ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 181 by ph10, Wed Jun 13 14:55:18 2007 UTC revision 947 by ph10, Sun Mar 4 16:51:13 2012 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4  Version 7.2 13-June-07  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.
76    Version 8.30 04-February-2012
77    -----------------------------
79    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
80        name is defined in ctype.h.
82    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
83        only in quite long subpatterns.
85    3.  Removed the function pcre_info(), which has been obsolete and deprecated
86        since it was replaced by pcre_fullinfo() in February 2000.
88    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
89        match a (*MARK), and the match failed at the start of the subject, a
90        reference to memory before the start of the subject could occur. This bug
91        was introduced by fix 17 of release 8.21.
93    5.  A reference to an unset group with zero minimum repetition was giving
94        totally wrong answers (in non-JavaScript-compatibility mode). For example,
95        /(another)?(\1?)test/ matched against "hello world test". This bug was
96        introduced in release 8.13.
98    6.  Add support for 16-bit character strings (a large amount of work involving
99        many changes and refactorings).
101    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
102        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
103        from a file.
105    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
106        rounding is not applied in this particular case).
108    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
109        if they appear, or are escaped, in patterns.
111    10. Get rid of a number of -Wunused-but-set-variable warnings.
113    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
114        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
115        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
116        also returns the mark "x". This bug applied to capturing parentheses,
117        non-capturing parentheses, and atomic parentheses. It also applied to some
118        assertions.
120    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
121        information out of configure.ac instead of relying on pcre.h.generic, which
122        is not stored in the repository.
124    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
125        -lreadline.
127    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
129    15. Applied Graycode's patch to put the top-level frame on the stack rather
130        than the heap when not using the stack for recursion. This gives a
131        performance improvement in many cases when recursion is not deep.
133    16. Experimental code added to "pcretest -C" to output the stack frame size.
136    Version 8.21 12-Dec-2011
137    ------------------------
139    1.  Updating the JIT compiler.
141    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
142        are added as well.
144    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
145        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
146        calling _pcre_jit_exec. Some extra comments are added.
148    4.  (*MARK) settings inside atomic groups that do not contain any capturing
149        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
150        was introduced by change 18 for 8.20.
152    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
153        ECMA-262 standard.
155    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
156        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
157        This bug was probably introduced by change 9 of 8.13.
159    7.  While fixing 6 above, I noticed that a number of other items were being
160        incorrectly rejected as "not fixed length". This arose partly because newer
161        opcodes had not been added to the fixed-length checking code. I have (a)
162        corrected the bug and added tests for these items, and (b) arranged for an
163        error to occur if an unknown opcode is encountered while checking for fixed
164        length instead of just assuming "not fixed length". The items that were
165        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
166        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
167        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
169    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
170        being incorrectly compiled and would have given unpredicatble results.
172    9.  A possessively repeated subpattern with minimum repeat count greater than
173        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
174        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
175        the first (A) could occur when it should not.
177    10. Add a cast and remove a redundant test from the code.
179    11. JIT should use pcre_malloc/pcre_free for allocation.
181    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
182        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
183        is anything other than /usr, -L is still shown).
185    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
187    14. Perl does not support \N without a following name in a [] class; PCRE now
188        also gives an error.
190    15. If a forward reference was repeated with an upper limit of around 2000,
191        it caused the error "internal error: overran compiling workspace". The
192        maximum number of forward references (including repeats) was limited by the
193        internal workspace, and dependent on the LINK_SIZE. The code has been
194        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
195        the default depends on LINK_SIZE. There is a new upper limit (for safety)
196        of around 200,000 forward references. While doing this, I also speeded up
197        the filling in of repeated forward references.
199    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
200        incorrectly expecting the subject to contain another "a" after the start.
202    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
203        in the match, the SKIP should be ignored. This was not happening; instead
204        the SKIP was being treated as NOMATCH. For patterns such as
205        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
206        tested.
208    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
209        now much more compatible with Perl, in particular in cases where the result
210        is a non-match for a non-anchored pattern. For example, if
211        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
212        "m", where previously it did not return a name. A side effect of this
213        change is that for partial matches, the last encountered mark name is
214        returned, as for non matches. A number of tests that were previously not
215        Perl-compatible have been moved into the Perl-compatible test files. The
216        refactoring has had the pleasing side effect of removing one argument from
217        the match() function, thus reducing its stack requirements.
219    19. If the /S+ option was used in pcretest to study a pattern using JIT,
220        subsequent uses of /S (without +) incorrectly behaved like /S+.
222    21. Retrieve executable code size support for the JIT compiler and fixing
223        some warnings.
225    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
226        not work when the shorter character appeared right at the end of the
227        subject string.
229    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
230        systems.
232    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
233        output it when the /M option is used in pcretest.
235    25. The CheckMan script was not being included in the distribution. Also, added
236        an explicit "perl" to run Perl scripts from the PrepareRelease script
237        because this is reportedly needed in Windows.
239    26. If study data was being save in a file and studying had not found a set of
240        "starts with" bytes for the pattern, the data written to the file (though
241        never used) was taken from uninitialized memory and so caused valgrind to
242        complain.
244    27. Updated RunTest.bat as provided by Sheri Pierce.
246    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
248    29. Computation of memory usage for the table of capturing group names was
249        giving an unnecessarily large value.
252    Version 8.20 21-Oct-2011
253    ------------------------
255    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
256        a POSIX class. After further experiments with Perl, which convinced me that
257        Perl has bugs and confusions, a closing square bracket is no longer allowed
258        in a POSIX name. This bug also affected patterns with classes that started
259        with full stops.
261    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
262        captured substring, but while checking the failing first alternative,
263        substring 1 is temporarily captured. If the output vector supplied to
264        pcre_exec() was not big enough for this capture, the yield of the function
265        was still zero ("insufficient space for captured substrings"). This cannot
266        be totally fixed without adding another stack variable, which seems a lot
267        of expense for a edge case. However, I have improved the situation in cases
268        such as /(a)(b)x|abc/ matched against "abc", where the return code
269        indicates that fewer than the maximum number of slots in the ovector have
270        been set.
272    3.  Related to (2) above: when there are more back references in a pattern than
273        slots in the output vector, pcre_exec() uses temporary memory during
274        matching, and copies in the captures as far as possible afterwards. It was
275        using the entire output vector, but this conflicts with the specification
276        that only 2/3 is used for passing back captured substrings. Now it uses
277        only the first 2/3, for compatibility. This is, of course, another edge
278        case.
280    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
281        main code base, and can be used by building with --enable-jit. When this is
282        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
283        runtime --no-jit option is given.
285    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
286        ovector, the return from the function was zero, implying that there were
287        other matches that did not fit. The correct "exactly full" value is now
288        returned.
290    6.  If a subpattern that was called recursively or as a subroutine contained
291        (*PRUNE) or any other control that caused it to give a non-standard return,
292        invalid errors such as "Error -26 (nested recursion at the same subject
293        position)" or even infinite loops could occur.
295    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
296        computing the minimum length on reaching *ACCEPT, and so ended up with the
297        wrong value of 1 rather than 0. Further investigation indicates that
298        computing a minimum subject length in the presence of *ACCEPT is difficult
299        (think back references, subroutine calls), and so I have changed the code
300        so that no minimum is registered for a pattern that contains *ACCEPT.
302    8.  If (*THEN) was present in the first (true) branch of a conditional group,
303        it was not handled as intended. [But see 16 below.]
305    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
306        Sheri Pierce.
308    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
309        the first byte in a match must be "a".
311    11. Change 17 for 8.13 increased the recursion depth for patterns like
312        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
313        pattern contains any instances of (*THEN). If it does not, the old
314        optimizations are restored. It would be nice to do this on a per-group
315        basis, but at the moment that is not feasible.
317    12. In some environments, the output of pcretest -C is CRLF terminated. This
318        broke RunTest's code that checks for the link size. A single white space
319        character after the value is now allowed for.
