations are restored. It would be nice to do this on a per-group
361        basis, but at the moment that is not feasible.
363    12. In some environments, the output of pcretest -C is CRLF terminated. This
364        broke RunTest's code that checks for the link size. A single white space
365        character after the value is now allowed for.
367    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
368        For "fr", it uses the Windows-specific input and output files.
370    14. If (*THEN) appeared in a group that was called recursively or as a
371        subroutine, it did not work as intended. [But see next item.]
373    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
374        pattern fragments (but not containing any | characters). If A and B are
375        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
376        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
377        D. In other words, Perl considers parentheses that do not contain any |
378        characters to be part of a surrounding alternative, whereas PCRE was
379        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
380        differently. PCRE now behaves in the same way as Perl, except in the case
381        of subroutine/recursion calls such as (?1) which have in any case always
382        been different (but PCRE had them first :-).
384    16. Related to 15 above: Perl does not treat the | in a conditional group as
385        creating alternatives. Such a group is treated in the same way as an
386        ordinary group without any | characters when processing (*THEN). PCRE has
387        been changed to match Perl's behaviour.
389    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
390        RunGrepTest script failed.
392    18. Change 22 for version 13 caused atomic groups to use more stack. This is
393        inevitable for groups that contain captures, but it can lead to a lot of
394        stack use in large patterns. The old behaviour has been restored for atomic
395        groups that do not contain any capturing parentheses.
397    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
398        suppress the check for a minimum subject length at run time. (If it was
399        given to pcre_exec() or pcre_dfa_exec() it did work.)
401    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
402        fail to work when decoding hex characters in data strings in EBCDIC
403        environments.
405    21. It appears that in at least one Mac OS environment, the isxdigit() function
406        is implemented as a macro that evaluates to its argument more than once,
407        contravening the C 90 Standard (I haven't checked a later standard). There
408        was an instance in pcretest which caused it to go wrong when processing
409        \x{...} escapes in subject strings. The has been rewritten to avoid using
410        things like p++ in the argument of isxdigit().
413    Version 8.13 16-Aug-2011
414    ------------------------
416    1.  The Unicode data tables have been updated to Unicode 6.0.0.
418    2.  Two minor typos in pcre_internal.h have been fixed.
420    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
421        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
422        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
424    4.  There were a number of related bugs in the code for matching backrefences
425        caselessly in UTF-8 mode when codes for the characters concerned were
426        different numbers of bytes. For example, U+023A and U+2C65 are an upper
427        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
428        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
429        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
430        2-byte code at the end of the subject (it thought there wasn't enough data
431        left).
433    5.  Comprehensive information about what went wrong is now returned by
434        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
435        as the output vector has at least 2 elements. The offset of the start of
436        the failing character and a reason code are placed in the vector.
438    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
439        now returned is for the first byte of the failing character, instead of the
440        last byte inspected. This is an incompatible change, but I hope it is small
441        enough not to be a problem. It makes the returned offset consistent with
442        pcre_exec() and pcre_dfa_exec().
444    7.  pcretest now gives a text phrase as well as the error number when
445        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
446        failure, the offset and reason code are output.
448    8.  When \R was used with a maximizing quantifier it failed to skip backwards
449        over a \r\n pair if the subsequent match failed. Instead, it just skipped
450        back over a single character (\n). This seems wrong (because it treated the
451        two characters as a single entity when going forwards), conflicts with the
452        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
453        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
454        has been changed.
456    9.  Some internal refactoring has changed the processing so that the handling
457        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
458        time (the PCRE_DOTALL option was changed this way some time ago: version
459        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
460        which was always a bit of a fudge. It also means that there is one less
461        argument for the match() function, which reduces its stack requirements
462        slightly. This change also fixes an incompatibility with Perl: the pattern
463        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
465    10. More internal refactoring has drastically reduced the number of recursive
466        calls to match() for possessively repeated groups such as (abc)++ when
467        using pcre_exec().
469    11. While implementing 10, a number of bugs in the handling of groups were
470        discovered and fixed:
472        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
473        (a|)*(?1) gave a compile-time internal error.
474        ((a|)+)+  did not notice that the outer group could match an empty string.
475        (^a|^)+   was not marked as anchored.
476        (.*a|.*)+ was not marked as matching at start or after a newline.
478    12. Yet more internal refactoring has removed another argument from the match()
479        function. Special calls to this function are now indicated by setting a
480        value in a variable in the "match data" data block.
482    13. Be more explicit in pcre_study() instead of relying on "default" for
483        opcodes that mean there is no starting character; this means that when new
484        ones are added and accidentally left out of pcre_study(), testing should
485        pick them up.
487    14. The -s option of pcretest has been documented for ages as being an old
488        synonym of -m (show memory usage). I have changed it to mean "force study
489        for every regex", that is, assume /S for every regex. This is similar to -i
490        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
491        using it. It makes it easier to run collections of tests with and without
492        study enabled, and thereby test pcre_study() more easily. All the standard
493        tests are now run with and without -s (but some patterns can be marked as
494        "never study" - see 20 below).
496    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
497        restoration of the capturing data to the outer values was not happening
498        correctly.
500    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
501        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
502        pattern had matched an empty string, and so incorrectly returned a no
503        match.
505    17. There was optimizing code for the last branch of non-capturing parentheses,
506        and also for the obeyed branch of a conditional subexpression, which used
507        tail recursion to cut down on stack usage. Unfortunately, now that there is
508        the possibility of (*THEN) occurring in these branches, tail recursion is
509        no longer possible because the return has to be checked for (*THEN). These
510        two optimizations have therefore been removed. [But see 8.20/11 above.]
512    18. If a pattern containing \R was studied, it was assumed that \R always
513        matched two bytes, thus causing the minimum subject length to be
514        incorrectly computed because \R can also match just one byte.
516    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
517        was incorrectly computed.
519    20. If /S is present twice on a test pattern in pcretest input, it now
520        *disables* studying, thereby overriding the use of -s on the command line
521        (see 14 above). This is necessary for one or two tests to keep the output
522        identical in both cases.
524    21. When (*ACCEPT) was used in an assertion that matched an empty string and
525        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
527    22. When an atomic group that contained a capturing parenthesis was
528        successfully matched, but the branch in which it appeared failed, the
529        capturing was not being forgotten if a higher numbered group was later
530        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
531        group 1 to "a", when in fact it should be unset. This applied to multi-
532        branched capturing and non-capturing groups, repeated or not, and also to
533        positive assertions (capturing in negative assertions does not happen
534        in PCRE) and also to nested atomic groups.
536    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
537        subject after a captured substring, to make it easier to tell which of a
538        number of identical substrings has been captured.
540    24. The way atomic groups are processed by pcre_exec() has been changed so that
541        if they are repeated, backtracking one repetition now resets captured
542        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
543        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
544        "aaa". Previously, it would have been "a". As part of this code
545        refactoring, the way recursive calls are handled has also been changed.
547    25. If an assertion condition captured any substrings, they were not passed
548        back unless some other capturing happened later. For example, if
549        (?(?=(a))a) was matched against "a", no capturing was returned.
551    26. When studying a pattern that contained subroutine calls or assertions,
552        the code for finding the minimum length of a possible match was handling
553        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
554        group 1 called group 2 while simultaneously a separate group 2 called group
555        1). A stack overflow occurred in this case. I have fixed this by limiting
556        the recursion depth to 10.
558    27. Updated RunTest.bat in the distribution to the version supplied by Tom
559        Fortmann. This supports explicit test numbers on the command line, and has
560        argument validation and error reporting.
562    28. An instance of \X with an unlimited repeat could fail if at any point the
563        first character it looked at was a mark character.
565    29. Some minor code refactoring concerning Unicode properties and scripts
566        should reduce the stack requirement of match() slightly.
568    30. Added the '=' option to pcretest to check the setting of unused capturing
569        slots at the end of the pattern, which are documented as being -1, but are
570        not included in the return count.
572    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
573        compiled something random. Now it gives a compile-time error (as does
574        Perl).
576    32. A *MARK encountered during the processing of a positive assertion is now
577        recorded and passed back (compatible with Perl).
579    33. If --only-matching or --colour was set on a pcregrep call whose pattern
580        had alternative anchored branches, the search for a second match in a line
581        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
582        matched the line "0102" twice. The same bug affected patterns that started
583        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
584        twice.
586    34. Previously, PCRE did not allow quantification of assertions. However, Perl
587        does, and because of capturing effects, quantifying parenthesized
588        assertions may at times be useful. Quantifiers are now allowed for
589        parenthesized assertions.
591    35. A minor code tidy in pcre_compile() when checking options for \R usage.
593    36. \g was being checked for fancy things in a character class, when it should
594        just be a literal "g".
596    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
597        appearance of a nested POSIX class supersedes an apparent external class.
598        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
599        unescaped square brackets may also appear as part of class names. For
600        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
601        more like Perl. (But see 8.20/1 above.)
603    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
604        was because it thought it was \N{name}, which is not supported).
606    39. Add minix to OS list not supporting the -S option in pcretest.
608    40. PCRE tries to detect cases of infinite recursion at compile time, but it
609        cannot analyze patterns in sufficient detail to catch mutual recursions
610        such as ((?1))((?2)). There is now a runtime test that gives an error if a
611        subgroup is called recursively as a subpattern for a second time at the
612        same position in the subject string. In previous releases this might have
613        been caught by the recursion limit, or it might have run out of stack.
615    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
616        happen only once. PCRE was, however incorrectly giving a compile time error
617        "recursive call could loop indefinitely" because it cannot analyze the
618        pattern in sufficient detail. The compile time test no longer happens when
619        PCRE is compiling a conditional subpattern, but actual runaway loops are
620        now caught at runtime (see 40 above).
622    42. It seems that Perl allows any characters other than a closing parenthesis
623        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
624        has been changed to be the same.
626    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
627        as not to get warnings when autogen.sh is called. Also changed
628        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
630    44. To help people who use pcregrep to scan files containing exceedingly long
631        lines, the following changes have been made:
633        (a) The default value of the buffer size parameter has been increased from
634            8K to 20K. (The actual buffer used is three times this size.)
636        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
637            PCRE is built.
639        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
640            to be set at run time.
642        (d) Numerical values in pcregrep options can be followed by K or M, for
643            example --buffer-size=50K.
645        (e) If a line being scanned overflows pcregrep's buffer, an error is now
646            given and the return code is set to 2.
648    45. Add a pointer to the latest mark to the callout data block.
650    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
651        partial match of an empty string instead of no match. This was specific to
652        the use of ".".
654    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
655        complete match instead of a partial match. This bug was dependent on both
656        the PCRE_UTF8 and PCRE_DOTALL options being set.
658    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
659        starting byte set, because \b was not being ignored.
662    Version 8.12 15-Jan-2011
663    ------------------------
665    1.  Fixed some typos in the markup of the man pages, and wrote a script that
666        checks for such things as part of the documentation building process.
668    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
669        --match-limit and --recursion-limit options (added for 8.11). In
670        particular, this made one of the standard tests fail. (The integer value
671        went into the wrong half of a long int.)
673    3.  If the --colour option was given to pcregrep with -v (invert match), it
674        did strange things, either producing crazy output, or crashing. It should,
675        of course, ignore a request for colour when reporting lines that do not
676        match.
678    4.  Another pcregrep bug caused similar problems if --colour was specified with
679        -M (multiline) and the pattern match finished with a line ending.
681    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
682        matched in multiline mode, the following line was shown as part of the
683        match. This seems wrong, so I have changed it.
685    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
686        the check for further matches in the same line (so they could be coloured)
687        to overrun the end of the current line. If another match was found, it was
688        incorrectly shown (and then shown again when found in the next line).
690    7.  If pcregrep was compiled under Windows, there was a reference to the
691        function pcregrep_exit() before it was defined. I am assuming this was
692        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
693        reported by a user. I've moved the definition above the reference.
696    Version 8.11 10-Dec-2010
697    ------------------------
699    1.  (*THEN) was not working properly if there were untried alternatives prior
700        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
701        backtracked to try for "b" instead of moving to the next alternative branch
702        at the same level (in this case, to look for "c"). The Perl documentation
703        is clear that when (*THEN) is backtracked onto, it goes to the "next
704        alternative in the innermost enclosing group".
706    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
707        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
708        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
709        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
710        (*THEN).
712    3.  If \s appeared in a character class, it removed the VT character from
713        the class, even if it had been included by some previous item, for example
714        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
715        of \s, but is part of the POSIX "space" class.)
717    4.  A partial match never returns an empty string (because you can always
718        match an empty string at the end of the subject); however the checking for
719        an empty string was starting at the "start of match" point. This has been
720        changed to the "earliest inspected character" point, because the returned
721        data for a partial match starts at this character. This means that, for
722        example, /(?<=abc)def/ gives a partial match for the subject "abc"
723        (previously it gave "no match").
725    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
726        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
727        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
728        has an implication that the given string is incomplete (because a partial
729        match is preferred over a full match). For this reason, these items now
730        give a partial match in this situation. [Aside: previously, the one case
731        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
732        match rather than a full match, which was wrong by the old rules, but is
733        now correct.]
735    6.  There was a bug in the handling of #-introduced comments, recognized when
736        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
737        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
738        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
739        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
740        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
741        places in pcre_compile().
743    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
744        comments when looking ahead for named forward references to subpatterns,
745        the only newline sequence it recognized was NL. It now handles newlines
746        according to the set newline convention.
748    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
749        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
750        cater for a lack of strerror(). These oversights have been fixed.
752    9.  Added --match-limit and --recursion-limit to pcregrep.
754    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
756    11. When the -o option was used, pcregrep was setting a return code of 1, even
757        when matches were found, and --line-buffered was not being honoured.
759    12. Added an optional parentheses number to the -o and --only-matching options
760        of pcregrep.