321    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
322        For "fr", it uses the Windows-specific input and output files.
324    14. If (*THEN) appeared in a group that was called recursively or as a
325        subroutine, it did not work as intended. [But see next item.]
327    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
328        pattern fragments (but not containing any | characters). If A and B are
329        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
330        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
331        D. In other words, Perl considers parentheses that do not contain any |
332        characters to be part of a surrounding alternative, whereas PCRE was
333        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
334        differently. PCRE now behaves in the same way as Perl, except in the case
335        of subroutine/recursion calls such as (?1) which have in any case always
336        been different (but PCRE had them first :-).
338    16. Related to 15 above: Perl does not treat the | in a conditional group as
339        creating alternatives. Such a group is treated in the same way as an
340        ordinary group without any | characters when processing (*THEN). PCRE has
341        been changed to match Perl's behaviour.
343    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
344        RunGrepTest script failed.
346    18. Change 22 for version 13 caused atomic groups to use more stack. This is
347        inevitable for groups that contain captures, but it can lead to a lot of
348        stack use in large patterns. The old behaviour has been restored for atomic
349        groups that do not contain any capturing parentheses.
351    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
352        suppress the check for a minimum subject length at run time. (If it was
353        given to pcre_exec() or pcre_dfa_exec() it did work.)
355    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
356        fail to work when decoding hex characters in data strings in EBCDIC
357        environments.
359    21. It appears that in at least one Mac OS environment, the isxdigit() function
360        is implemented as a macro that evaluates to its argument more than once,
361        contravening the C 90 Standard (I haven't checked a later standard). There
362        was an instance in pcretest which caused it to go wrong when processing
363        \x{...} escapes in subject strings. The has been rewritten to avoid using
364        things like p++ in the argument of isxdigit().
367    Version 8.13 16-Aug-2011
368    ------------------------
370    1.  The Unicode data tables have been updated to Unicode 6.0.0.
372    2.  Two minor typos in pcre_internal.h have been fixed.
374    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
375        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
376        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
378    4.  There were a number of related bugs in the code for matching backrefences
379        caselessly in UTF-8 mode when codes for the characters concerned were
380        different numbers of bytes. For example, U+023A and U+2C65 are an upper
381        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
382        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
383        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
384        2-byte code at the end of the subject (it thought there wasn't enough data
385        left).
387    5.  Comprehensive information about what went wrong is now returned by
388        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
389        as the output vector has at least 2 elements. The offset of the start of
390        the failing character and a reason code are placed in the vector.
392    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
393        now returned is for the first byte of the failing character, instead of the
394        last byte inspected. This is an incompatible change, but I hope it is small
395        enough not to be a problem. It makes the returned offset consistent with
396        pcre_exec() and pcre_dfa_exec().
398    7.  pcretest now gives a text phrase as well as the error number when
399        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
400        failure, the offset and reason code are output.
402    8.  When \R was used with a maximizing quantifier it failed to skip backwards
403        over a \r\n pair if the subsequent match failed. Instead, it just skipped
404        back over a single character (\n). This seems wrong (because it treated the
405        two characters as a single entity when going forwards), conflicts with the
406        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
407        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
408        has been changed.
410    9.  Some internal refactoring has changed the processing so that the handling
411        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
412        time (the PCRE_DOTALL option was changed this way some time ago: version
413        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
414        which was always a bit of a fudge. It also means that there is one less
415        argument for the match() function, which reduces its stack requirements
416        slightly. This change also fixes an incompatibility with Perl: the pattern
417        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
419    10. More internal refactoring has drastically reduced the number of recursive
420        calls to match() for possessively repeated groups such as (abc)++ when
421        using pcre_exec().
423    11. While implementing 10, a number of bugs in the handling of groups were
424        discovered and fixed:
426        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
427        (a|)*(?1) gave a compile-time internal error.
428        ((a|)+)+  did not notice that the outer group could match an empty string.
429        (^a|^)+   was not marked as anchored.
430        (.*a|.*)+ was not marked as matching at start or after a newline.
432    12. Yet more internal refactoring has removed another argument from the match()
433        function. Special calls to this function are now indicated by setting a
434        value in a variable in the "match data" data block.
436    13. Be more explicit in pcre_study() instead of relying on "default" for
437        opcodes that mean there is no starting character; this means that when new
438        ones are added and accidentally left out of pcre_study(), testing should
439        pick them up.
441    14. The -s option of pcretest has been documented for ages as being an old
442        synonym of -m (show memory usage). I have changed it to mean "force study
443        for every regex", that is, assume /S for every regex. This is similar to -i
444        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
445        using it. It makes it easier to run collections of tests with and without
446        study enabled, and thereby test pcre_study() more easily. All the standard
447        tests are now run with and without -s (but some patterns can be marked as
448        "never study" - see 20 below).
450    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
451        restoration of the capturing data to the outer values was not happening
452        correctly.
454    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
455        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
456        pattern had matched an empty string, and so incorrectly returned a no
457        match.
459    17. There was optimizing code for the last branch of non-capturing parentheses,
460        and also for the obeyed branch of a conditional subexpression, which used
461        tail recursion to cut down on stack usage. Unfortunately, now that there is
462        the possibility of (*THEN) occurring in these branches, tail recursion is
463        no longer possible because the return has to be checked for (*THEN). These
464        two optimizations have therefore been removed. [But see 8.20/11 above.]
466    18. If a pattern containing \R was studied, it was assumed that \R always
467        matched two bytes, thus causing the minimum subject length to be
468        incorrectly computed because \R can also match just one byte.
470    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
471        was incorrectly computed.
473    20. If /S is present twice on a test pattern in pcretest input, it now
474        *disables* studying, thereby overriding the use of -s on the command line
475        (see 14 above). This is necessary for one or two tests to keep the output
476        identical in both cases.
478    21. When (*ACCEPT) was used in an assertion that matched an empty string and
479        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
481    22. When an atomic group that contained a capturing parenthesis was
482        successfully matched, but the branch in which it appeared failed, the
483        capturing was not being forgotten if a higher numbered group was later
484        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
485        group 1 to "a", when in fact it should be unset. This applied to multi-
486        branched capturing and non-capturing groups, repeated or not, and also to
487        positive assertions (capturing in negative assertions does not happen
488        in PCRE) and also to nested atomic groups.
490    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
491        subject after a captured substring, to make it easier to tell which of a
492        number of identical substrings has been captured.
494    24. The way atomic groups are processed by pcre_exec() has been changed so that
495        if they are repeated, backtracking one repetition now resets captured
496        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
497        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
498        "aaa". Previously, it would have been "a". As part of this code
499        refactoring, the way recursive calls are handled has also been changed.
501    25. If an assertion condition captured any substrings, they were not passed
502        back unless some other capturing happened later. For example, if
503        (?(?=(a))a) was matched against "a", no capturing was returned.
505    26. When studying a pattern that contained subroutine calls or assertions,
506        the code for finding the minimum length of a possible match was handling
507        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
508        group 1 called group 2 while simultaneously a separate group 2 called group
509        1). A stack overflow occurred in this case. I have fixed this by limiting
510        the recursion depth to 10.
512    27. Updated RunTest.bat in the distribution to the version supplied by Tom
513        Fortmann. This supports explicit test numbers on the command line, and has
514        argument validation and error reporting.
516    28. An instance of \X with an unlimited repeat could fail if at any point the
517        first character it looked at was a mark character.
519    29. Some minor code refactoring concerning Unicode properties and scripts
520        should reduce the stack requirement of match() slightly.
522    30. Added the '=' option to pcretest to check the setting of unused capturing
523        slots at the end of the pattern, which are documented as being -1, but are
524        not included in the return count.
526    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
527        compiled something random. Now it gives a compile-time error (as does
528        Perl).
530    32. A *MARK encountered during the processing of a positive assertion is now
531        recorded and passed back (compatible with Perl).
533    33. If --only-matching or --colour was set on a pcregrep call whose pattern
534        had alternative anchored branches, the search for a second match in a line
535        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
536        matched the line "0102" twice. The same bug affected patterns that started
537        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
538        twice.
540    34. Previously, PCRE did not allow quantification of assertions. However, Perl
541        does, and because of capturing effects, quantifying parenthesized
542        assertions may at times be useful. Quantifiers are now allowed for
543        parenthesized assertions.
545    35. A minor code tidy in pcre_compile() when checking options for \R usage.
547    36. \g was being checked for fancy things in a character class, when it should
548        just be a literal "g".
550    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
551        appearance of a nested POSIX class supersedes an apparent external class.
552        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
553        unescaped square brackets may also appear as part of class names. For
554        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
555        more like Perl. (But see 8.20/1 above.)
557    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
558        was because it thought it was \N{name}, which is not supported).
560    39. Add minix to OS list not supporting the -S option in pcretest.
562    40. PCRE tries to detect cases of infinite recursion at compile time, but it
563        cannot analyze patterns in sufficient detail to catch mutual recursions
564        such as ((?1))((?2)). There is now a runtime test that gives an error if a
565        subgroup is called recursively as a subpattern for a second time at the
566        same position in the subject string. In previous releases this might have
567        been caught by the recursion limit, or it might have run out of stack.
569    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
570        happen only once. PCRE was, however incorrectly giving a compile time error
571        "recursive call could loop indefinitely" because it cannot analyze the
572        pattern in sufficient detail. The compile time test no longer happens when
573        PCRE is compiling a conditional subpattern, but actual runaway loops are
574        now caught at runtime (see 40 above).
576    42. It seems that Perl allows any characters other than a closing parenthesis
577        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
578        has been changed to be the same.
580    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
581        as not to get warnings when autogen.sh is called. Also changed
582        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
584    44. To help people who use pcregrep to scan files containing exceedingly long
585        lines, the following changes have been made:
587        (a) The default value of the buffer size parameter has been increased from
588            8K to 20K. (The actual buffer used is three times this size.)
590        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
591            PCRE is built.
593        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
594            to be set at run time.
596        (d) Numerical values in pcregrep options can be followed by K or M, for
597            example --buffer-size=50K.
599        (e) If a line being scanned overflows pcregrep's buffer, an error is now
600            given and the return code is set to 2.
602    45. Add a pointer to the latest mark to the callout data block.
604    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
605        partial match of an empty string instead of no match. This was specific to
606        the use of ".".
608    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
609        complete match instead of a partial match. This bug was dependent on both
610        the PCRE_UTF8 and PCRE_DOTALL options being set.
612    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
613        starting byte set, because \b was not being ignored.
616    Version 8.12 15-Jan-2011
617    ------------------------
619    1.  Fixed some typos in the markup of the man pages, and wrote a script that
620        checks for such things as part of the documentation building process.
622    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
623        --match-limit and --recursion-limit options (added for 8.11). In
624        particular, this made one of the standard tests fail. (The integer value
625        went into the wrong half of a long int.)
627    3.  If the --colour option was given to pcregrep with -v (invert match), it
628        did strange things, either producing crazy output, or crashing. It should,
629        of course, ignore a request for colour when reporting lines that do not
630        match.
632    4.  Another pcregrep bug caused similar problems if --colour was specified with
633        -M (multiline) and the pattern match finished with a line ending.
635    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
636        matched in multiline mode, the following line was shown as part of the
637        match. This seems wrong, so I have changed it.
639    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
640        the check for further matches in the same line (so they could be coloured)
641        to overrun the end of the current line. If another match was found, it was
642        incorrectly shown (and then shown again when found in the next line).
644    7.  If pcregrep was compiled under Windows, there was a reference to the
645        function pcregrep_exit() before it was defined. I am assuming this was
646        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
647        reported by a user. I've moved the definition above the reference.
650    Version 8.11 10-Dec-2010
651    ------------------------
653    1.  (*THEN) was not working properly if there were untried alternatives prior
654        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
655        backtracked to try for "b" instead of moving to the next alternative branch
656        at the same level (in this case, to look for "c"). The Perl documentation
657        is clear that when (*THEN) is backtracked onto, it goes to the "next
658        alternative in the innermost enclosing group".
660    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
661        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
662        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
663        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
664        (*THEN).
666    3.  If \s appeared in a character class, it removed the VT character from
667        the class, even if it had been included by some previous item, for example
668        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
669        of \s, but is part of the POSIX "space" class.)
671    4.  A partial match never returns an empty string (because you can always
672        match an empty string at the end of the subject); however the checking for
673        an empty string was starting at the "start of match" point. This has been
674        changed to the "earliest inspected character" point, because the returned
675        data for a partial match starts at this character. This means that, for
676        example, /(?<=abc)def/ gives a partial match for the subject "abc"
677        (previously it gave "no match").
679    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
680        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
681        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
682        has an implication that the given string is incomplete (because a partial
683        match is preferred over a full match). For this reason, these items now
684        give a partial match in this situation. [Aside: previously, the one case
685        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
686        match rather than a full match, which was wrong by the old rules, but is
687        now correct.]
689    6.  There was a bug in the handling of #-introduced comments, recognized when
690        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
691        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
692        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
693        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
694        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
695        places in pcre_compile().
697    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
698        comments when looking ahead for named forward references to subpatterns,
699        the only newline sequence it recognized was NL. It now handles newlines
700        according to the set newline convention.
702    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
703        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
704        cater for a lack of strerror(). These oversights have been fixed.
706    9.  Added --match-limit and --recursion-limit to pcregrep.
708    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
710    11. When the -o option was used, pcregrep was setting a return code of 1, even
711        when matches were found, and --line-buffered was not being honoured.
713    12. Added an optional parentheses number to the -o and --only-matching options
714        of pcregrep.
716    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
717        can match an empty string. The code to do it in pcretest and pcredemo
718        needed fixing:
720        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
721            only one byte after an empty string match just before CRLF (this case
722            just got forgotten; "any" and "anycrlf" were OK).
724        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
725            mode when an empty string match preceded an ASCII character followed by
726            a non-ASCII character. (The code for advancing by one character rather
727            than one byte was nonsense.)
729        (c) The pcredemo.c sample program did not have any code at all to handle
730            the cases when CRLF is a valid newline sequence.
732    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
733        as a starting offset was within the subject string. There is now a new
734        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
735        negative or greater than the length of the string. In order to test this,
736        pcretest is extended to allow the setting of negative starting offsets.
738    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
739        starting offset points to the beginning of a UTF-8 character was
740        unnecessarily clumsy. I tidied it up.
742    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
743        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
745    17. Nobody had reported that the --include_dir option, which was added in
746        release 7.7 should have been called --include-dir (hyphen, not underscore)
747        for compatibility with GNU grep. I have changed it to --include-dir, but
748        left --include_dir as an undocumented synonym, and the same for
749        --exclude-dir, though that is not available in GNU grep, at least as of
750        release 2.5.4.
752    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
753        characters from a string of bytes) have been redefined so as not to use
754        loops, in order to improve performance in some environments. At the same
755        time, I abstracted some of the common code into auxiliary macros to save
756        repetition (this should not affect the compiled code).
758    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
759        compile-time error is now given if \c is not followed by an ASCII
760        character, that is, a byte less than 128. (In EBCDIC mode, the code is
761        different, and any byte value is allowed.)
763    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
764        START_OPTIMIZE option, which is now allowed at compile time - but just
765        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
766        to pcregrep and other applications that have no direct access to PCRE
767        options. The new /Y option in pcretest sets this option when calling
768        pcre_compile().
770    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
771        back references. Groups containing recursive back references were forced to
772        be atomic by that change, but in the case of named groups, the amount of
773        memory required was incorrectly computed, leading to "Failed: internal
774        error: code overflow". This has been fixed.