762    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
763        can match an empty string. The code to do it in pcretest and pcredemo
764        needed fixing:
766        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
767            only one byte after an empty string match just before CRLF (this case
768            just got forgotten; "any" and "anycrlf" were OK).
770        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
771            mode when an empty string match preceded an ASCII character followed by
772            a non-ASCII character. (The code for advancing by one character rather
773            than one byte was nonsense.)
775        (c) The pcredemo.c sample program did not have any code at all to handle
776            the cases when CRLF is a valid newline sequence.
778    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
779        as a starting offset was within the subject string. There is now a new
780        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
781        negative or greater than the length of the string. In order to test this,
782        pcretest is extended to allow the setting of negative starting offsets.
784    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
785        starting offset points to the beginning of a UTF-8 character was
786        unnecessarily clumsy. I tidied it up.
788    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
789        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
791    17. Nobody had reported that the --include_dir option, which was added in
792        release 7.7 should have been called --include-dir (hyphen, not underscore)
793        for compatibility with GNU grep. I have changed it to --include-dir, but
794        left --include_dir as an undocumented synonym, and the same for
795        --exclude-dir, though that is not available in GNU grep, at least as of
796        release 2.5.4.
798    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
799        characters from a string of bytes) have been redefined so as not to use
800        loops, in order to improve performance in some environments. At the same
801        time, I abstracted some of the common code into auxiliary macros to save
802        repetition (this should not affect the compiled code).
804    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
805        compile-time error is now given if \c is not followed by an ASCII
806        character, that is, a byte less than 128. (In EBCDIC mode, the code is
807        different, and any byte value is allowed.)
809    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
810        START_OPTIMIZE option, which is now allowed at compile time - but just
811        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
812        to pcregrep and other applications that have no direct access to PCRE
813        options. The new /Y option in pcretest sets this option when calling
814        pcre_compile().
816    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
817        back references. Groups containing recursive back references were forced to
818        be atomic by that change, but in the case of named groups, the amount of
819        memory required was incorrectly computed, leading to "Failed: internal
820        error: code overflow". This has been fixed.
822    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
823        pcretest.c, to avoid build problems in some Borland environments.
826    Version 8.10 25-Jun-2010
827    ------------------------
829    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
830        THEN.
832    2.  (*ACCEPT) was not working when inside an atomic group.
834    3.  Inside a character class, \B is treated as a literal by default, but
835        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
836        causes the error). The code is unchanged, but I tidied the documentation.
838    4.  Inside a character class, PCRE always treated \R and \X as literals,
839        whereas Perl faults them if its -w option is set. I have changed PCRE so
840        that it faults them when PCRE_EXTRA is set.
842    5.  Added support for \N, which always matches any character other than
843        newline. (It is the same as "." when PCRE_DOTALL is not set.)
845    6.  When compiling pcregrep with newer versions of gcc which may have
846        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
847        declared with attribute warn_unused_result" were given. Just casting the
848        result to (void) does not stop the warnings; a more elaborate fudge is
849        needed. I've used a macro to implement this.
851    7.  Minor change to pcretest.c to avoid a compiler warning.
853    8.  Added four artifical Unicode properties to help with an option to make
854        \s etc use properties (see next item). The new properties are: Xan
855        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
857    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
858        use Unicode properties. (*UCP) at the start of a pattern can be used to set
859        this option. Modified pcretest to add /W to test this facility. Added
860        REG_UCP to make it available via the POSIX interface.
862    10. Added --line-buffered to pcregrep.
864    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
865        studied, and the match started with a letter with a code point greater than
866        127 whose first byte was different to the first byte of the other case of
867        the letter, the other case of this starting letter was not recognized
868        (#976).
870    12. If a pattern that was studied started with a repeated Unicode property
871        test, for example, \p{Nd}+, there was the theoretical possibility of
872        setting up an incorrect bitmap of starting bytes, but fortunately it could
873        not have actually happened in practice until change 8 above was made (it
874        added property types that matched character-matching opcodes).
876    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
877        possible starting bytes for non-anchored patterns.
879    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
880        \R, and also a number of cases that involve Unicode properties, both
881        explicit and implicit when PCRE_UCP is set.
883    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
884        input, it could crash or give wrong results if characters with values
885        greater than 0xc0 were present in the subject string. (Detail: it assumed
886        UTF-8 input when processing these items.)
888    16. Added a lot of (int) casts to avoid compiler warnings in systems where
889        size_t is 64-bit (#991).
891    17. Added a check for running out of memory when PCRE is compiled with
892        --disable-stack-for-recursion (#990).
894    18. If the last data line in a file for pcretest does not have a newline on
895        the end, a newline was missing in the output.
897    19. The default pcre_chartables.c file recognizes only ASCII characters (values
898        less than 128) in its various bitmaps. However, there is a facility for
899        generating tables according to the current locale when PCRE is compiled. It
900        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
901        characters, are recognized by isspace() and therefore were getting set in
902        these tables, and indeed these tables seem to approximate to ISO 8859. This
903        caused a problem in UTF-8 mode when pcre_study() was used to create a list
904        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
905        which of course cannot start UTF-8 characters. I have changed the code so
906        that only real ASCII characters (less than 128) and the correct starting
907        bytes for UTF-8 encodings are set for characters greater than 127 when in
908        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
909        altogether.)
911    20. Added the /T option to pcretest so as to be able to run tests with non-
912        standard character tables, thus making it possible to include the tests
913        used for 19 above in the standard set of tests.
915    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
916        reference to a subpattern the other side of a comment that contains an
917        opening parenthesis caused either an internal compiling error, or a
918        reference to the wrong subpattern.
921    Version 8.02 19-Mar-2010
922    ------------------------
924    1.  The Unicode data tables have been updated to Unicode 5.2.0.
926    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
927        configured.
929    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
930        original author of that file, following a query about its status.
932    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
933        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
935    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
936        quantifier applied to a forward-referencing subroutine call, could compile
937        incorrect code or give the error "internal error: previously-checked
938        referenced subpattern not found".
940    6.  Both MS Visual Studio and Symbian OS have problems with initializing
941        variables to point to external functions. For these systems, therefore,
942        pcre_malloc etc. are now initialized to local functions that call the
943        relevant global functions.
945    7.  There were two entries missing in the vectors called coptable and poptable
946        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
947        I've fixed the data, and added a kludgy way of testing at compile time that
948        the lengths are correct (equal to the number of opcodes).
950    8.  Following on from 7, I added a similar kludge to check the length of the
951        eint vector in pcreposix.c.
953    9.  Error texts for pcre_compile() are held as one long string to avoid too
954        much relocation at load time. To find a text, the string is searched,
955        counting zeros. There was no check for running off the end of the string,
956        which could happen if a new error number was added without updating the
957        string.
959    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
961    11. \K was not working if it appeared in an atomic group or in a group that
962        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
963        \K is "not well defined" if used in an assertion. PCRE now accepts it if
964        the assertion is positive, but not if it is negative.
966    12. Change 11 fortuitously reduced the size of the stack frame used in the
967        "match()" function of pcre_exec.c by one pointer. Forthcoming
968        implementation of support for (*MARK) will need an extra pointer on the
969        stack; I have reserved it now, so that the stack frame size does not
970        decrease.
972    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
973        item in branch that calls a recursion is a subroutine call - as in the
974        second branch in the above example - was incorrectly given the compile-
975        time error "recursive call could loop indefinitely" because pcre_compile()
976        was not correctly checking the subroutine for matching a non-empty string.
978    14. The checks for overrunning compiling workspace could trigger after an
979        overrun had occurred. This is a "should never occur" error, but it can be
980        triggered by pathological patterns such as hundreds of nested parentheses.
981        The checks now trigger 100 bytes before the end of the workspace.
983    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
986    Version 8.01 19-Jan-2010
987    ------------------------
989    1.  If a pattern contained a conditional subpattern with only one branch (in
990        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
991        computed the wrong minimum data length (which is of course zero for such
992        subpatterns). This could cause incorrect "no match" results.
994    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
995        the pattern is reset in the first branch, pcre_compile() failed with
996        "internal error: code overflow at offset...". This happened only when
997        the reset was to the original external option setting. (An optimization
998        abstracts leading options settings into an external setting, which was the
999        cause of this.)
1001    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1002        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1003        assertion pattern did not match (meaning that the assertion was true), it
1004        was incorrectly treated as false if the SKIP had been reached during the
1005        matching. This also applied to assertions used as conditions.
1007    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1008        assertion subpattern, including such a pattern used as a condition,
1009        unpredictable results occurred, instead of the error return
1012    5.  The C++ GlobalReplace function was not working like Perl for the special
1013        situation when an empty string is matched. It now does the fancy magic
1014        stuff that is necessary.
1016    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1017        removed. (These were left over from very, very early versions of PCRE.)
1019    7.  Some cosmetic changes to the code to make life easier when compiling it
1020        as part of something else:
1022        (a) Change DEBUG to PCRE_DEBUG.
1024        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1025            called "current" as "current_branch", to prevent a collision with the
1026            Linux macro when compiled as a kernel module.
1028        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1029            prevent a collision with the Linux macro when compiled as a kernel
1030            module.
1032    8.  In pcre_compile() there are some checks for integer overflows that used to
1033        cast potentially large values to (double). This has been changed to that
1034        when building, a check for int64_t is made, and if it is found, it is used
1035        instead, thus avoiding the use of floating point arithmetic. (There is no
1036        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1037        double.
1039    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1040        2005 (difference between two addresses compared to an unsigned value).
1042    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1043        custom one, because of the following reported problem in Windows:
1045          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1046              under Win32.
1047          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1048              therefore missing the function definition.
1049          - The compiler thus generates a "C" signature for the test function.
1050          - The linker fails to find the "C" function.
1051          - PCRE fails to configure if asked to do so against libbz2.
1053    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1054        messages were output:
1056          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1057          rerunning libtoolize, to keep the correct libtool macros in-tree.
1058          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1060        I have done both of these things.
1062    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1063        most of the time, it *can* run out if it is given a pattern that contains a
1064        runaway infinite recursion. I updated the discussion in the pcrestack man
1065        page.
1067    13. Now that we have gone to the x.xx style of version numbers, the minor
1068        version may start with zero. Using 08 or 09 is a bad idea because users
1069        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1070        interpreted as invalid octal numbers. I've updated the previous comment in
1071        configure.ac, and also added a check that gives an error if 08 or 09 are
1072        used.
1074    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1075        causing partial matching to fail when the end of the subject matched \W
1076        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1078    15. There were some discrepancies between the declarations in pcre_internal.h
1079        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1080        their definitions. The declarations used "const uschar *" and the
1081        definitions used USPTR. Even though USPTR is normally defined as "const
1082        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1083        reported that: "This difference in casting confuses some C++ compilers, for
1084        example, SunCC recognizes above declarations as different functions and
1085        generates broken code for hbpcre." I have changed the declarations to use
1086        USPTR.
1088    16. GNU libtool is named differently on some systems. The autogen.sh script now
1089        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1090        (FreeBSD).
1092    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1093        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1094        comment: "Figure out how to create a longlong from a string: strtoll and
1095        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1096        instance, but it only takes 2 args instead of 3!"
1098    18. A subtle bug concerned with back references has been fixed by a change of
1099        specification, with a corresponding code fix. A pattern such as
1100        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1101        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1102        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1103        same bug. Such groups have to be quantified to be useful, or contained
1104        inside another quantified group. (If there's no repetition, the reference
1105        can never match.) The problem arises because, having left the group and
1106        moved on to the rest of the pattern, a later failure that backtracks into
1107        the group uses the captured value from the final iteration of the group
1108        rather than the correct earlier one. I have fixed this in PCRE by forcing
1109        any group that contains a reference to itself to be an atomic group; that
1110        is, there cannot be any backtracking into it once it has completed. This is
1111        similar to recursive and subroutine calls.
1114    Version 8.00 19-Oct-09
1115    ----------------------
1117    1.  The table for translating pcre_compile() error codes into POSIX error codes
1118        was out-of-date, and there was no check on the pcre_compile() error code
1119        being within the table. This could lead to an OK return being given in
1120        error.
1122    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1123        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1124        in a Windows environment.
1126    3.  The pcregrep --count option prints the count for each file even when it is
1127        zero, as does GNU grep. However, pcregrep was also printing all files when
1128        --files-with-matches was added. Now, when both options are given, it prints
1129        counts only for those files that have at least one match. (GNU grep just
1130        prints the file name in this circumstance, but including the count seems
1131        more useful - otherwise, why use --count?) Also ensured that the
1132        combination -clh just lists non-zero counts, with no names.
1134    4.  The long form of the pcregrep -F option was incorrectly implemented as
1135        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1136        but it seems right to fix it, and I didn't think it was worth preserving
1137        the old behaviour.
1139    5.  The command line items --regex=pattern and --regexp=pattern were not
1140        recognized by pcregrep, which required --regex pattern or --regexp pattern
1141        (with a space rather than an '='). The man page documented the '=' forms,
1142        which are compatible with GNU grep; these now work.
1144    6.  No libpcreposix.pc file was created for pkg-config; there was just
1145        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1147    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1148        when UCP support is not needed, by modifying the Python script that
1149        generates it from Unicode data files. This should not matter if the module
1150        is correctly used as a library, but I received one complaint about 50K of
1151        unwanted data. My guess is that the person linked everything into his
1152        program rather than using a library. Anyway, it does no harm.
1154    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1155        was a minimum greater than 1 for a wide character in a possessive
1156        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1157        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1158        character. Chaos in the form of incorrect output or a compiling loop could
1159        result.
1161    9.  The restrictions on what a pattern can contain when partial matching is
1162        requested for pcre_exec() have been removed. All patterns can now be
1163        partially matched by this function. In addition, if there are at least two
1164        slots in the offset vector, the offset of the earliest inspected character
1165        for the match and the offset of the end of the subject are set in them when
1166        PCRE_ERROR_PARTIAL is returned.
1168    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1169        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1170        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1171        and may be more useful for multi-segment matching.
1173    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1174        used to be given if ever the end of the subject was reached; now it is
1175        given only if matching could not proceed because another character was
1176        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1177        string "Z", which now yields "no match" instead of "partial match". In the
1178        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1179        final character ended with (*FAIL).
1181    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1182        if the pattern had a "must contain" character that was already found in the
1183        earlier partial match, unless partial matching was again requested. For
1184        example, with the pattern /dog.(body)?/, the "must contain" character is
1185        "g". If the first part-match was for the string "dog", restarting with
1186        "sbody" failed. This bug has been fixed.
1188    13. The string returned by pcre_dfa_exec() after a partial match has been
1189        changed so that it starts at the first inspected character rather than the
1190        first character of the match. This makes a difference only if the pattern
1191        starts with a lookbehind assertion or \b or \B (\K is not supported by
1192        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1193        matching functions compatible, and I think it's the right thing to do.
1195    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1196        so that the demonstration program is easily available in environments where
1197        PCRE has not been installed from source.
1199    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1200        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1201        library.
1203    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1204        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1205        is not the first non-POSIX option to be added. Clearly some people find
1206        these options useful.
1208    17. If a caller to the POSIX matching function regexec() passes a non-zero
1209        value for nmatch with a NULL value for pmatch, the value of
1210        nmatch is forced to zero.
1212    18. RunGrepTest did not have a test for the availability of the -u option of
1213        the diff command, as RunTest does. It now checks in the same way as
1214        RunTest, and also checks for the -b option.
1216    19. If an odd number of negated classes containing just a single character
1217        interposed, within parentheses, between a forward reference to a named
1218        subpattern and the definition of the subpattern, compilation crashed with
1219        an internal error, complaining that it could not find the referenced
1220        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1221        [The bug was that it was starting one character too far in when skipping
1222        over the character class, thus treating the ] as data rather than
1223        terminating the class. This meant it could skip too much.]
1225    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1226        /g option in pcretest when the pattern contains \K, which makes it possible
1227        to have an empty string match not at the start, even when the pattern is
1228        anchored. Updated pcretest and pcredemo to use this option.
1230    21. If the maximum number of capturing subpatterns in a recursion was greater
1231        than the maximum at the outer level, the higher number was returned, but
1232        with unset values at the outer level. The correct (outer level) value is
1233        now given.
1235    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1236        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1237        make it do so. The string so far is captured, making this feature
1238        compatible with Perl.
1240    23. The tests have been re-organized, adding tests 11 and 12, to make it
1241        possible to check the Perl 5.10 features against Perl 5.10.
1243    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1244        pattern matches a fixed length string. PCRE did not allow this; now it
1245        does. Neither allows recursion.
1247    25. I finally figured out how to implement a request to provide the minimum
1248        length of subject string that was needed in order to match a given pattern.
1249        (It was back references and recursion that I had previously got hung up
1250        on.) This code has now been added to pcre_study(); it finds a lower bound
1251        to the length of subject needed. It is not necessarily the greatest lower
1252        bound, but using it to avoid searching strings that are too short does give
1253        some useful speed-ups. The value is available to calling programs via
1254        pcre_fullinfo().
1256    26. While implementing 25, I discovered to my embarrassment that pcretest had
1257        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1258        study optimizations had never been tested with that matching function.
1259        Oops. What is worse, even when it was passed study data, there was a bug in
1260        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1261        were also very few tests of studied patterns with pcre_dfa_exec().
1263    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1264        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1265        on the other side of the coin, they are no longer allowed to have different
1266        names, because these cannot be distinguished in PCRE, and this has caused
1267        confusion. (This is a difference from Perl.)
1269    28. When duplicate subpattern names are present (necessarily with different
1270        numbers, as required by 27 above), and a test is made by name in a
1271        conditional pattern, either for a subpattern having been matched, or for
1272        recursion in such a pattern, all the associated numbered subpatterns are
1273        tested, and the overall condition is true if the condition is true for any
1274        one of them. This is the way Perl works, and is also more like the way
1275        testing by number works.
1278    Version 7.9 11-Apr-09
1279    ---------------------
1281    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1282        (pcretest), all targets were linked against these libraries. This included
1283        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1284        libraries. This caused unwanted dependencies to be created. This problem
1285        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1286        pcretest is linked with readline.
1288    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1289        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1290        moved outside it again, because FALSE and TRUE are already defined in AIX,
1291        but BOOL is not.
1293    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1294        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1296    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1297        hyphens) following file names and line numbers when outputting matching
1298        lines. This is not true; no spaces are inserted. I have also clarified the
1299        wording for the --colour (or --color) option.
1301    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1302        was not coloured; this is different to GNU grep, so I have changed it to be
1303        the same.
1305    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1306        each matching line was coloured. Now it goes on to look for further matches
1307        of any of the test patterns, which is the same behaviour as GNU grep.
1309    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1310        doesn't make sense to accept an empty string match in pcregrep, so I have
1311        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1312        seems to be how GNU grep behaves.
1314    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1315        start or after a newline", because the conditional assertion was not being
1316        correctly handled. The rule now is that both the assertion and what follows
1317        in the first alternative must satisfy the test.
1319    9.  If auto-callout was enabled in a pattern with a conditional group whose
1320        condition was an assertion, PCRE could crash during matching, both with
1321        pcre_exec() and pcre_dfa_exec().
1323    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1324        used for matching.
1326    11. Unicode property support in character classes was not working for
1327        characters (bytes) greater than 127 when not in UTF-8 mode.
1329    12. Added the -M command line option to pcretest.
1331    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1333    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1335    16. Added comments and documentation about mis-use of no_arg in the C++
1336        wrapper.
1338    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1339        from Martin Jerabek that uses macro names for all relevant character and
1340        string constants.
1342    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1343        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1344        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1345        these, but not everybody uses configure.
1347    19. A conditional group that had only one branch was not being correctly
1348        recognized as an item that could match an empty string. This meant that an
1349        enclosing group might also not be so recognized, causing infinite looping
1350        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1351        with the subject "ab", where knowledge that the repeated group can match
1352        nothing is needed in order to break the loop.
1354    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1355        exec(), but without supplying a callout function, matching went wrong.
1357    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1358        leak if the size of the offset vector was greater than 30. When the vector
1359        is smaller, the saved offsets during recursion go onto a local stack
1360        vector, but for larger vectors malloc() is used. It was failing to free
1361        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1362        error, in fact).
1364    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1365        heapframe that is used only when UTF-8 support is enabled. This caused no
1366        problem, but was untidy.
1368    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1369        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1370        included within another project.
1372    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1373        slightly modified by me:
1375          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1376              not building pcregrep.
1378          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1379              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1381    25. Forward references, both numeric and by name, in patterns that made use of
1382        duplicate group numbers, could behave incorrectly or give incorrect errors,
1383        because when scanning forward to find the reference group, PCRE was not
1384        taking into account the duplicate group numbers. A pattern such as
1385        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1387    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1388        the feature of a custom pointer more persuasive (as requested by a user).
1390    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1391        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1392        pre-defined.
1394    28. Added support for (*UTF8) at the start of a pattern.
1396    29. Arrange for flags added by the "release type" setting in CMake to be shown
1397        in the configuration summary.
1400    Version 7.8 05-Sep-08
1401    ---------------------
1403    1.  Replaced UCP searching code with optimized version as implemented for Ad
1404        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1405        stage table and inline lookup instead of a function, giving speed ups of 2
1406        to 5 times on some simple patterns that I tested. Permission was given to
1407        distribute the MultiStage2.py script that generates the tables (it's not in
1408        the tarball, but is in the Subversion repository).
1410    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1411        scripts.
1413    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1414        a group with a zero qualifier. The result of the study could be incorrect,
1415        or the function might crash, depending on the pattern.
1417    4.  Caseless matching was not working for non-ASCII characters in back
1418        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1419        It now works when Unicode Property Support is available.
1421    5.  In pcretest, an escape such as \x{de} in the data was always generating
1422        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1423        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1424        truncation.
1426    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1428    7.  Added two (int) casts to pcregrep when printing the difference of two
1429        pointers, in case they are 64-bit values.
1431    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1432        test 2 if it fails.
1434    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1435        and a #define of that name to empty if it is not externally set. This is to
1436        allow users of MSVC to set it if necessary.
1438    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1439        the convenience functions in the pcre_get.c source file.
1441    11. An option change at the start of a pattern that had top-level alternatives
1442        could cause overwriting and/or a crash. This command provoked a crash in
1443        some environments:
1445          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1447        This potential security problem was recorded as CVE-2008-2371.
1449    12. For a pattern where the match had to start at the beginning or immediately
1450        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1451        pcre_dfa_exec() could read past the end of the passed subject if there was
1452        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1453        pcretest so that it places the subject at the end of its malloc-ed buffer.
1455    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1456        exec() might read past the end of the data buffer in UTF-8 mode.
1458    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1459        the data contained the byte 0x85 as part of a UTF-8 character within its
1460        first line. This applied both to normal and DFA matching.
1462    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1463        /^[^d]*?$/8 failed to match "abc".
1465    16. Added a missing copyright notice to pcrecpp_internal.h.
1467    17. Make it more clear in the documentation that values returned from
1468        pcre_exec() in ovector are byte offsets, not character counts.
1470    18. Tidied a few places to stop certain compilers from issuing warnings.
1472    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1473        supplied by Stefan Weber. I made a further small update for 7.8 because
1474        there is a change of source arrangements: the pcre_searchfuncs.c module is
1475        replaced by pcre_ucd.c.
1478    Version 7.7 07-May-08
1479    ---------------------
1481    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1482        a string to a long long, pretend we don't even have a long long." This is
1483        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1485    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1486        pre-7.6 versions, which defined a global no_arg variable instead of putting
1487        it in the RE class. (See also #8 below.)
1489    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1490        Lopes.
1492    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1494        (1) The include/exclude patterns were being applied to the whole pathnames
1495            of files, instead of just to the final components.
1497        (2) If there was more than one level of directory, the subdirectories were
1498            skipped unless they satisfied the include/exclude conditions. This is
1499            inconsistent with GNU grep (and could even be seen as contrary to the
1500            pcregrep specification - which I improved to make it absolutely clear).
1501            The action now is always to scan all levels of directory, and just
1502            apply the include/exclude patterns to regular files.
1504    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1505        --exclude_dir in the tests to avoid scanning .svn directories.
1507    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1508        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1509        doesn't support NULs in patterns.
1511    7.  Added some missing "const"s to declarations of static tables in
1512        pcre_compile.c and pcre_dfa_exec.c.
1514    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1515        caused by fix #2  above. (Subsequently also a second patch to fix the
1516        first patch. And a third patch - this was a messy problem.)
1518    9.  Applied Craig's patch to remove the use of push_back().
1520    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1521        matching function regexec().
1523    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1524        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1525        references. PCRE supports relative numbers with this syntax (I don't think
1526        Oniguruma does).
1528    12. Previously, a group with a zero repeat such as (...){0} was completely
1529        omitted from the compiled regex. However, this means that if the group
1530        was called as a subroutine from elsewhere in the pattern, things went wrong
1531        (an internal error was given). Such groups are now left in the compiled
1532        pattern, with a new opcode that causes them to be skipped at execution
1533        time.
1535    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1536        to the way PCRE behaves:
1538        (a) A lone ] character is dis-allowed (Perl treats it as data).
1540        (b) A back reference to an unmatched subpattern matches an empty string
1541            (Perl fails the current match path).
1543        (c) A data ] in a character class must be notated as \] because if the
1544            first data character in a class is ], it defines an empty class. (In
1545            Perl it is not possible to have an empty class.) The empty class []
1546            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1547            The negative empty class [^] matches any one character, independently
1548            of the DOTALL setting.
1550    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1551        non-existent subpattern following a character class starting with ']' and
1552        containing () gave an internal compiling error instead of "reference to
1553        non-existent subpattern". Fortunately, when the pattern did exist, the
1554        compiled code was correct. (When scanning forwards to check for the
1555        existencd of the subpattern, it was treating the data ']' as terminating
1556        the class, so got the count wrong. When actually compiling, the reference
1557        was subsequently set up correctly.)
1559    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1560        it was being rejected as not supported by pcre_dfa_exec(), even though
1561        other assertions are supported. I have made pcre_dfa_exec() support
1562        (*FAIL).
1564    16. The implementation of 13c above involved the invention of a new opcode,
1565        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1566        cannot be changed at match time, I realized I could make a small
1567        improvement to matching performance by compiling OP_ALLANY instead of
1568        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1569        on the OP_ANY path.
1571    17. Compiling pcretest on Windows with readline support failed without the
1572        following two fixes: (1) Make the unistd.h include conditional on
1573        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1575    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1576        ncurses library to be included for pcretest when ReadLine support is
1577        requested, but also to allow for it to be overridden. This patch came from
1578        Daniel Bergström.
1580    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1581        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1582        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1583        spotting this.
1586    Version 7.6 28-Jan-08
1587    ---------------------
1589    1.  A character class containing a very large number of characters with
1590        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1591        overflow.
1593    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1594        HAVE_LONG_LONG is not defined.
1596    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1597        bring them up to date and include new features. This patch includes:
1599        - Fixed PH's badly added libz and libbz2 support.
1600        - Fixed a problem with static linking.
1601        - Added pcredemo. [But later removed - see 7 below.]
1602        - Fixed dftables problem and added an option.
1603        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1604            HAVE_LONG_LONG.
1605        - Added readline support for pcretest.
1606        - Added an listing of the option settings after cmake has run.