776    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
777        pcretest.c, to avoid build problems in some Borland environments.
780    Version 8.10 25-Jun-2010
781    ------------------------
783    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
784        THEN.
786    2.  (*ACCEPT) was not working when inside an atomic group.
788    3.  Inside a character class, \B is treated as a literal by default, but
789        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
790        causes the error). The code is unchanged, but I tidied the documentation.
792    4.  Inside a character class, PCRE always treated \R and \X as literals,
793        whereas Perl faults them if its -w option is set. I have changed PCRE so
794        that it faults them when PCRE_EXTRA is set.
796    5.  Added support for \N, which always matches any character other than
797        newline. (It is the same as "." when PCRE_DOTALL is not set.)
799    6.  When compiling pcregrep with newer versions of gcc which may have
800        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
801        declared with attribute warn_unused_result" were given. Just casting the
802        result to (void) does not stop the warnings; a more elaborate fudge is
803        needed. I've used a macro to implement this.
805    7.  Minor change to pcretest.c to avoid a compiler warning.
807    8.  Added four artifical Unicode properties to help with an option to make
808        \s etc use properties (see next item). The new properties are: Xan
809        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
811    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
812        use Unicode properties. (*UCP) at the start of a pattern can be used to set
813        this option. Modified pcretest to add /W to test this facility. Added
814        REG_UCP to make it available via the POSIX interface.
816    10. Added --line-buffered to pcregrep.
818    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
819        studied, and the match started with a letter with a code point greater than
820        127 whose first byte was different to the first byte of the other case of
821        the letter, the other case of this starting letter was not recognized
822        (#976).
824    12. If a pattern that was studied started with a repeated Unicode property
825        test, for example, \p{Nd}+, there was the theoretical possibility of
826        setting up an incorrect bitmap of starting bytes, but fortunately it could
827        not have actually happened in practice until change 8 above was made (it
828        added property types that matched character-matching opcodes).
830    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
831        possible starting bytes for non-anchored patterns.
833    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
834        \R, and also a number of cases that involve Unicode properties, both
835        explicit and implicit when PCRE_UCP is set.
837    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
838        input, it could crash or give wrong results if characters with values
839        greater than 0xc0 were present in the subject string. (Detail: it assumed
840        UTF-8 input when processing these items.)
842    16. Added a lot of (int) casts to avoid compiler warnings in systems where
843        size_t is 64-bit (#991).
845    17. Added a check for running out of memory when PCRE is compiled with
846        --disable-stack-for-recursion (#990).
848    18. If the last data line in a file for pcretest does not have a newline on
849        the end, a newline was missing in the output.
851    19. The default pcre_chartables.c file recognizes only ASCII characters (values
852        less than 128) in its various bitmaps. However, there is a facility for
853        generating tables according to the current locale when PCRE is compiled. It
854        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
855        characters, are recognized by isspace() and therefore were getting set in
856        these tables, and indeed these tables seem to approximate to ISO 8859. This
857        caused a problem in UTF-8 mode when pcre_study() was used to create a list
858        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
859        which of course cannot start UTF-8 characters. I have changed the code so
860        that only real ASCII characters (less than 128) and the correct starting
861        bytes for UTF-8 encodings are set for characters greater than 127 when in
862        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
863        altogether.)
865    20. Added the /T option to pcretest so as to be able to run tests with non-
866        standard character tables, thus making it possible to include the tests
867        used for 19 above in the standard set of tests.
869    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
870        reference to a subpattern the other side of a comment that contains an
871        opening parenthesis caused either an internal compiling error, or a
872        reference to the wrong subpattern.
875    Version 8.02 19-Mar-2010
876    ------------------------
878    1.  The Unicode data tables have been updated to Unicode 5.2.0.
880    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
881        configured.
883    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
884        original author of that file, following a query about its status.
886    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
887        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
889    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
890        quantifier applied to a forward-referencing subroutine call, could compile
891        incorrect code or give the error "internal error: previously-checked
892        referenced subpattern not found".
894    6.  Both MS Visual Studio and Symbian OS have problems with initializing
895        variables to point to external functions. For these systems, therefore,
896        pcre_malloc etc. are now initialized to local functions that call the
897        relevant global functions.
899    7.  There were two entries missing in the vectors called coptable and poptable
900        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
901        I've fixed the data, and added a kludgy way of testing at compile time that
902        the lengths are correct (equal to the number of opcodes).
904    8.  Following on from 7, I added a similar kludge to check the length of the
905        eint vector in pcreposix.c.
907    9.  Error texts for pcre_compile() are held as one long string to avoid too
908        much relocation at load time. To find a text, the string is searched,
909        counting zeros. There was no check for running off the end of the string,
910        which could happen if a new error number was added without updating the
911        string.
913    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
915    11. \K was not working if it appeared in an atomic group or in a group that
916        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
917        \K is "not well defined" if used in an assertion. PCRE now accepts it if
918        the assertion is positive, but not if it is negative.
920    12. Change 11 fortuitously reduced the size of the stack frame used in the
921        "match()" function of pcre_exec.c by one pointer. Forthcoming
922        implementation of support for (*MARK) will need an extra pointer on the
923        stack; I have reserved it now, so that the stack frame size does not
924        decrease.
926    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
927        item in branch that calls a recursion is a subroutine call - as in the
928        second branch in the above example - was incorrectly given the compile-
929        time error "recursive call could loop indefinitely" because pcre_compile()
930        was not correctly checking the subroutine for matching a non-empty string.
932    14. The checks for overrunning compiling workspace could trigger after an
933        overrun had occurred. This is a "should never occur" error, but it can be
934        triggered by pathological patterns such as hundreds of nested parentheses.
935        The checks now trigger 100 bytes before the end of the workspace.
937    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
940    Version 8.01 19-Jan-2010
941    ------------------------
943    1.  If a pattern contained a conditional subpattern with only one branch (in
944        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
945        computed the wrong minimum data length (which is of course zero for such
946        subpatterns). This could cause incorrect "no match" results.
948    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
949        the pattern is reset in the first branch, pcre_compile() failed with
950        "internal error: code overflow at offset...". This happened only when
951        the reset was to the original external option setting. (An optimization
952        abstracts leading options settings into an external setting, which was the
953        cause of this.)
955    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
956        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
957        assertion pattern did not match (meaning that the assertion was true), it
958        was incorrectly treated as false if the SKIP had been reached during the
959        matching. This also applied to assertions used as conditions.
961    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
962        assertion subpattern, including such a pattern used as a condition,
963        unpredictable results occurred, instead of the error return
966    5.  The C++ GlobalReplace function was not working like Perl for the special
967        situation when an empty string is matched. It now does the fancy magic
968        stuff that is necessary.
970    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
971        removed. (These were left over from very, very early versions of PCRE.)
973    7.  Some cosmetic changes to the code to make life easier when compiling it
974        as part of something else:
976        (a) Change DEBUG to PCRE_DEBUG.
978        (b) In pcre_compile(), rename the member of the "branch_chain" structure
979            called "current" as "current_branch", to prevent a collision with the
980            Linux macro when compiled as a kernel module.
982        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
983            prevent a collision with the Linux macro when compiled as a kernel
984            module.
986    8.  In pcre_compile() there are some checks for integer overflows that used to
987        cast potentially large values to (double). This has been changed to that
988        when building, a check for int64_t is made, and if it is found, it is used
989        instead, thus avoiding the use of floating point arithmetic. (There is no
990        other use of FP in PCRE.) If int64_t is not found, the fallback is to
991        double.
993    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
994        2005 (difference between two addresses compared to an unsigned value).
996    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
997        custom one, because of the following reported problem in Windows:
999          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1000              under Win32.
1001          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1002              therefore missing the function definition.
1003          - The compiler thus generates a "C" signature for the test function.
1004          - The linker fails to find the "C" function.
1005          - PCRE fails to configure if asked to do so against libbz2.