1608    4.  A user submitted a patch to Makefile that makes it easy to create
1609        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1610        Makefile.am that cause it to include this special target, without
1611        affecting anything else. Note that the same mingw target plus all
1612        the other distribution libraries and programs are now supported
1613        when configuring with CMake (see 6 below) instead of with
1614        Configure/Make.
1616    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1617        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1618        exported in the Windows port". It has not yet been confirmed that the patch
1619        solves the problem, but it does no harm.
1621    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1622        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1623        with CMake, and also correct the comment about stack recursion.
1625    7.  Remove the automatic building of pcredemo from the ./configure system and
1626        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1627        of a program that users should build themselves after PCRE is installed, so
1628        building it automatically is not really right. What is more, it gave
1629        trouble in some build environments.
1631    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1634    Version 7.5 10-Jan-08
1635    ---------------------
1637    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1638        values in parens when parsing an RE using the C++ wrapper."
1640    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1641        Characters greater than 255 were excluded from the class instead of being
1642        included.
1644    3.  The same bug as (2) above applied to negated POSIX classes such as
1645        [:^space:].
1647    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1648        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1649        I have changed it.
1651    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1652        first named subpattern) and a construct such as (?&a) would reference the
1653        first named subpattern whose name started with "a" (in other words, the
1654        length check was missing). Both these problems are fixed. "Subpattern name
1655        expected" is now given for (?&) (a zero-length name), and this patch also
1656        makes it give the same error for \k'' (previously it complained that that
1657        was a reference to a non-existent subpattern).
1659    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1660        this is right because (?- can be followed by option settings as well as by
1661        digits. I have, however, made the messages clearer.
1663    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1664        than the number used in the conditional) now cause a compile-time error.
1665        This is actually not compatible with Perl, which accepts such patterns, but
1666        treats the conditional as always being FALSE (as PCRE used to), but it
1667        seems to me that giving a diagnostic is better.
1669    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1670        and messages.
1672    9.  Fix two occurrences of "backslash" in comments that should have been
1673        "backspace".
1675    10. Remove two redundant lines of code that can never be obeyed (their function
1676        was moved elsewhere).
1678    11. The program that makes PCRE's Unicode character property table had a bug
1679        which caused it to generate incorrect table entries for sequences of
1680        characters that have the same character type, but are in different scripts.
1681        It amalgamated them into a single range, with the script of the first of
1682        them. In other words, some characters were in the wrong script. There were
1683        thirteen such cases, affecting characters in the following ranges:
1685          U+002b0 - U+002c1
1686          U+0060c - U+0060d
1687          U+0061e - U+00612
1688          U+0064b - U+0065e
1689          U+0074d - U+0076d
1690          U+01800 - U+01805
1691          U+01d00 - U+01d77
1692          U+01d9b - U+01dbf
1693          U+0200b - U+0200f
1694          U+030fc - U+030fe
1695          U+03260 - U+0327f
1696          U+0fb46 - U+0fbb1
1697          U+10450 - U+1049d
1699    12. The -o option (show only the matching part of a line) for pcregrep was not
1700        compatible with GNU grep in that, if there was more than one match in a
1701        line, it showed only the first of them. It now behaves in the same way as
1702        GNU grep.
1704    13. If the -o and -v options were combined for pcregrep, it printed a blank
1705        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1706        does the same. The return code can be used to tell if there were any
1707        non-matching lines.
1709    14. Added --file-offsets and --line-offsets to pcregrep.
1711    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1712        infinitely looping recursion. The bug was that positive lookaheads were not
1713        being skipped when checking for a possible empty match (negative lookaheads
1714        and both kinds of lookbehind were skipped).
1716    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1717        inclusion of <windows.h> to before rather than after the definition of
1718        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1720    17. Specifying a possessive quantifier with a specific limit for a Unicode
1721        character property caused pcre_compile() to compile bad code, which led at
1722        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1723        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1724        caused the error; without that there was no problem.
1726    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1728    19. Added --enable-pcretest-libreadline.
1730    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1731        RE::GlobalReplace(). As a result, the number of replacements returned was
1732        double what it should be. I removed one of the increments, but Craig sent a
1733        later patch that removed the other one (the right fix) and added unit tests
1734        that check the return values (which was not done before).
1736    21. Several CMake things:
1738        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1739            the names libpcre and libpcreposix, not just pcre and pcreposix.
1741        (2) The above change means that pcretest and pcregrep are now correctly
1742            linked with the newly-built libraries, not previously installed ones.
1746    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1747        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1748        UTF-8 newline character). The key issue is that the pattern starts .*;
1749        this means that the match must be either at the beginning, or after a
1750        newline. The bug was in the code for advancing after a failed match and
1751        checking that the new position followed a newline. It was not taking
1752        account of UTF-8 characters correctly.
1754    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1755        character classes. PCRE was not treating the sequence [:...:] as a
1756        character class unless the ... were all letters. Perl, however, seems to
1757        allow any characters between [: and :], though of course it rejects as
1758        unknown any "names" that contain non-letters, because all the known class
1759        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1760        for example, whereas PCRE did not - it did not recognize a POSIX character
1761        class. This seemed a bit dangerous, so the code has been changed to be
1762        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1763        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1764        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1765        Perl does, and where it didn't before.
1767    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1768        Windows environments %n is disabled by default.
1771    Version 7.4 21-Sep-07
1772    ---------------------
1774    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1775        means that a class such as [\s] counted as "explicit reference to CR or
1776        LF". That isn't really right - the whole point of the change was to try to
1777        help when there was an actual mention of one of the two characters. So now
1778        the change happens only if \r or \n (or a literal CR or LF) character is
1779        encountered.
1781    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1782        of both had grown to the point where there were only 3 bits left.
1783        Fortunately, there was spare space in the data structure, and so I have
1784        moved the internal flags into a new 16-bit field to free up more option
1785        bits.
1787    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1788        but did not set the internal JCHANGED flag - either of these is enough to
1789        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1790        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1791        start sets both bits.
1793    4.  Added options (at build time, compile time, exec time) to change \R from
1794        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1796    5.  doc/pcresyntax.html was missing from the distribution.
1798    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1799        compatibility, even though it is no longer used.
1801    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1802        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1803        windows.h file is present (where different names are used). [This was
1804        reversed later after testing - see 16 below.]
1806    8.  Changed all #include <config.h> to #include "config.h". There were also
1807        some further <pcre.h> cases that I changed to "pcre.h".
1809    9.  When pcregrep was used with the --colour option, it missed the line ending
1810        sequence off the lines that it output.
1812    10. It was pointed out to me that arrays of string pointers cause lots of
1813        relocations when a shared library is dynamically loaded. A technique of
1814        using a single long string with a table of offsets can drastically reduce
1815        these. I have refactored PCRE in four places to do this. The result is
1816        dramatic:
1818          Originally:                          290
1819          After changing UCP table:            187
1820          After changing error message table:   43
1821          After changing table of "verbs"       36
1822          After changing table of Posix names   22
1824        Thanks to the folks working on Gregex for glib for this insight.
1826    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1827        unicode-properties was also set.
1829    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1831    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1832        checked only for CRLF.
1834    14. Added casts to pcretest.c to avoid compiler warnings.
1836    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1838    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1839        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1840        entirely. This removes changes made in 7 above.
1842    17. The CMake files have been updated, and there is now more information about
1843        building with CMake in the NON-UNIX-USE document.
1846    Version 7.3 28-Aug-07
1847    ---------------------
1849     1. In the rejigging of the build system that eventually resulted in 7.1, the
1850        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1851        brackets there is not right, since it causes compilers to look for an
1852        installed pcre.h, not the version that is in the source that is being
1853        compiled (which of course may be different). I have changed it back to:
1855          #include "pcre.h"
1857        I have a vague recollection that the change was concerned with compiling in
1858        different directories, but in the new build system, that is taken care of
1859        by the VPATH setting the Makefile.
1861     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1862        when the subject happened to end in the byte 0x85 (e.g. if the last
1863        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1864        characters but of course it shouldn't be taken as a newline when it is part
1865        of another character. The bug was that, for an unlimited repeat of . in
1866        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1867        characters when looking for a newline.
1869     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1871     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1872        in debug output.
1874     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1875        long printing in the pcrecpp unittest when running under MinGW.
1877     6. ESC_K was left out of the EBCDIC table.
1879     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1880        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1881        limit also applies to "virtual nesting" when a pattern is recursive, and in
1882        this case 1000 isn't so big. I have been able to remove this limit at the
1883        expense of backing off one optimization in certain circumstances. Normally,
1884        when pcre_exec() would call its internal match() function recursively and
1885        immediately return the result unconditionally, it uses a "tail recursion"
1886        feature to save stack. However, when a subpattern that can match an empty
1887        string has an unlimited repetition quantifier, it no longer makes this
1888        optimization. That gives it a stack frame in which to save the data for
1889        checking that an empty string has been matched. Previously this was taken
1890        from the 1000-entry workspace that had been reserved. So now there is no
1891        explicit limit, but more stack is used.
1893     8. Applied Daniel's patches to solve problems with the import/export magic
1894        syntax that is required for Windows, and which was going wrong for the
1895        pcreposix and pcrecpp parts of the library. These were overlooked when this
1896        problem was solved for the main library.
1898     9. There were some crude static tests to avoid integer overflow when computing
1899        the size of patterns that contain repeated groups with explicit upper
1900        limits. As the maximum quantifier is 65535, the maximum group length was
1901        set at 30,000 so that the product of these two numbers did not overflow a
1902        32-bit integer. However, it turns out that people want to use groups that
1903        are longer than 30,000 bytes (though not repeat them that many times).
1904        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1905        made it possible to implement the integer overflow checks in a much more
1906        dynamic way, which I have now done. The artificial limitation on group
1907        length has been removed - we now have only the limit on the total length of
1908        the compiled pattern, which depends on the LINK_SIZE setting.
1910    10. Fixed a bug in the documentation for get/copy named substring when
1911        duplicate names are permitted. If none of the named substrings are set, the
1912        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1913        empty string.
1915    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1916        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1917        because the ] is interpreted as the first data character and the
1918        terminating ] is not found. PCRE has been made compatible with Perl in this
1919        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1920        cause memory overwriting.
1922    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1923        string has been matched (to stop an infinite loop). It was not recognizing
1924        a conditional subpattern that could match an empty string if that
1925        subpattern was within another subpattern. For example, it looped when
1926        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1927        condition was not nested. This bug has been fixed.
1929    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1930        past the start of the subject in the presence of bytes with the top bit
1931        set, for example "\x8aBCD".
1933    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1934        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1936    14. Optimized (?!) to (*FAIL).
1938    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1939        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1940        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1941        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1942        does: it's just the validity check that is more restrictive.
1944    16. Inserted checks for integer overflows during escape sequence (backslash)
1945        processing, and also fixed erroneous offset values for syntax errors during
1946        backslash processing.
1948    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1949        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1951    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1952        caused an overrun.
1954    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1955        something other than just ASCII characters) inside a group that had an
1956        unlimited repeat caused a loop at compile time (while checking to see
1957        whether the group could match an empty string).
1959    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1960        [\P{Any}] did so. (Error in the code for printing property names.)
1962    21. An orphan \E inside a character class could cause a crash.
1964    22. A repeated capturing bracket such as (A)? could cause a wild memory
1965        reference during compilation.
1967    23. There are several functions in pcre_compile() that scan along a compiled
1968        expression for various reasons (e.g. to see if it's fixed length for look
1969        behind). There were bugs in these functions when a repeated \p or \P was
1970        present in the pattern. These operators have additional parameters compared
1971        with \d, etc, and these were not being taken into account when moving along
1972        the compiled data. Specifically:
1974        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1975            length.
1977        (b) An item such as \pL+ within a repeated group could cause crashes or
1978            loops.
1980        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1981            "reference to non-existent subpattern" error.
1983        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1985    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1986        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1988    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1990    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1991        character were causing crashes (broken optimization).
1993    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1994        \p or \P) caused a compile-time loop.
1996    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1997        break. For example, the unstudied pattern [\r\n]A does not match the string
1998        "\r\nA" because change 7.0/46 below moves the current point on by two
1999        characters after failing to match at the start. However, the pattern \nA
2000        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2001        the same is true. There doesn't seem any very clean way out of this, but
2002        what I have chosen to do makes the common cases work: PCRE now takes note
2003        of whether there can be an explicit match for \r or \n anywhere in the
2004        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2005        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2006        pattern has explicit CR or LF references.
2008    29. Added (*CR) etc for changing newline setting at start of pattern.
2011    Version 7.2 19-Jun-07
2012    ---------------------
2014     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
2015        which is apparently normally available under Windows.
2017     2. Re-jig the pcregrep tests with different newline settings in an attempt
2018        to make them independent of the local environment's newline setting.
2020     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
2022     4. Some of the "internals" tests were previously cut out when the link size
2023        was not 2, because the output contained actual offsets. The recent new
2024        "Z" feature of pcretest means that these can be cut out, making the tests
2025        usable with all link sizes.
2027     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
2028        stack recursion. This gives a massive performance boost under BSD, but just
2029        a small improvement under Linux. However, it saves one field in the frame
2030        in all cases.
2032     6. Added more features from the forthcoming Perl 5.10:
2034        (a) (?-n) (where n is a string of digits) is a relative subroutine or
2035            recursion call. It refers to the nth most recently opened parentheses.
2037        (b) (?+n) is also a relative subroutine call; it refers to the nth next
2038            to be opened parentheses.
2040        (c) Conditions that refer to capturing parentheses can be specified
2041            relatively, for example, (?(-2)... or (?(+3)...
2043        (d) \K resets the start of the current match so that everything before
2044            is not part of it.
2046        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2048        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2049            reference syntax.