1007    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1008        messages were output:
1010          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1011          rerunning libtoolize, to keep the correct libtool macros in-tree.
1012          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1014        I have done both of these things.
1016    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1017        most of the time, it *can* run out if it is given a pattern that contains a
1018        runaway infinite recursion. I updated the discussion in the pcrestack man
1019        page.
1021    13. Now that we have gone to the x.xx style of version numbers, the minor
1022        version may start with zero. Using 08 or 09 is a bad idea because users
1023        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1024        interpreted as invalid octal numbers. I've updated the previous comment in
1025        configure.ac, and also added a check that gives an error if 08 or 09 are
1026        used.
1028    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1029        causing partial matching to fail when the end of the subject matched \W
1030        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1032    15. There were some discrepancies between the declarations in pcre_internal.h
1033        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1034        their definitions. The declarations used "const uschar *" and the
1035        definitions used USPTR. Even though USPTR is normally defined as "const
1036        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1037        reported that: "This difference in casting confuses some C++ compilers, for
1038        example, SunCC recognizes above declarations as different functions and
1039        generates broken code for hbpcre." I have changed the declarations to use
1040        USPTR.
1042    16. GNU libtool is named differently on some systems. The autogen.sh script now
1043        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1044        (FreeBSD).
1046    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1047        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1048        comment: "Figure out how to create a longlong from a string: strtoll and
1049        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1050        instance, but it only takes 2 args instead of 3!"
1052    18. A subtle bug concerned with back references has been fixed by a change of
1053        specification, with a corresponding code fix. A pattern such as
1054        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1055        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1056        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1057        same bug. Such groups have to be quantified to be useful, or contained
1058        inside another quantified group. (If there's no repetition, the reference
1059        can never match.) The problem arises because, having left the group and
1060        moved on to the rest of the pattern, a later failure that backtracks into
1061        the group uses the captured value from the final iteration of the group
1062        rather than the correct earlier one. I have fixed this in PCRE by forcing
1063        any group that contains a reference to itself to be an atomic group; that
1064        is, there cannot be any backtracking into it once it has completed. This is
1065        similar to recursive and subroutine calls.
1068    Version 8.00 19-Oct-09
1069    ----------------------
1071    1.  The table for translating pcre_compile() error codes into POSIX error codes
1072        was out-of-date, and there was no check on the pcre_compile() error code
1073        being within the table. This could lead to an OK return being given in
1074        error.
1076    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1077        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1078        in a Windows environment.
1080    3.  The pcregrep --count option prints the count for each file even when it is
1081        zero, as does GNU grep. However, pcregrep was also printing all files when
1082        --files-with-matches was added. Now, when both options are given, it prints
1083        counts only for those files that have at least one match. (GNU grep just
1084        prints the file name in this circumstance, but including the count seems
1085        more useful - otherwise, why use --count?) Also ensured that the
1086        combination -clh just lists non-zero counts, with no names.
1088    4.  The long form of the pcregrep -F option was incorrectly implemented as
1089        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1090        but it seems right to fix it, and I didn't think it was worth preserving
1091        the old behaviour.
1093    5.  The command line items --regex=pattern and --regexp=pattern were not
1094        recognized by pcregrep, which required --regex pattern or --regexp pattern
1095        (with a space rather than an '='). The man page documented the '=' forms,
1096        which are compatible with GNU grep; these now work.
1098    6.  No libpcreposix.pc file was created for pkg-config; there was just
1099        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1101    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1102        when UCP support is not needed, by modifying the Python script that
1103        generates it from Unicode data files. This should not matter if the module
1104        is correctly used as a library, but I received one complaint about 50K of
1105        unwanted data. My guess is that the person linked everything into his
1106        program rather than using a library. Anyway, it does no harm.
1108    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1109        was a minimum greater than 1 for a wide character in a possessive
1110        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1111        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1112        character. Chaos in the form of incorrect output or a compiling loop could
1113        result.
1115    9.  The restrictions on what a pattern can contain when partial matching is
1116        requested for pcre_exec() have been removed. All patterns can now be
1117        partially matched by this function. In addition, if there are at least two
1118        slots in the offset vector, the offset of the earliest inspected character
1119        for the match and the offset of the end of the subject are set in them when
1120        PCRE_ERROR_PARTIAL is returned.
1122    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1123        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1124        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1125        and may be more useful for multi-segment matching.
1127    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1128        used to be given if ever the end of the subject was reached; now it is
1129        given only if matching could not proceed because another character was
1130        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1131        string "Z", which now yields "no match" instead of "partial match". In the
1132        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1133        final character ended with (*FAIL).
1135    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1136        if the pattern had a "must contain" character that was already found in the
1137        earlier partial match, unless partial matching was again requested. For
1138        example, with the pattern /dog.(body)?/, the "must contain" character is
1139        "g". If the first part-match was for the string "dog", restarting with
1140        "sbody" failed. This bug has been fixed.
1142    13. The string returned by pcre_dfa_exec() after a partial match has been
1143        changed so that it starts at the first inspected character rather than the
1144        first character of the match. This makes a difference only if the pattern
1145        starts with a lookbehind assertion or \b or \B (\K is not supported by
1146        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1147        matching functions compatible, and I think it's the right thing to do.
1149    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1150        so that the demonstration program is easily available in environments where
1151        PCRE has not been installed from source.
1153    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1154        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1155        library.
1157    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1158        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1159        is not the first non-POSIX option to be added. Clearly some people find
1160        these options useful.
1162    17. If a caller to the POSIX matching function regexec() passes a non-zero
1163        value for nmatch with a NULL value for pmatch, the value of
1164        nmatch is forced to zero.
1166    18. RunGrepTest did not have a test for the availability of the -u option of
1167        the diff command, as RunTest does. It now checks in the same way as
1168        RunTest, and also checks for the -b option.
1170    19. If an odd number of negated classes containing just a single character
1171        interposed, within parentheses, between a forward reference to a named
1172        subpattern and the definition of the subpattern, compilation crashed with
1173        an internal error, complaining that it could not find the referenced
1174        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1175        [The bug was that it was starting one character too far in when skipping
1176        over the character class, thus treating the ] as data rather than
1177        terminating the class. This meant it could skip too much.]
1179    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1180        /g option in pcretest when the pattern contains \K, which makes it possible
1181        to have an empty string match not at the start, even when the pattern is
1182        anchored. Updated pcretest and pcredemo to use this option.
1184    21. If the maximum number of capturing subpatterns in a recursion was greater
1185        than the maximum at the outer level, the higher number was returned, but
1186        with unset values at the outer level. The correct (outer level) value is
1187        now given.
1189    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1190        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1191        make it do so. The string so far is captured, making this feature
1192        compatible with Perl.
1194    23. The tests have been re-organized, adding tests 11 and 12, to make it
1195        possible to check the Perl 5.10 features against Perl 5.10.
1197    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1198        pattern matches a fixed length string. PCRE did not allow this; now it
1199        does. Neither allows recursion.
1201    25. I finally figured out how to implement a request to provide the minimum
1202        length of subject string that was needed in order to match a given pattern.
1203        (It was back references and recursion that I had previously got hung up
1204        on.) This code has now been added to pcre_study(); it finds a lower bound
1205        to the length of subject needed. It is not necessarily the greatest lower
1206        bound, but using it to avoid searching strings that are too short does give
1207        some useful speed-ups. The value is available to calling programs via
1208        pcre_fullinfo().
1210    26. While implementing 25, I discovered to my embarrassment that pcretest had
1211        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1212        study optimizations had never been tested with that matching function.
1213        Oops. What is worse, even when it was passed study data, there was a bug in
1214        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1215        were also very few tests of studied patterns with pcre_dfa_exec().
1217    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1218        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1219        on the other side of the coin, they are no longer allowed to have different
1220        names, because these cannot be distinguished in PCRE, and this has caused
1221        confusion. (This is a difference from Perl.)