2051        (g) (?| introduces a group in which the numbering of parentheses in each
2052            alternative starts with the same number.
2054        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2056     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2059     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2060        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2061        for detecting groups that can match an empty string.
2063     9. A pattern with a very large number of alternatives (more than several
2064        hundred) was running out of internal workspace during the pre-compile
2065        phase, where pcre_compile() figures out how much memory will be needed. A
2066        bit of new cunning has reduced the workspace needed for groups with
2067        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2068        workspace instead of running out of the 4096 that are available.
2070    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2072    11. Applied patch from Google to remove an optimization that didn't quite work.
2073        The report of the bug said:
2075          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2076          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2077          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2079    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2080        it matched the wrong number of bytes.
2083    Version 7.1 24-Apr-07
2084    ---------------------
2086     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
2087        that is more "standard", making use of automake and other Autotools. There
2088        is some re-arrangement of the files and adjustment of comments consequent
2089        on this.
2091     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
2092        for recursive directory scanning broke on some systems because the files
2093        are not scanned in any specific order and on different systems the order
2094        was different. A call to "sort" has been inserted into RunGrepTest for the
2095        approprate test as a short-term fix. In the longer term there may be an
2096        alternative.
2098     3. I had an email from Eric Raymond about problems translating some of PCRE's
2099        man pages to HTML (despite the fact that I distribute HTML pages, some
2100        people do their own conversions for various reasons). The problems
2101        concerned the use of low-level troff macros .br and .in. I have therefore
2102        removed all such uses from the man pages (some were redundant, some could
2103        be replaced by .nf/.fi pairs). The 132html script that I use to generate
2104        HTML has been updated to handle .nf/.fi and to complain if it encounters
2105        .br or .in.
2107     4. Updated comments in configure.ac that get placed in config.h.in and also
2108        arranged for config.h to be included in the distribution, with the name
2109        config.h.generic, for the benefit of those who have to compile without
2110        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
2112     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
2113        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
2114        makevp.bat for the current PCRE, using the additional files
2115        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
2117     6. A Windows user reported a minor discrepancy with test 2, which turned out
2118        to be caused by a trailing space on an input line that had got lost in his
2119        copy. The trailing space was an accident, so I've just removed it.
2121     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
2122        that is needed.
2124     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
2125        as "const" (a) because they are and (b) because it helps the PHP
2126        maintainers who have recently made a script to detect big data structures
2127        in the php code that should be moved to the .rodata section. I remembered
2128        to update Builducptable as well, so it won't revert if ucptable.h is ever
2129        re-created.
2131     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
2132        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
2133        order to be able to cut out the UTF-8 tables in the latter when UTF-8
2134        support is not required. This saves 1.5-2K of code, which is important in
2135        some applications.
2137        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
2138        so as not to refer to the tables, even though these functions will never be
2139        called when UTF-8 support is disabled. Otherwise there are problems with a
2140        shared library.
2142    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
2144        (a) It was defining its arguments as char * instead of void *.
2146        (b) It was assuming that all moves were upwards in memory; this was true
2147            a long time ago when I wrote it, but is no longer the case.
2149        The emulated memove() is provided for those environments that have neither
2150        memmove() nor bcopy(). I didn't think anyone used it these days, but that
2151        is clearly not the case, as these two bugs were recently reported.
2153    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
2154        and Detrail to create the HTML documentation, the .txt form of the man
2155        pages, and it removes trailing spaces from listed files. It also creates
2156        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
2157        case, it wraps all the #defines with #ifndefs. This script should be run
2158        before "make dist".
2160    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
2161        with Unicode property support.
2163        (a) For a maximizing quantifier, if the two different cases of the
2164            character were of different lengths in their UTF-8 codings (there are
2165            some cases like this - I found 11), and the matching function had to
2166            back up over a mixture of the two cases, it incorrectly assumed they
2167            were both the same length.
2169        (b) When PCRE was configured to use the heap rather than the stack for
2170            recursion during matching, it was not correctly preserving the data for
2171            the other case of a UTF-8 character when checking ahead for a match
2172            while processing a minimizing repeat. If the check also involved
2173            matching a wide character, but failed, corruption could cause an
2174            erroneous result when trying to check for a repeat of the original
2175            character.
2177    13. Some tidying changes to the testing mechanism:
2179        (a) The RunTest script now detects the internal link size and whether there
2180            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
2181            values substituted by "configure". (The RunGrepTest script already did
2182            this for UTF-8.) The configure.ac script no longer substitutes the
2183            relevant variables.
2185        (b) The debugging options /B and /D in pcretest show the compiled bytecode
2186            with length and offset values. This means that the output is different
2187            for different internal link sizes. Test 2 is skipped for link sizes
2188            other than 2 because of this, bypassing the problem. Unfortunately,
2189            there was also a test in test 3 (the locale tests) that used /B and
2190            failed for link sizes other than 2. Rather than cut the whole test out,
2191            I have added a new /Z option to pcretest that replaces the length and
2192            offset values with spaces. This is now used to make test 3 independent
2193            of link size. (Test 2 will be tidied up later.)
2195    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
2196        segmentation fault instead of returning the appropriate error message.
2198    15. In multiline mode when the newline sequence was set to "any", the pattern
2199        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
2200        This doesn't seem right; it now treats the CRLF combination as the line
2201        ending, and so does not match in that case. It's only a pattern such as ^$
2202        that would hit this one: something like ^ABC$ would have failed after \r
2203        and then tried again after \r\n.
2205    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
2206        in an attempt to make files that differ only in their line terminators
2207        compare equal. This works on Linux.
2209    17. Under certain error circumstances pcregrep might try to free random memory
2210        as it exited. This is now fixed, thanks to valgrind.
2212    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
2213        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
2214        was because its rules for how to advance for /g after matching an empty
2215        string at the end of a line did not allow for this case. They now check for
2216        it specially.
2218    20. pcretest is supposed to handle patterns and data of any length, by
2219        extending its buffers when necessary. It was getting this wrong when the
2220        buffer for a data line had to be extended.
2222    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
2223        CRLF as a newline sequence.
2225    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
2226        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
2227        I have nevertheless tidied it up.
2229    23. Added some casts to kill warnings from HP-UX ia64 compiler.
2231    24. Added a man page for pcre-config.
2234    Version 7.0 19-Dec-06
2235    ---------------------
2237     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
2238        moving to gcc 4.1.1.
2240     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
2241        sys/time.h, which is documented as needed for this function. It doesn't
2242        seem to matter on Linux, but it showed up on some releases of OS X.
2244     3. It seems that there are systems where bytes whose values are greater than
2245        127 match isprint() in the "C" locale. The "C" locale should be the
2246        default when a C program starts up. In most systems, only ASCII printing
2247        characters match isprint(). This difference caused the output from pcretest
2248        to vary, making some of the tests fail. I have changed pcretest so that:
2250        (a) When it is outputting text in the compiled version of a pattern, bytes
2251            other than 32-126 are always shown as hex escapes.
2253        (b) When it is outputting text that is a matched part of a subject string,
2254            it does the same, unless a different locale has been set for the match
2255            (using the /L modifier). In this case, it uses isprint() to decide.
2257     4. Fixed a major bug that caused incorrect computation of the amount of memory
2258        required for a compiled pattern when options that changed within the
2259        pattern affected the logic of the preliminary scan that determines the
2260        length. The relevant options are -x, and -i in UTF-8 mode. The result was
2261        that the computed length was too small. The symptoms of this bug were
2262        either the PCRE error "internal error: code overflow" from pcre_compile(),
2263        or a glibc crash with a message such as "pcretest: free(): invalid next
2264        size (fast)". Examples of patterns that provoked this bug (shown in
2265        pcretest format) are:
2267          /(?-x: )/x
2268          /(?x)(?-x: \s*#\s*)/
2269          /((?i)[\x{c0}])/8
2270          /(?i:[\x{c0}])/8
2272        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
2273        is now done differently.
2275     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
2276        wrapper classes; (b) implement a new function in the C++ scanner that is
2277        more efficient than the old way of doing things because it avoids levels of
2278        recursion in the regex matching; (c) add a paragraph to the documentation
2279        for the FullMatch() function.
2281     6. The escape sequence \n was being treated as whatever was defined as
2282        "newline". Not only was this contrary to the documentation, which states
2283        that \n is character 10 (hex 0A), but it also went horribly wrong when
2284        "newline" was defined as CRLF. This has been fixed.
2286     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
2287        was being set to -1 for the "end of line" case (supposedly a value that no
2288        character can have). Though this value is never used (the check for end of
2289        line is "zero bytes in current character"), it caused compiler complaints.
2290        I've changed it to 0xffffffff.
2292     8. In pcre_version.c, the version string was being built by a sequence of
2293        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
2294        string (as it is for production releases) called a macro with an empty
2295        argument. The C standard says the result of this is undefined. The gcc
2296        compiler treats it as an empty string (which was what was wanted) but it is
2297        reported that Visual C gives an error. The source has been hacked around to
2298        avoid this problem.
2300     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
2301        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
2302        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
2303        of them did).
2305    10. Originally, pcretest opened its input and output without "b"; then I was
2306        told that "b" was needed in some environments, so it was added for release
2307        5.0 to both the input and output. (It makes no difference on Unix-like
2308        systems.) Later I was told that it is wrong for the input on Windows. I've
2309        now abstracted the modes into two macros, to make it easier to fiddle with
2310        them, and removed "b" from the input mode under Windows.
2312    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
2314    12. Added -help and --help to pcretest as an official way of being reminded
2315        of the options.
2317    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
2318        and pcrecpp.cc because they annoy compilers at high warning levels.
2320    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
2322    15. Fixed an occurrence of == in configure.ac that should have been = (shell
2323        scripts are not C programs :-) and which was not noticed because it works
2324        on Linux.
2326    16. pcretest is supposed to handle any length of pattern and data line (as one
2327        line or as a continued sequence of lines) by extending its input buffer if
2328        necessary. This feature was broken for very long pattern lines, leading to
2329        a string of junk being passed to pcre_compile() if the pattern was longer
2330        than about 50K.
2332    17. I have done a major re-factoring of the way pcre_compile() computes the
2333        amount of memory needed for a compiled pattern. Previously, there was code
2334        that made a preliminary scan of the pattern in order to do this. That was
2335        OK when PCRE was new, but as the facilities have expanded, it has become
2336        harder and harder to keep it in step with the real compile phase, and there
2337        have been a number of bugs (see for example, 4 above). I have now found a
2338        cunning way of running the real compile function in a "fake" mode that
2339        enables it to compute how much memory it would need, while actually only
2340        ever using a few hundred bytes of working memory and without too many
2341        tests of the mode. This should make future maintenance and development
2342        easier. A side effect of this work is that the limit of 200 on the nesting
2343        depth of parentheses has been removed (though this was never a serious
2344        limitation, I suspect). However, there is a downside: pcre_compile() now
2345        runs more slowly than before (30% or more, depending on the pattern). I
2346        hope this isn't a big issue. There is no effect on runtime performance.
2348    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
2349        newline (only possible for the last line of a file) and it was a
2350        pattern that set a locale (followed by /Lsomething), pcretest crashed.
2352    19. Added additional timing features to pcretest. (1) The -tm option now times
2353        matching only, not compiling. (2) Both -t and -tm can be followed, as a
2354        separate command line item, by a number that specifies the number of
2355        repeats to use when timing. The default is 50000; this gives better
2356        precision, but takes uncomfortably long for very large patterns.
2358    20. Extended pcre_study() to be more clever in cases where a branch of a
2359        subpattern has no definite first character. For example, (a*|b*)[cd] would
2360        previously give no result from pcre_study(). Now it recognizes that the
2361        first character must be a, b, c, or d.
2363    21. There was an incorrect error "recursive call could loop indefinitely" if
2364        a subpattern (or the entire pattern) that was being tested for matching an
2365        empty string contained only one non-empty item after a nested subpattern.
2366        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
2367        incorrectly, because the \d was being skipped in the check.
2369    22. The pcretest program now has a new pattern option /B and a command line
2370        option -b, which is equivalent to adding /B to every pattern. This causes
2371        it to show the compiled bytecode, without the additional information that
2372        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
2373        is the same as /B/I).
2375    23. A new optimization is now able automatically to treat some sequences such
2376        as a*b as a*+b. More specifically, if something simple (such as a character
2377        or a simple class like \d) has an unlimited quantifier, and is followed by
2378        something that cannot possibly match the quantified thing, the quantifier
2379        is automatically "possessified".
2381    24. A recursive reference to a subpattern whose number was greater than 39
2382        went wrong under certain circumstances in UTF-8 mode. This bug could also
2383        have affected the operation of pcre_study().
2385    25. Realized that a little bit of performance could be had by replacing
2386        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
2388    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
2390    27. Possessive quantifiers such as a++ were previously implemented by turning
2391        them into atomic groups such as ($>a+). Now they have their own opcodes,
2392        which improves performance. This includes the automatically created ones
2393        from 23 above.
2395    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
2396        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
2397        the first matched character to be a colon. This applied both to named and
2398        numbered groups.
2400    29. The ucpinternal.h header file was missing its idempotency #ifdef.
2402    30. I was sent a "project" file called libpcre.a.dev which I understand makes
2403        building PCRE on Windows easier, so I have included it in the distribution.
2405    31. There is now a check in pcretest against a ridiculously large number being
2406        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
2407        loop, the loop is abandoned.
2409    32. Forward references to subpatterns in conditions such as (?(2)...) where
2410        subpattern 2 is defined later cause pcre_compile() to search forwards in
2411        the pattern for the relevant set of parentheses. This search went wrong
2412        when there were unescaped parentheses in a character class, parentheses
2413        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2415    33. "Subroutine" calls and backreferences were previously restricted to
2416        referencing subpatterns earlier in the regex. This restriction has now
2417        been removed.