1223    28. When duplicate subpattern names are present (necessarily with different
1224        numbers, as required by 27 above), and a test is made by name in a
1225        conditional pattern, either for a subpattern having been matched, or for
1226        recursion in such a pattern, all the associated numbered subpatterns are
1227        tested, and the overall condition is true if the condition is true for any
1228        one of them. This is the way Perl works, and is also more like the way
1229        testing by number works.
1232    Version 7.9 11-Apr-09
1233    ---------------------
1235    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1236        (pcretest), all targets were linked against these libraries. This included
1237        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1238        libraries. This caused unwanted dependencies to be created. This problem
1239        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1240        pcretest is linked with readline.
1242    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1243        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1244        moved outside it again, because FALSE and TRUE are already defined in AIX,
1245        but BOOL is not.
1247    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1248        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1250    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1251        hyphens) following file names and line numbers when outputting matching
1252        lines. This is not true; no spaces are inserted. I have also clarified the
1253        wording for the --colour (or --color) option.
1255    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1256        was not coloured; this is different to GNU grep, so I have changed it to be
1257        the same.
1259    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1260        each matching line was coloured. Now it goes on to look for further matches
1261        of any of the test patterns, which is the same behaviour as GNU grep.
1263    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1264        doesn't make sense to accept an empty string match in pcregrep, so I have
1265        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1266        seems to be how GNU grep behaves.
1268    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1269        start or after a newline", because the conditional assertion was not being
1270        correctly handled. The rule now is that both the assertion and what follows
1271        in the first alternative must satisfy the test.
1273    9.  If auto-callout was enabled in a pattern with a conditional group whose
1274        condition was an assertion, PCRE could crash during matching, both with
1275        pcre_exec() and pcre_dfa_exec().
1277    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1278        used for matching.
1280    11. Unicode property support in character classes was not working for
1281        characters (bytes) greater than 127 when not in UTF-8 mode.
1283    12. Added the -M command line option to pcretest.
1285    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1287    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1289    16. Added comments and documentation about mis-use of no_arg in the C++
1290        wrapper.
1292    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1293        from Martin Jerabek that uses macro names for all relevant character and
1294        string constants.
1296    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1297        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1298        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1299        these, but not everybody uses configure.
1301    19. A conditional group that had only one branch was not being correctly
1302        recognized as an item that could match an empty string. This meant that an
1303        enclosing group might also not be so recognized, causing infinite looping
1304        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1305        with the subject "ab", where knowledge that the repeated group can match
1306        nothing is needed in order to break the loop.
1308    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1309        exec(), but without supplying a callout function, matching went wrong.
1311    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1312        leak if the size of the offset vector was greater than 30. When the vector
1313        is smaller, the saved offsets during recursion go onto a local stack
1314        vector, but for larger vectors malloc() is used. It was failing to free
1315        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1316        error, in fact).
1318    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1319        heapframe that is used only when UTF-8 support is enabled. This caused no
1320        problem, but was untidy.
1322    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1323        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1324        included within another project.
1326    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1327        slightly modified by me:
1329          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1330              not building pcregrep.
1332          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1333              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1335    25. Forward references, both numeric and by name, in patterns that made use of
1336        duplicate group numbers, could behave incorrectly or give incorrect errors,
1337        because when scanning forward to find the reference group, PCRE was not
1338        taking into account the duplicate group numbers. A pattern such as
1339        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1341    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1342        the feature of a custom pointer more persuasive (as requested by a user).
1344    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1345        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1346        pre-defined.
1348    28. Added support for (*UTF8) at the start of a pattern.
1350    29. Arrange for flags added by the "release type" setting in CMake to be shown
1351        in the configuration summary.
1354    Version 7.8 05-Sep-08
1355    ---------------------
1357    1.  Replaced UCP searching code with optimized version as implemented for Ad
1358        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1359        stage table and inline lookup instead of a function, giving speed ups of 2
1360        to 5 times on some simple patterns that I tested. Permission was given to
1361        distribute the MultiStage2.py script that generates the tables (it's not in
1362        the tarball, but is in the Subversion repository).
1364    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1365        scripts.
1367    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1368        a group with a zero qualifier. The result of the study could be incorrect,
1369        or the function might crash, depending on the pattern.
1371    4.  Caseless matching was not working for non-ASCII characters in back
1372        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1373        It now works when Unicode Property Support is available.
1375    5.  In pcretest, an escape such as \x{de} in the data was always generating
1376        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1377        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1378        truncation.
1380    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1382    7.  Added two (int) casts to pcregrep when printing the difference of two
1383        pointers, in case they are 64-bit values.
1385    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1386        test 2 if it fails.
1388    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1389        and a #define of that name to empty if it is not externally set. This is to
1390        allow users of MSVC to set it if necessary.
1392    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1393        the convenience functions in the pcre_get.c source file.
1395    11. An option change at the start of a pattern that had top-level alternatives
1396        could cause overwriting and/or a crash. This command provoked a crash in
1397        some environments:
1399          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1401        This potential security problem was recorded as CVE-2008-2371.
1403    12. For a pattern where the match had to start at the beginning or immediately
1404        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1405        pcre_dfa_exec() could read past the end of the passed subject if there was
1406        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1407        pcretest so that it places the subject at the end of its malloc-ed buffer.
1409    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1410        exec() might read past the end of the data buffer in UTF-8 mode.
1412    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1413        the data contained the byte 0x85 as part of a UTF-8 character within its
1414        first line. This applied both to normal and DFA matching.
1416    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1417        /^[^d]*?$/8 failed to match "abc".
1419    16. Added a missing copyright notice to pcrecpp_internal.h.
1421    17. Make it more clear in the documentation that values returned from
1422        pcre_exec() in ovector are byte offsets, not character counts.
1424    18. Tidied a few places to stop certain compilers from issuing warnings.
1426    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1427        supplied by Stefan Weber. I made a further small update for 7.8 because
1428        there is a change of source arrangements: the pcre_searchfuncs.c module is
1429        replaced by pcre_ucd.c.
1432    Version 7.7 07-May-08
1433    ---------------------
1435    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1436        a string to a long long, pretend we don't even have a long long." This is
1437        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1439    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1440        pre-7.6 versions, which defined a global no_arg variable instead of putting
1441        it in the RE class. (See also #8 below.)
1443    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1444        Lopes.
1446    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1448        (1) The include/exclude patterns were being applied to the whole pathnames
1449            of files, instead of just to the final components.
1451        (2) If there was more than one level of directory, the subdirectories were
1452            skipped unless they satisfied the include/exclude conditions. This is
1453            inconsistent with GNU grep (and could even be seen as contrary to the
1454            pcregrep specification - which I improved to make it absolutely clear).
1455            The action now is always to scan all levels of directory, and just
1456            apply the include/exclude patterns to regular files.
1458    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1459        --exclude_dir in the tests to avoid scanning .svn directories.
1461    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1462        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1463        doesn't support NULs in patterns.
1465    7.  Added some missing "const"s to declarations of static tables in
1466        pcre_compile.c and pcre_dfa_exec.c.
1468    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1469        caused by fix #2  above. (Subsequently also a second patch to fix the
1470        first patch. And a third patch - this was a messy problem.)
1472    9.  Applied Craig's patch to remove the use of push_back().
1474    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1475        matching function regexec().
1477    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1478        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1479        references. PCRE supports relative numbers with this syntax (I don't think
1480        Oniguruma does).
1482    12. Previously, a group with a zero repeat such as (...){0} was completely
1483        omitted from the compiled regex. However, this means that if the group
1484        was called as a subroutine from elsewhere in the pattern, things went wrong
1485        (an internal error was given). Such groups are now left in the compiled
1486        pattern, with a new opcode that causes them to be skipped at execution
1487        time.
1489    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1490        to the way PCRE behaves:
1492        (a) A lone ] character is dis-allowed (Perl treats it as data).