2419    34. Added a number of extra features that are going to be in Perl 5.10. On the
2420        whole, these are just syntactic alternatives for features that PCRE had
2421        previously implemented using the Python syntax or my own invention. The
2422        other formats are all retained for compatibility.
2424        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2425            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2426            also .NET compatible.
2428        (b) A recursion or subroutine call to a named group can now be defined as
2429            (?&name) as well as (?P>name).
2431        (c) A backreference to a named group can now be defined as \k<name> or
2432            \k'name' as well as (?P=name). The new forms, as well as being in Perl
2433            5.10, are also .NET compatible.
2435        (d) A conditional reference to a named group can now use the syntax
2436            (?(<name>) or (?('name') as well as (?(name).
2438        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2439            groups (named and numbered) that are never evaluated inline, but can be
2440            called as "subroutines" from elsewhere. In effect, the DEFINE condition
2441            is always false. There may be only one alternative in such a group.
2443        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2444            as the simple (?(R). The condition is true only if the most recent
2445            recursion is that of the given number or name. It does not search out
2446            through the entire recursion stack.
2448        (g) The escape \gN or \g{N} has been added, where N is a positive or
2449            negative number, specifying an absolute or relative reference.
2451    35. Tidied to get rid of some further signed/unsigned compiler warnings and
2452        some "unreachable code" warnings.
2454    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2455        things, this adds five new scripts.
2457    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2458        There were also incompatibilities regarding the handling of \Q..\E inside
2459        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2460        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2462    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2463        matches an empty string, and forcibly breaks the loop. There were bugs in
2464        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2465        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2466        separate and independent bugs (that affected different cases) have been
2467        fixed.
2469    39. Refactored the code to abolish the use of different opcodes for small
2470        capturing bracket numbers. This is a tidy that I avoided doing when I
2471        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2472        The new approach is not only tidier, it makes it possible to reduce the
2473        memory needed to fix the previous bug (38).
2475    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2476        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2477        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2478        mode.
2480    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2481        report.
2483    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2484        copy construction and assignment in the C++ wrapper.
2486    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2487        couple of bugs that could have given wrong results in the "--newline=crlf"
2488        case.
2490    44. Added a number of casts and did some reorganization of signed/unsigned int
2491        variables following suggestions from Dair Grant. Also renamed the variable
2492        "this" as "item" because it is a C++ keyword.
2494    45. Arranged for dftables to add
2496          #include "pcre_internal.h"
2498        to pcre_chartables.c because without it, gcc 4.x may remove the array
2499        definition from the final binary if PCRE is built into a static library and
2500        dead code stripping is activated.
2502    46. For an unanchored pattern, if a match attempt fails at the start of a
2503        newline sequence, and the newline setting is CRLF or ANY, and the next two
2504        characters are CRLF, advance by two characters instead of one.
2507    Version 6.7 04-Jul-06
2508    ---------------------
2510     1. In order to handle tests when input lines are enormously long, pcretest has
2511        been re-factored so that it automatically extends its buffers when
2512        necessary. The code is crude, but this _is_ just a test program. The
2513        default size has been increased from 32K to 50K.
2515     2. The code in pcre_study() was using the value of the re argument before
2516        testing it for NULL. (Of course, in any sensible call of the function, it
2517        won't be NULL.)
2519     3. The memmove() emulation function in pcre_internal.h, which is used on
2520        systems that lack both memmove() and bcopy() - that is, hardly ever -
2521        was missing a "static" storage class specifier.
2523     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2524        containing an extended class (one that cannot be represented by a bitmap
2525        because it contains high-valued characters or Unicode property items, e.g.
2526        [\pZ]). Almost always one would set UTF-8 mode when processing such a
2527        pattern, but PCRE should not loop if you do not (it no longer does).
2528        [Detail: two cases were found: (a) a repeated subpattern containing an
2529        extended class; (b) a recursive reference to a subpattern that followed a
2530        previous extended class. It wasn't skipping over the extended class
2531        correctly when UTF-8 mode was not set.]
2533     5. A negated single-character class was not being recognized as fixed-length
2534        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2535        compile error "lookbehind assertion is not fixed length".
2537     6. The RunPerlTest auxiliary script was showing an unexpected difference
2538        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2539        write a Perl script that can interpret lines of an input file either as
2540        byte characters or as UTF-8, which is what "perltest" was being required to
2541        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2542        can't do is switch easily at run time between having the "use utf8;" pragma
2543        or not. In the end, I fudged it by using the RunPerlTest script to insert
2544        "use utf8;" explicitly for the UTF-8 tests.
2546     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2547        the end of the subject string, contrary to the documentation and to what
2548        Perl does. This was true of both matching functions. Now it matches only at
2549        the start of the subject and immediately after *internal* newlines.
2551     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2552        a pointer to an int instead of a pointer to an unsigned long int. This
2553        caused problems on 64-bit systems.
2555     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2556        instance of the 'standard' template library not being so standard".
2558    10. There was no check on the number of named subpatterns nor the maximum
2559        length of a subpattern name. The product of these values is used to compute
2560        the size of the memory block for a compiled pattern. By supplying a very
2561        long subpattern name and a large number of named subpatterns, the size
2562        computation could be caused to overflow. This is now prevented by limiting
2563        the length of names to 32 characters, and the number of named subpatterns
2564        to 10,000.
2566    11. Subpatterns that are repeated with specific counts have to be replicated in
2567        the compiled pattern. The size of memory for this was computed from the
2568        length of the subpattern and the repeat count. The latter is limited to
2569        65535, but there was no limit on the former, meaning that integer overflow
2570        could in principle occur. The compiled length of a repeated subpattern is
2571        now limited to 30,000 bytes in order to prevent this.
2573    12. Added the optional facility to have named substrings with the same name.
2575    13. Added the ability to use a named substring as a condition, using the
2576        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2577        are numbers (not recommended). Forward references are permitted.
2579    14. Added forward references in named backreferences (if you see what I mean).
2581    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2582        pattern could run off the end of the subject. For example, the pattern
2583        "(?s)(.{1,5})"8 did this with the subject "ab".
2585    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2586        PCRE_CASELESS was set when matching characters that were quantified with ?
2587        or *.
2589    17. A character class other than a single negated character that had a minimum
2590        but no maximum quantifier - for example [ab]{6,} - was not handled
2591        correctly by pce_dfa_exec(). It would match only one character.
2593    18. A valid (though odd) pattern that looked like a POSIX character
2594        class but used an invalid character after [ (for example [[,abc,]]) caused
2595        pcre_compile() to give the error "Failed: internal error: code overflow" or
2596        in some cases to crash with a glibc free() error. This could even happen if
2597        the pattern terminated after [[ but there just happened to be a sequence of
2598        letters, a binary zero, and a closing ] in the memory that followed.
2600    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2601        over the years. Originally (before any Unicode support), just the bottom 8
2602        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2603        output from "man perlunicode" includes this:
2605          The regular expression compiler produces polymorphic opcodes.  That
2606          is, the pattern adapts to the data and automatically switches to
2607          the Unicode character scheme when presented with Unicode data--or
2608          instead uses a traditional byte scheme when presented with byte
2609          data.
2611        Sadly, a wide octal escape does not cause a switch, and in a string with
2612        no other multibyte characters, these octal escapes are treated as before.
2613        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2614        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2615        Unicode string.
2617        I have not perpetrated such confusion in PCRE. Up till now, it took just
2618        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2619        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2620        translate to the appropriate multibyte character.
2622    29. Applied some refactoring to reduce the number of warnings from Microsoft
2623        and Borland compilers. This has included removing the fudge introduced
2624        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2625        a warning about an unused variable.
2627    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2628        characters since release 4.0, because Perl (from release 5.004) does not.
2629        [Or at least, is documented not to: some releases seem to be in conflict
2630        with the documentation.] However, when a pattern was studied with
2631        pcre_study() and all its branches started with \s, PCRE still included VT
2632        as a possible starting character. Of course, this did no harm; it just
2633        caused an unnecessary match attempt.
2635    22. Removed a now-redundant internal flag bit that recorded the fact that case
2636        dependency changed within the pattern. This was once needed for "required
2637        byte" processing, but is no longer used. This recovers a now-scarce options
2638        bit. Also moved the least significant internal flag bit to the most-
2639        significant bit of the word, which was not previously used (hangover from
2640        the days when it was an int rather than a uint) to free up another bit for
2641        the future.
2643    23. Added support for CRLF line endings as well as CR and LF. As well as the
2644        default being selectable at build time, it can now be changed at runtime
2645        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2646        specify that it is scanning data with non-default line endings.
2648    24. Changed the definition of CXXLINK to make it agree with the definition of
2649        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2651    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2652        recursions. This makes a big different to stack usage for some patterns.
2654    26. If a subpattern containing a named recursion or subroutine reference such
2655        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2656        the space required for the compiled pattern went wrong and gave too small a
2657        value. Depending on the environment, this could lead to "Failed: internal
2658        error: code overflow at offset 49" or "glibc detected double free or
2659        corruption" errors.
2661    27. Applied patches from Google (a) to support the new newline modes and (b) to
2662        advance over multibyte UTF-8 characters in GlobalReplace.
2664    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2665        difference for some implementation of PCRE in some Windows version.
2667    29. Added some extra testing facilities to pcretest:
2669        \q<number>   in a data line sets the "match limit" value
2670        \Q<number>   in a data line sets the "match recursion limt" value
2671        -S <number>  sets the stack size, where <number> is in megabytes
2673        The -S option isn't available for Windows.
2676    Version 6.6 06-Feb-06
2677    ---------------------
2679     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2680        in pcreposix.h. I have copied the definition from pcre.h.
2682     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2683        because pcre.h is no longer a built file.
2685     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2686        not normally included in the compiled code.
2689    Version 6.5 01-Feb-06
2690    ---------------------
2692     1. When using the partial match feature with pcre_dfa_exec(), it was not
2693        anchoring the second and subsequent partial matches at the new starting
2694        point. This could lead to incorrect results. For example, with the pattern
2695        /1234/, partially matching against "123" and then "a4" gave a match.
2697     2. Changes to pcregrep:
2699        (a) All non-match returns from pcre_exec() were being treated as failures
2700            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2701            error message is output. Some extra information is given for the
2702            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2703            probably the only errors that are likely to be caused by users (by
2704            specifying a regex that has nested indefinite repeats, for instance).
2705            If there are more than 20 of these errors, pcregrep is abandoned.
2707        (b) A binary zero was treated as data while matching, but terminated the
2708            output line if it was written out. This has been fixed: binary zeroes
2709            are now no different to any other data bytes.
2711        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2712            used to set a locale for matching. The --locale=xxxx long option has
2713            been added (no short equivalent) to specify a locale explicitly on the
2714            pcregrep command, overriding the environment variables.
2716        (d) When -B was used with -n, some line numbers in the output were one less
2717            than they should have been.
2719        (e) Added the -o (--only-matching) option.
2721        (f) If -A or -C was used with -c (count only), some lines of context were
2722            accidentally printed for the final match.
2724        (g) Added the -H (--with-filename) option.
2726        (h) The combination of options -rh failed to suppress file names for files
2727            that were found from directory arguments.
2729        (i) Added the -D (--devices) and -d (--directories) options.
2731        (j) Added the -F (--fixed-strings) option.
2733        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2735        (l) Added the --colo(u)r option.
2737        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2738            is not present by default.
2740     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2741        items such as (?R) or (?1), when the recursion could match a number of
2742        alternatives. If it matched one of the alternatives, but subsequently,
2743        outside the recursion, there was a failure, the code tried to back up into
2744        the recursion. However, because of the way PCRE is implemented, this is not
2745        possible, and the result was an incorrect result from the match.
2747        In order to prevent this happening, the specification of recursion has
2748        been changed so that all such subpatterns are automatically treated as
2749        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2751     4. I had overlooked the fact that, in some locales, there are characters for
2752        which isalpha() is true but neither isupper() nor islower() are true. In
2753        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2754        and ordfeminine) are like this. This affected the treatment of \w and \W
2755        when they appeared in character classes, but not when they appeared outside
2756        a character class. The bit map for "word" characters is now created
2757        separately from the results of isalnum() instead of just taking it from the
2758        upper, lower, and digit maps. (Plus the underscore character, of course.)
2760     5. The above bug also affected the handling of POSIX character classes such as
2761        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2762        permanent tables. Instead, the bit maps for such a class were previously
2763        created as the appropriate unions of the upper, lower, and digit bitmaps.
2764        Now they are created by subtraction from the [[:word:]] class, which has
2765        its own bitmap.
2767     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2768        It is created by subtracting the vertical space characters (\x09, \x0a,
2769        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2770        subtraction was done in the overall bitmap for a character class, meaning
2771        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2772        be recognized. This bug has been fixed.
2774     7. Patches from the folks at Google:
2776          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2777          real life, but is still worth protecting against".
2779          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2780          regular expressions".
2782          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2783          have it.
2785          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2786          "configure" and the latter not, in order to fix a problem somebody had
2787          with compiling the Arg class on HP-UX.
2789          (e) Improve the error-handling of the C++ wrapper a little bit.
2791          (f) New tests for checking recursion limiting.
2793     8. The pcre_memmove() function, which is used only if the environment does not
2794        have a standard memmove() function (and is therefore rarely compiled),
2795        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2796        returning a result (though PCRE never actually uses the result).
2798     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2799        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2800        returned instead of calling malloc() with an overflowing number that would
2801        most likely cause subsequent chaos.
2803    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2805    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2806        with this option is matched, the nmatch and pmatch options of regexec() are
2807        ignored.
2809    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2810        provided in case anyone wants to the the POSIX interface with UTF-8
2811        strings.
2813    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2814        C++ linking (needed for some HP-UX environments).
2816    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2817        (unused parameter) and in the pcre_printint() function (omitted "default"
2818        switch label when the default is to do nothing).
2820    15. Added some code to make it possible, when PCRE is compiled as a C++
2821        library, to replace subject pointers for pcre_exec() with a smart pointer
2822        class, thus making it possible to process discontinuous strings.
2824    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2825        much the same function. They were added by different people who were trying
2826        to make PCRE easy to compile on non-Unix systems. It has been suggested
2827        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2828        for compiling on Windows systems. I have therefore replaced it with
2829        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2830        defaults to "extern" for C or "extern C" for C++, which works fine on
2831        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2832        SCOPE with something explicit in config.h. In addition:
2834        (a) pcreposix.h still had just "extern" instead of either of these macros;
2835            I have replaced it with PCRE_DATA_SCOPE.
2837        (b) Functions such as _pcre_xclass(), which are internal to the library,
2838            but external in the C sense, all had PCRE_EXPORT in their definitions.
2839            This is apparently wrong for the Windows case, so I have removed it.
2840            (It makes no difference on Unix-like systems.)
2842    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2843        of recursive calls to match(). This is different to MATCH_LIMIT because
2844        that limits the total number of calls to match(), not all of which increase
2845        the depth of recursion. Limiting the recursion depth limits the amount of
2846        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2847        when PCRE is compiled, and changed at run time. A patch from Google adds
2848        this functionality to the C++ interface.
2850    18. Changes to the handling of Unicode character properties:
2852        (a) Updated the table to Unicode 4.1.0.
2854        (b) Recognize characters that are not in the table as "Cn" (undefined).
2856        (c) I revised the way the table is implemented to a much improved format
2857            which includes recognition of ranges. It now supports the ranges that
2858            are defined in UnicodeData.txt, and it also amalgamates other
2859            characters into ranges. This has reduced the number of entries in the
2860            table from around 16,000 to around 3,000, thus reducing its size
2861            considerably. I realized I did not need to use a tree structure after
2862            all - a binary chop search is just as efficient. Having reduced the
2863            number of entries, I extended their size from 6 bytes to 8 bytes to
2864            allow for more data.
2866        (d) Added support for Unicode script names via properties such as \p{Han}.
2868    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2869        matching that character.
2871    20. When matching a repeated Unicode property with a minimum greater than zero,
2872        (for example \pL{2,}), PCRE could look past the end of the subject if it
2873        reached it while seeking the minimum number of characters. This could
2874        happen only if some of the characters were more than one byte long, because
2875        there is a check for at least the minimum number of bytes.
2877    21. Refactored the implementation of \p and \P so as to be more general, to
2878        allow for more different types of property in future. This has changed the
2879        compiled form incompatibly. Anybody with saved compiled patterns that use
2880        \p or \P will have to recompile them.
2882    22. Added "Any" and "L&" to the supported property types.
2884    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2885        but give a compile time error if the value is greater than 0xff.
2887    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2888        accidentally not being installed or uninstalled.
2890    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2891        made were to insert the current release number. This seemed silly, because
2892        it made things harder for people building PCRE on systems that don't run
2893        "configure". I have turned pcre.h into a distributed file, no longer built
2894        by "configure", with the version identification directly included. There is
2895        no longer a pcre.h.in file.
2897        However, this change necessitated a change to the pcre-config script as
2898        well. It is built from pcre-config.in, and one of the substitutions was the
2899        release number. I have updated configure.ac so that ./configure now finds
2900        the release number by grepping pcre.h.
2902    26. Added the ability to run the tests under valgrind.
2905    Version 6.4 05-Sep-05
2906    ---------------------
2908     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2909        "--" to be printed when multiple files were scanned, even when none of the
2910        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2911        consider it to be a bug, and have restored the previous behaviour.
2913     2. A couple of code tidies to get rid of compiler warnings.
2915     3. The pcretest program used to cheat by referring to symbols in the library
2916        whose names begin with _pcre_. These are internal symbols that are not
2917        really supposed to be visible externally, and in some environments it is
2918        possible to suppress them. The cheating is now confined to including
2919        certain files from the library's source, which is a bit cleaner.
2921     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2922        file's purpose clearer.
2924     5. Reorganized pcre_ucp_findchar().
2927    Version 6.3 15-Aug-05
2928    ---------------------
2930     1. The file libpcre.pc.in did not have general read permission in the tarball.
2932     2. There were some problems when building without C++ support:
2934        (a) If C++ support was not built, "make install" and "make test" still
2935            tried to test it.
2937        (b) There were problems when the value of CXX was explicitly set. Some
2938            changes have been made to try to fix these, and ...
2940        (c) --disable-cpp can now be used to explicitly disable C++ support.
2942        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2943            backslash in a target when C++ was disabled. This confuses some
2944            versions of "make", apparently. Using an intermediate variable solves
2945            this. (Same for CPP_LOBJ.)
2947     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2948        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2949        necessary on certain architectures.
2951     4. Added a setting of -export-symbols-regex to the link command to remove
2952        those symbols that are exported in the C sense, but actually are local
2953        within the library, and not documented. Their names all begin with
2954        "_pcre_". This is not a perfect job, because (a) we have to except some
2955        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2956        available (and never for static libraries). I have made a note to try to
2957        find a way round (a) in the future.
2960    Version 6.2 01-Aug-05
2961    ---------------------
2963     1. There was no test for integer overflow of quantifier values. A construction
2964        such as {1111111111111111} would give undefined results. What is worse, if
2965        a minimum quantifier for a parenthesized subpattern overflowed and became
2966        negative, the calculation of the memory size went wrong. This could have
2967        led to memory overwriting.
2969     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2971     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2972        operating environments where this matters.
2974     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2975        PCRE options from within the C++ wrapper.
2977     5. Named capturing subpatterns were not being correctly counted when a pattern
2978        was compiled. This caused two problems: (a) If there were more than 100
2979        such subpatterns, the calculation of the memory needed for the whole
2980        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2981        back references of the form \12, where the number was greater than 9, were
2982        not recognized as back references, even though there were sufficient
2983        previous subpatterns.
2985     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2986        versions of gcc, e.g. 2.95.4.
2989    Version 6.1 21-Jun-05
2990    ---------------------
2992     1. There was one reference to the variable "posix" in pcretest.c that was not
2993        surrounded by "#if !defined NOPOSIX".
2995     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2996        the cross-check on the old pcre_info() function, for the benefit of the
2997        cut-down version of PCRE that is currently imported into Exim.
2999     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
3000        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
3001        patterns that start (?i) but I don't think that matters. The use of (?i) is
3002        just an example; this all applies to the other options as well.
3004     4. Since libtool seems to echo the compile commands it is issuing, the output
3005        from "make" can be reduced a bit by putting "@" in front of each libtool
3006        compile command.
3008     5. Patch from the folks at Google for configure.in to be a bit more thorough
3009        in checking for a suitable C++ installation before trying to compile the
3010        C++ stuff. This should fix a reported problem when a compiler was present,
3011        but no suitable headers.
3013     6. The man pages all had just "PCRE" as their title. I have changed them to
3014        be the relevant file name. I have also arranged that these names are
3015        retained in the file doc/pcre.txt, which is a concatenation in text format
3016        of all the man pages except the little individual ones for each function.
3018     7. The NON-UNIX-USE file had not been updated for the different set of source
3019        files that come with release 6. I also added a few comments about the C++
3020        wrapper.
3023    Version 6.0 07-Jun-05
3024    ---------------------
3026     1. Some minor internal re-organization to help with my DFA experiments.
3028     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
3029        didn't matter for the library itself when fully configured, but did matter
3030        when compiling without UCP support, or within Exim, where the ucp files are
3031        not imported.
3033     3. Refactoring of the library code to split up the various functions into
3034        different source modules. The addition of the new DFA matching code (see
3035        below) to a single monolithic source would have made it really too
3036        unwieldy, quite apart from causing all the code to be include in a
3037        statically linked application, when only some functions are used. This is
3038        relevant even without the DFA addition now that patterns can be compiled in
3039        one application and matched in another.
3041        The downside of splitting up is that there have to be some external
3042        functions and data tables that are used internally in different modules of
3043        the library but which are not part of the API. These have all had their
3044        names changed to start with "_pcre_" so that they are unlikely to clash
3045        with other external names.
3047     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
3048        a different (DFA) algorithm. Although it is slower than the original
3049        function, it does have some advantages for certain types of matching
3050        problem.
3052     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
3053        including restarting after a partial match.
3055     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
3056        defined when compiling for Windows was sent to me. I have put it into the
3057        code, though I have no means of testing or verifying it.
3059     7. Added the pcre_refcount() auxiliary function.
3061     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
3062        match before or at the first newline in the subject string. In pcretest,
3063        the /f option on a pattern can be used to set this.
3065     9. A repeated \w when used in UTF-8 mode with characters greater than 256
3066        would behave wrongly. This has been present in PCRE since release 4.0.
3068    10. A number of changes to the pcregrep command:
3070        (a) Refactored how -x works; insert ^(...)$ instead of setting
3071            PCRE_ANCHORED and checking the length, in preparation for adding
3072            something similar for -w.
3074        (b) Added the -w (match as a word) option.
3076        (c) Refactored the way lines are read and buffered so as to have more
3077            than one at a time available.
3079        (d) Implemented a pcregrep test script.
3081        (e) Added the -M (multiline match) option. This allows patterns to match
3082            over several lines of the subject. The buffering ensures that at least
3083            8K, or the rest of the document (whichever is the shorter) is available
3084            for matching (and similarly the previous 8K for lookbehind assertions).
3086        (f) Changed the --help output so that it now says
3088              -w, --word-regex(p)
3090            instead of two lines, one with "regex" and the other with "regexp"
3091            because that confused at least one person since the short forms are the
3092            same. (This required a bit of code, as the output is generated
3093            automatically from a table. It wasn't just a text change.)
3095        (g) -- can be used to terminate pcregrep options if the next thing isn't an
3096            option but starts with a hyphen. Could be a pattern or a path name
3097            starting with a hyphen, for instance.
3099        (h) "-" can be given as a file name to represent stdin.
3101        (i) When file names are being printed, "(standard input)" is used for
3102            the standard input, for compatibility with GNU grep. Previously
3103            "<stdin>" was used.
3105        (j) The option --label=xxx can be used to supply a name to be used for
3106            stdin when file names are being printed. There is no short form.
3108        (k) Re-factored the options decoding logic because we are going to add
3109            two more options that take data. Such options can now be given in four
3110            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
3112        (l) Added the -A, -B, and -C options for requesting that lines of context
3113            around matches be printed.
3115        (m) Added the -L option to print the names of files that do not contain
3116            any matching lines, that is, the complement of -l.
3118        (n) The return code is 2 if any file cannot be opened, but pcregrep does
3119            continue to scan other files.
3121        (o) The -s option was incorrectly implemented. For compatibility with other
3122            greps, it now suppresses the error message for a non-existent or non-
3123            accessible file (but not the return code). There is a new option called
3124            -q that suppresses the output of matching lines, which was what -s was
3125            previously doing.
3127        (p) Added --include and --exclude options to specify files for inclusion
3128            and exclusion when recursing.
3130    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
3131        Hopefully, it now does.
3133    12. Missing cast in pcre_study().
3135    13. Added an "uninstall" target to the makefile.
3137    14. Replaced "extern" in the function prototypes in Makefile.in with
3138        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
3139        world, but is set differently for Windows.
3141    15. Added a second compiling function called pcre_compile2(). The only
3142        difference is that it has an extra argument, which is a pointer to an
3143        integer error code. When there is a compile-time failure, this is set
3144        non-zero, in addition to the error test pointer being set to point to an
3145        error message. The new argument may be NULL if no error number is required
3146        (but then you may as well call pcre_compile(), which is now just a
3147        wrapper). This facility is provided because some applications need a
3148        numeric error indication, but it has also enabled me to tidy up the way
3149        compile-time errors are handled in the POSIX wrapper.
3151    16. Added VPATH=.libs to the makefile; this should help when building with one
3152        prefix path and installing with another. (Or so I'm told by someone who
3153        knows more about this stuff than I do.)
3155    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
3156        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
3157        match everything, including newlines. This is not POSIX-compatible, but
3158        somebody wanted the feature. From pcretest it can be activated by using
3159        both the P and the s flags.
3161    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
3163    19. libpcre.pc was being incorrectly installed as executable.
3165    20. A couple of places in pcretest check for end-of-line by looking for '\n';
3166        it now also looks for '\r' so that it will work unmodified on Windows.
3168    21. Added Google's contributed C++ wrapper to the distribution.
3170    22. Added some untidy missing memory free() calls in pcretest, to keep
3171        Electric Fence happy when testing.
3175    Version 5.0 13-Sep-04
3176    ---------------------
3178     1. Internal change: literal characters are no longer packed up into items
3179        containing multiple characters in a single byte-string. Each character
3180        is now matched using a separate opcode. However, there may be more than one
3181        byte in the character in UTF-8 mode.
3183     2. The pcre_callout_block structure has two new fields: pattern_position and
3184        next_item_length. These contain the offset in the pattern to the next match
3185        item, and its length, respectively.
3187     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
3188        insertion of callouts before each pattern item. Added the /C option to
3189        pcretest to make use of this.
3191     4. On the advice of a Windows user, the lines
3193          #if defined(_WIN32) || defined(WIN32)
3194          _setmode( _fileno( stdout ), 0x8000 );
3195          #endif  /* defined(_WIN32) || defined(WIN32) */
3197        have been added to the source of pcretest. This apparently does useful
3198        magic in relation to line terminators.
3200     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
3201        for the benefit of those environments where the "b" makes a difference.