1494        (b) A back reference to an unmatched subpattern matches an empty string
1495            (Perl fails the current match path).
1497        (c) A data ] in a character class must be notated as \] because if the
1498            first data character in a class is ], it defines an empty class. (In
1499            Perl it is not possible to have an empty class.) The empty class []
1500            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1501            The negative empty class [^] matches any one character, independently
1502            of the DOTALL setting.
1504    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1505        non-existent subpattern following a character class starting with ']' and
1506        containing () gave an internal compiling error instead of "reference to
1507        non-existent subpattern". Fortunately, when the pattern did exist, the
1508        compiled code was correct. (When scanning forwards to check for the
1509        existencd of the subpattern, it was treating the data ']' as terminating
1510        the class, so got the count wrong. When actually compiling, the reference
1511        was subsequently set up correctly.)
1513    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1514        it was being rejected as not supported by pcre_dfa_exec(), even though
1515        other assertions are supported. I have made pcre_dfa_exec() support
1516        (*FAIL).
1518    16. The implementation of 13c above involved the invention of a new opcode,
1519        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1520        cannot be changed at match time, I realized I could make a small
1521        improvement to matching performance by compiling OP_ALLANY instead of
1522        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1523        on the OP_ANY path.
1525    17. Compiling pcretest on Windows with readline support failed without the
1526        following two fixes: (1) Make the unistd.h include conditional on
1527        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1529    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1530        ncurses library to be included for pcretest when ReadLine support is
1531        requested, but also to allow for it to be overridden. This patch came from
1532        Daniel Bergström.
1534    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1535        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1536        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1537        spotting this.
1540    Version 7.6 28-Jan-08
1541    ---------------------
1543    1.  A character class containing a very large number of characters with
1544        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1545        overflow.
1547    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1548        HAVE_LONG_LONG is not defined.
1550    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1551        bring them up to date and include new features. This patch includes:
1553        - Fixed PH's badly added libz and libbz2 support.
1554        - Fixed a problem with static linking.
1555        - Added pcredemo. [But later removed - see 7 below.]
1556        - Fixed dftables problem and added an option.
1557        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1558            HAVE_LONG_LONG.
1559        - Added readline support for pcretest.
1560        - Added an listing of the option settings after cmake has run.
1562    4.  A user submitted a patch to Makefile that makes it easy to create
1563        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1564        Makefile.am that cause it to include this special target, without
1565        affecting anything else. Note that the same mingw target plus all
1566        the other distribution libraries and programs are now supported
1567        when configuring with CMake (see 6 below) instead of with
1568        Configure/Make.
1570    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1571        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1572        exported in the Windows port". It has not yet been confirmed that the patch
1573        solves the problem, but it does no harm.
1575    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1576        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1577        with CMake, and also correct the comment about stack recursion.
1579    7.  Remove the automatic building of pcredemo from the ./configure system and
1580        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1581        of a program that users should build themselves after PCRE is installed, so
1582        building it automatically is not really right. What is more, it gave
1583        trouble in some build environments.
1585    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1588    Version 7.5 10-Jan-08
1589    ---------------------
1591    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1592        values in parens when parsing an RE using the C++ wrapper."
1594    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1595        Characters greater than 255 were excluded from the class instead of being
1596        included.
1598    3.  The same bug as (2) above applied to negated POSIX classes such as
1599        [:^space:].
1601    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1602        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1603        I have changed it.
1605    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1606        first named subpattern) and a construct such as (?&a) would reference the
1607        first named subpattern whose name started with "a" (in other words, the
1608        length check was missing). Both these problems are fixed. "Subpattern name
1609        expected" is now given for (?&) (a zero-length name), and this patch also
1610        makes it give the same error for \k'' (previously it complained that that
1611        was a reference to a non-existent subpattern).
1613    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1614        this is right because (?- can be followed by option settings as well as by
1615        digits. I have, however, made the messages clearer.
1617    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1618        than the number used in the conditional) now cause a compile-time error.
1619        This is actually not compatible with Perl, which accepts such patterns, but
1620        treats the conditional as always being FALSE (as PCRE used to), but it
1621        seems to me that giving a diagnostic is better.
1623    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1624        and messages.
1626    9.  Fix two occurrences of "backslash" in comments that should have been
1627        "backspace".
1629    10. Remove two redundant lines of code that can never be obeyed (their function
1630        was moved elsewhere).
1632    11. The program that makes PCRE's Unicode character property table had a bug
1633        which caused it to generate incorrect table entries for sequences of
1634        characters that have the same character type, but are in different scripts.
1635        It amalgamated them into a single range, with the script of the first of
1636        them. In other words, some characters were in the wrong script. There were
1637        thirteen such cases, affecting characters in the following ranges:
1639          U+002b0 - U+002c1
1640          U+0060c - U+0060d
1641          U+0061e - U+00612
1642          U+0064b - U+0065e
1643          U+0074d - U+0076d
1644          U+01800 - U+01805
1645          U+01d00 - U+01d77
1646          U+01d9b - U+01dbf
1647          U+0200b - U+0200f
1648          U+030fc - U+030fe
1649          U+03260 - U+0327f
1650          U+0fb46 - U+0fbb1
1651          U+10450 - U+1049d
1653    12. The -o option (show only the matching part of a line) for pcregrep was not
1654        compatible with GNU grep in that, if there was more than one match in a
1655        line, it showed only the first of them. It now behaves in the same way as
1656        GNU grep.
1658    13. If the -o and -v options were combined for pcregrep, it printed a blank
1659        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1660        does the same. The return code can be used to tell if there were any
1661        non-matching lines.
1663    14. Added --file-offsets and --line-offsets to pcregrep.
1665    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1666        infinitely looping recursion. The bug was that positive lookaheads were not
1667        being skipped when checking for a possible empty match (negative lookaheads
1668        and both kinds of lookbehind were skipped).
1670    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1671        inclusion of <windows.h> to before rather than after the definition of
1672        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1674    17. Specifying a possessive quantifier with a specific limit for a Unicode
1675        character property caused pcre_compile() to compile bad code, which led at
1676        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1677        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1678        caused the error; without that there was no problem.
1680    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1682    19. Added --enable-pcretest-libreadline.
1684    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1685        RE::GlobalReplace(). As a result, the number of replacements returned was
1686        double what it should be. I removed one of the increments, but Craig sent a
1687        later patch that removed the other one (the right fix) and added unit tests
1688        that check the return values (which was not done before).
1690    21. Several CMake things:
1692        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1693            the names libpcre and libpcreposix, not just pcre and pcreposix.
1695        (2) The above change means that pcretest and pcregrep are now correctly
1696            linked with the newly-built libraries, not previously installed ones.
1700    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1701        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1702        UTF-8 newline character). The key issue is that the pattern starts .*;
1703        this means that the match must be either at the beginning, or after a
1704        newline. The bug was in the code for advancing after a failed match and
1705        checking that the new position followed a newline. It was not taking
1706        account of UTF-8 characters correctly.
1708    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1709        character classes. PCRE was not treating the sequence [:...:] as a
1710        character class unless the ... were all letters. Perl, however, seems to
1711        allow any characters between [: and :], though of course it rejects as
1712        unknown any "names" that contain non-letters, because all the known class
1713        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1714        for example, whereas PCRE did not - it did not recognize a POSIX character
1715        class. This seemed a bit dangerous, so the code has been changed to be
1716        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1717        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1718        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1719        Perl does, and where it didn't before.
1721    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1722        Windows environments %n is disabled by default.
1725    Version 7.4 21-Sep-07
1726    ---------------------
1728    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1729        means that a class such as [\s] counted as "explicit reference to CR or
1730        LF". That isn't really right - the whole point of the change was to try to
1731        help when there was an actual mention of one of the two characters. So now
1732        the change happens only if \r or \n (or a literal CR or LF) character is
1733        encountered.
1735    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1736        of both had grown to the point where there were only 3 bits left.