3203     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
3204        to know about it. I have put a hack into configure.in that adds in code
3205        to set GCC=yes if CC=icc. This seems to end up at a point in the
3206        generated configure script that is early enough to affect the setting of
3207        compiler options, which is what is needed, but I have no means of testing
3208        whether it really works. (The user who reported this had patched the
3209        generated configure script, which of course I cannot do.)
3211        LATER: After change 22 below (new libtool files), the configure script
3212        seems to know about icc (and also ecc). Therefore, I have commented out
3213        this hack in configure.in.
3215     7. Added support for pkg-config (2 patches were sent in).
3217     8. Negated POSIX character classes that used a combination of internal tables
3218        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
3219        [[:^ascii]]. Typically, they would match almost any characters. The other
3220        POSIX classes were not broken in this way.
3222     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
3223        to find the match, as PCRE was deluded into thinking that the match had to
3224        start at the start point or following a newline. The same bug applied to
3225        patterns with negative forward assertions or any backward assertions
3226        preceding ".*" at the start, unless the pattern required a fixed first
3227        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
3229    10. In UTF-8 mode, when moving forwards in the subject after a failed match
3230        starting at the last subject character, bytes beyond the end of the subject
3231        string were read.
3233    11. Renamed the variable "class" as "classbits" to make life easier for C++
3234        users. (Previously there was a macro definition, but it apparently wasn't
3235        enough.)
3237    12. Added the new field "tables" to the extra data so that tables can be passed
3238        in at exec time, or the internal tables can be re-selected. This allows
3239        a compiled regex to be saved and re-used at a later time by a different
3240        program that might have everything at different addresses.
3242    13. Modified the pcre-config script so that, when run on Solaris, it shows a
3243        -R library as well as a -L library.
3245    14. The debugging options of pcretest (-d on the command line or D on a
3246        pattern) showed incorrect output for anything following an extended class
3247        that contained multibyte characters and which was followed by a quantifier.
3249    15. Added optional support for general category Unicode character properties
3250        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
3251        support. It adds about 90K to the size of the library. The meanings of the
3252        inbuilt class escapes such as \d and \s have NOT been changed.
3254    16. Updated pcredemo.c to include calls to free() to release the memory for the
3255        compiled pattern.
3257    17. The generated file chartables.c was being created in the source directory
3258        instead of in the building directory. This caused the build to fail if the
3259        source directory was different from the building directory, and was
3260        read-only.
3262    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
3263        file. No doubt somebody will tell me if they don't make sense... Also added
3264        Dan Mooney's comments about building on OpenVMS.
3266    19. Added support for partial matching via the PCRE_PARTIAL option for
3267        pcre_exec() and the \P data escape in pcretest.
3269    20. Extended pcretest with 3 new pattern features:
3271        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
3272              write the compiled pattern to the file whose name is "rest-of-line".
3273              This is a straight binary dump of the data, with the saved pointer to
3274              the character tables forced to be NULL. The study data, if any, is
3275              written too. After writing, pcretest reads a new pattern.
3277        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
3278              compiled pattern from the given file. There must not be any
3279              occurrences of "<" in the file name (pretty unlikely); if there are,
3280              pcretest will instead treat the initial "<" as a pattern delimiter.
3281              After reading in the pattern, pcretest goes on to read data lines as
3282              usual.
3284        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
3285              and 16-bit fields in a compiled pattern, to simulate a pattern that
3286              was compiled on a host of opposite endianness.
3288    21. The pcre-exec() function can now cope with patterns that were compiled on
3289        hosts of opposite endianness, with this restriction:
3291          As for any compiled expression that is saved and used later, the tables
3292          pointer field cannot be preserved; the extra_data field in the arguments
3293          to pcre_exec() should be used to pass in a tables address if a value
3294          other than the default internal tables were used at compile time.
3296    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
3297        now diagnosed as an error. Previously, most of the time, a negative number
3298        would have been treated as zero, but if in addition "ovector" was passed as
3299        NULL, a crash could occur.
3301    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
3302        new versions from the libtool 1.5 distribution (the last one is a copy of
3303        a file called libtool.m4). This seems to have fixed the need to patch
3304        "configure" to support Darwin 1.3 (which I used to do). However, I still
3305        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
3306        workstation).
3308    24. Changed the PCRE licence to be the more standard "BSD" licence.
3311    Version 4.5 01-Dec-03
3312    ---------------------
3314     1. There has been some re-arrangement of the code for the match() function so
3315        that it can be compiled in a version that does not call itself recursively.
3316        Instead, it keeps those local variables that need separate instances for
3317        each "recursion" in a frame on the heap, and gets/frees frames whenever it
3318        needs to "recurse". Keeping track of where control must go is done by means
3319        of setjmp/longjmp. The whole thing is implemented by a set of macros that
3320        hide most of the details from the main code, and operates only if
3321        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
3322        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
3323        operating.
3325        To make it easier for callers to provide specially tailored get/free
3326        functions for this usage, two new functions, pcre_stack_malloc, and
3327        pcre_stack_free, are used. They are always called in strict stacking order,
3328        and the size of block requested is always the same.
3330        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
3331        PCRE has been compiled to use the stack or the heap for recursion. The
3332        -C option of pcretest uses this to show which version is compiled.
3334        A new data escape \S, is added to pcretest; it causes the amounts of store
3335        obtained and freed by both kinds of malloc/free at match time to be added
3336        to the output.
3338     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
3339        what's available on my current Linux desktop machine.
3341     3. When matching a UTF-8 string, the test for a valid string at the start has
3342        been extended. If start_offset is not zero, PCRE now checks that it points
3343        to a byte that is the start of a UTF-8 character. If not, it returns
3344        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
3345        this is necessary because there may be backward assertions in the pattern.
3346        When matching the same subject several times, it may save resources to use
3347        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
3349     4. The code for checking the validity of UTF-8 strings has been tightened so
3350        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
3351        containing "overlong sequences".
3353     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
3354        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
3355        should have been "&", but it just so happened that all the cases this let
3356        through by mistake were picked up later in the function.
3358     6. I had used a variable called "isblank" - this is a C99 function, causing
3359        some compilers to warn. To avoid this, I renamed it (as "blankclass").
3361     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
3362        prompting; (b) run "./pcretest /dev/null" at the start of the test script
3363        so the version is shown; (c) stop "make test" echoing "./RunTest".
3365     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
3367     9. The prototype for memmove() for systems that don't have it was using
3368        size_t, but the inclusion of the header that defines size_t was later. I've
3369        moved the #includes for the C headers earlier to avoid this.
3371    10. Added some adjustments to the code to make it easier to compiler on certain
3372        special systems:
3374          (a) Some "const" qualifiers were missing.
3375          (b) Added the macro EXPORT before all exported functions; by default this
3376              is defined to be empty.
3377          (c) Changed the dftables auxiliary program (that builds chartables.c) so
3378              that it reads its output file name as an argument instead of writing
3379              to the standard output and assuming this can be redirected.
3381    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
3382        class containing characters with values greater than 255, PCRE compilation
3383        went into a loop.
3385    12. A recursive reference to a subpattern that was within another subpattern
3386        that had a minimum quantifier of zero caused PCRE to crash. For example,
3387        (x(y(?2))z)? provoked this bug with a subject that got as far as the
3388        recursion. If the recursively-called subpattern itself had a zero repeat,
3389        that was OK.
3391    13. In pcretest, the buffer for reading a data line was set at 30K, but the
3392        buffer into which it was copied (for escape processing) was still set at
3393        1024, so long lines caused crashes.
3395    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
3396        "internal error: code overflow...". This applied to any character class
3397        that was followed by a possessive quantifier.
3399    15. Modified the Makefile to add libpcre.la as a prerequisite for
3400        libpcreposix.la because I was told this is needed for a parallel build to
3401        work.
3403    16. If a pattern that contained .* following optional items at the start was
3404        studied, the wrong optimizing data was generated, leading to matching
3405        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
3406        matching string must start with a or b or c. The correct conclusion for
3407        this pattern is that a match can start with any character.
3410    Version 4.4 13-Aug-03
3411    ---------------------
3413     1. In UTF-8 mode, a character class containing characters with values between
3414        127 and 255 was not handled correctly if the compiled pattern was studied.
3415        In fixing this, I have also improved the studying algorithm for such
3416        classes (slightly).
3418     2. Three internal functions had redundant arguments passed to them. Removal
3419        might give a very teeny performance improvement.
3421     3. Documentation bug: the value of the capture_top field in a callout is *one
3422        more than* the number of the hightest numbered captured substring.
3424     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
3425        in incorrectly linking with a previously installed version. They now link
3426        explicitly with libpcre.la.
3428     5. configure.in no longer needs to recognize Cygwin specially.
3430     6. A problem in pcre.in for Windows platforms is fixed.
3432     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
3433        pcretest, it used to include the size of the study block as part of its
3434        output. Unfortunately, the structure contains a field that has a different
3435        size on different hardware architectures. This meant that the tests that
3436        showed this size failed. As the block is currently always of a fixed size,
3437        this information isn't actually particularly useful in pcretest output, so
3438        I have just removed it.
3440     8. Three pre-processor statements accidentally did not start in column 1.
3441        Sadly, there are *still* compilers around that complain, even though
3442        standard C has not required this for well over a decade. Sigh.
3444     9. In pcretest, the code for checking callouts passed small integers in the
3445        callout_data field, which is a void * field. However, some picky compilers
3446        complained about the casts involved for this on 64-bit systems. Now
3447        pcretest passes the address of the small integer instead, which should get
3448        rid of the warnings.
3450    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
3451        both compile and run time, and gives an error if an invalid UTF-8 sequence
3452        is found. There is a option for disabling this check in cases where the
3453        string is known to be correct and/or the maximum performance is wanted.
3455    11. In response to a bug report, I changed one line in Makefile.in from
3457            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
3458        to
3459            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
3461        to look similar to other lines, but I have no way of telling whether this
3462        is the right thing to do, as I do not use Windows. No doubt I'll get told
3463        if it's wrong...
3466    Version 4.3 21-May-03
3467    ---------------------
3469    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
3470       Makefile.
3472    2. Some refactoring to improve the quality of the code:
3474       (i)   The utf8_table... variables are now declared "const".
3476       (ii)  The code for \cx, which used the "case flipping" table to upper case
3477             lower case letters, now just substracts 32. This is ASCII-specific,
3478             but the whole concept of \cx is ASCII-specific, so it seems
3479             reasonable.
3481       (iii) PCRE was using its character types table to recognize decimal and
3482             hexadecimal digits in the pattern. This is silly, because it handles
3483             only 0-9, a-f, and A-F, but the character types table is locale-
3484             specific, which means strange things might happen. A private
3485             table is now used for this - though it costs 256 bytes, a table is
3486             much faster than multiple explicit tests. Of course, the standard
3487             character types table is still used for matching digits in subject
3488             strings against \d.
3490       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
3491             ending in _t are). So I've renamed it as ESC_tee.
3493    3. The first argument for regexec() in the POSIX wrapper should have been
3494       defined as "const".
3496    4. Changed pcretest to use malloc() for its buffers so that they can be
3497       Electric Fenced for debugging.
3499    5. There were several places in the code where, in UTF-8 mode, PCRE would try
3500       to read one or more bytes before the start of the subject string. Often this
3501       had no effect on PCRE's behaviour, but in some circumstances it could
3502       provoke a segmentation fault.
3504    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
3505       to try to read one or more bytes before the start of the subject string.
3507    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3508       UTF-8 support could misbehave in various ways if the subject string
3509       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3510       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3511       back over UTF-8 characters.)
3514    Version 4.2 14-Apr-03
3515    ---------------------
3517    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
3519    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
3520         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
3521         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
3522         [WIN_PREFIX]: new variable, "cyg" for Cygwin
3523         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
3524           and BUILD_EXEEXT
3525         Note: automatic setting of the BUILD variables is not yet working
3526         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
3527           compile-time but not at link-time
3528         [LINK]: use for linking executables only
3529         make different versions for Windows and non-Windows
3530         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
3531           libraries
3532         [LINK_FOR_BUILD]: new variable
3533         [OBJEXT]: use throughout
3534         [EXEEXT]: use throughout
3535         <winshared>: new target
3536         <wininstall>: new target
3537         <dftables.o>: use native compiler
3538         <dftables>: use native linker
3539         <install>: handle Windows platform correctly
3540         <clean>: ditto
3541         <check>: ditto
3542         copy DLL to top builddir before testing
3544       As part of these changes, -no-undefined was removed again. This was reported
3545       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3546       in any case.
3548    3. Some tidies to get rid of compiler warnings:
3550       . In the match_data structure, match_limit was an unsigned long int, whereas
3551         match_call_count was an int. I've made them both unsigned long ints.
3553       . In pcretest the fact that a const uschar * doesn't automatically cast to
3554         a void * provoked a warning.
3556       . Turning on some more compiler warnings threw up some "shadow" variables
3557         and a few more missing casts.
3559    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3560       option, a class that contained a single character with a value between 128
3561       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3563    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3564       option, a class that contained several characters, but with at least one
3565       whose value was between 128 and 255 caused PCRE to crash.
3568    Version 4.1 12-Mar-03
3569    ---------------------
3571    1. Compiling with gcc -pedantic found a couple of places where casts were
3572    needed, and a string in dftables.c that was longer than standard compilers are
3573    required to support.
3575    2. Compiling with Sun's compiler found a few more places where the code could
3576    be tidied up in order to avoid warnings.
3578    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3579    first of these names is deprecated in the latest Autoconf in favour of the name
3580    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3581    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3582    analogy I have changed it to CFLAGS_FOR_BUILD.
3584    4. Added -no-undefined to the linking command in the Makefile, because this is
3585    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
3586    linking step for the pcreposix library.
3588    5. PCRE was failing to diagnose the case of two named groups with the same
3589    name.