1737        Fortunately, there was spare space in the data structure, and so I have
1738        moved the internal flags into a new 16-bit field to free up more option
1739        bits.
1741    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1742        but did not set the internal JCHANGED flag - either of these is enough to
1743        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1744        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1745        start sets both bits.
1747    4.  Added options (at build time, compile time, exec time) to change \R from
1748        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1750    5.  doc/pcresyntax.html was missing from the distribution.
1752    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1753        compatibility, even though it is no longer used.
1755    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1756        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1757        windows.h file is present (where different names are used). [This was
1758        reversed later after testing - see 16 below.]
1760    8.  Changed all #include <config.h> to #include "config.h". There were also
1761        some further <pcre.h> cases that I changed to "pcre.h".
1763    9.  When pcregrep was used with the --colour option, it missed the line ending
1764        sequence off the lines that it output.
1766    10. It was pointed out to me that arrays of string pointers cause lots of
1767        relocations when a shared library is dynamically loaded. A technique of
1768        using a single long string with a table of offsets can drastically reduce
1769        these. I have refactored PCRE in four places to do this. The result is
1770        dramatic:
1772          Originally:                          290
1773          After changing UCP table:            187
1774          After changing error message table:   43
1775          After changing table of "verbs"       36
1776          After changing table of Posix names   22
1778        Thanks to the folks working on Gregex for glib for this insight.
1780    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1781        unicode-properties was also set.
1783    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1785    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1786        checked only for CRLF.
1788    14. Added casts to pcretest.c to avoid compiler warnings.
1790    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1792    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1793        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1794        entirely. This removes changes made in 7 above.
1796    17. The CMake files have been updated, and there is now more information about
1797        building with CMake in the NON-UNIX-USE document.
1800    Version 7.3 28-Aug-07
1801    ---------------------
1803     1. In the rejigging of the build system that eventually resulted in 7.1, the
1804        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1805        brackets there is not right, since it causes compilers to look for an
1806        installed pcre.h, not the version that is in the source that is being
1807        compiled (which of course may be different). I have changed it back to:
1809          #include "pcre.h"
1811        I have a vague recollection that the change was concerned with compiling in
1812        different directories, but in the new build system, that is taken care of
1813        by the VPATH setting the Makefile.
1815     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1816        when the subject happened to end in the byte 0x85 (e.g. if the last
1817        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1818        characters but of course it shouldn't be taken as a newline when it is part
1819        of another character. The bug was that, for an unlimited repeat of . in
1820        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1821        characters when looking for a newline.
1823     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1825     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1826        in debug output.
1828     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1829        long printing in the pcrecpp unittest when running under MinGW.
1831     6. ESC_K was left out of the EBCDIC table.
1833     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1834        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1835        limit also applies to "virtual nesting" when a pattern is recursive, and in
1836        this case 1000 isn't so big. I have been able to remove this limit at the
1837        expense of backing off one optimization in certain circumstances. Normally,
1838        when pcre_exec() would call its internal match() function recursively and
1839        immediately return the result unconditionally, it uses a "tail recursion"
1840        feature to save stack. However, when a subpattern that can match an empty
1841        string has an unlimited repetition quantifier, it no longer makes this
1842        optimization. That gives it a stack frame in which to save the data for
1843        checking that an empty string has been matched. Previously this was taken
1844        from the 1000-entry workspace that had been reserved. So now there is no
1845        explicit limit, but more stack is used.
1847     8. Applied Daniel's patches to solve problems with the import/export magic
1848        syntax that is required for Windows, and which was going wrong for the
1849        pcreposix and pcrecpp parts of the library. These were overlooked when this
1850        problem was solved for the main library.
1852     9. There were some crude static tests to avoid integer overflow when computing
1853        the size of patterns that contain repeated groups with explicit upper
1854        limits. As the maximum quantifier is 65535, the maximum group length was
1855        set at 30,000 so that the product of these two numbers did not overflow a
1856        32-bit integer. However, it turns out that people want to use groups that
1857        are longer than 30,000 bytes (though not repeat them that many times).
1858        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1859        made it possible to implement the integer overflow checks in a much more
1860        dynamic way, which I have now done. The artificial limitation on group
1861        length has been removed - we now have only the limit on the total length of
1862        the compiled pattern, which depends on the LINK_SIZE setting.
1864    10. Fixed a bug in the documentation for get/copy named substring when
1865        duplicate names are permitted. If none of the named substrings are set, the
1866        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1867        empty string.
1869    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1870        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1871        because the ] is interpreted as the first data character and the
1872        terminating ] is not found. PCRE has been made compatible with Perl in this
1873        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1874        cause memory overwriting.
1876    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1877        string has been matched (to stop an infinite loop). It was not recognizing
1878        a conditional subpattern that could match an empty string if that
1879        subpattern was within another subpattern. For example, it looped when
1880        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1881        condition was not nested. This bug has been fixed.
1883    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1884        past the start of the subject in the presence of bytes with the top bit
1885        set, for example "\x8aBCD".
1887    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1888        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1890    14. Optimized (?!) to (*FAIL).
1892    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1893        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1894        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1895        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1896        does: it's just the validity check that is more restrictive.
1898    16. Inserted checks for integer overflows during escape sequence (backslash)
1899        processing, and also fixed erroneous offset values for syntax errors during
1900        backslash processing.
1902    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1903        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1905    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1906        caused an overrun.
1908    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1909        something other than just ASCII characters) inside a group that had an
1910        unlimited repeat caused a loop at compile time (while checking to see
1911        whether the group could match an empty string).
1913    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1914        [\P{Any}] did so. (Error in the code for printing property names.)
1916    21. An orphan \E inside a character class could cause a crash.
1918    22. A repeated capturing bracket such as (A)? could cause a wild memory
1919        reference during compilation.
1921    23. There are several functions in pcre_compile() that scan along a compiled
1922        expression for various reasons (e.g. to see if it's fixed length for look
1923        behind). There were bugs in these functions when a repeated \p or \P was
1924        present in the pattern. These operators have additional parameters compared
1925        with \d, etc, and these were not being taken into account when moving along
1926        the compiled data. Specifically:
1928        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1929            length.
1931        (b) An item such as \pL+ within a repeated group could cause crashes or
1932            loops.
1934        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1935            "reference to non-existent subpattern" error.
1937        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1939    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1940        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1942    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1944    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1945        character were causing crashes (broken optimization).
1947    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1948        \p or \P) caused a compile-time loop.
1950    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1951        break. For example, the unstudied pattern [\r\n]A does not match the string
1952        "\r\nA" because change 7.0/46 below moves the current point on by two
1953        characters after failing to match at the start. However, the pattern \nA
1954        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1955        the same is true. There doesn't seem any very clean way out of this, but
1956        what I have chosen to do makes the common cases work: PCRE now takes note
1957        of whether there can be an explicit match for \r or \n anywhere in the
1958        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1959        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1960        pattern has explicit CR or LF references.
1962    29. Added (*CR) etc for changing newline setting at start of pattern.
1965    Version 7.2 19-Jun-07
1966  ---------------------  ---------------------
1968   1. If the fr_FR locale cannot be found for test 3, try the "french" locale,   1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
# Line 40  Version 7.2 13-June-07 Line 2001  Version 7.2 13-June-07
2002      (f) \g{name} is another synonym - part of Perl 5.10's unification of      (f) \g{name} is another synonym - part of Perl 5.10's unification of
2003          reference syntax.          reference syntax.
2005      (g) (?| introduces a group in which the numbering of parentheses in each      (g) (?| introduces a group in which the numbering of parentheses in each
2006          alternative starts with the same number.          alternative starts with the same number.
2008      (h) \h, \H, \v, and \V match horizontal and vertical whitespace.      (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2010   7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and   7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
# Line 59  Version 7.2 13-June-07 Line 2020  Version 7.2 13-June-07
2020      bit of new cunning has reduced the workspace needed for groups with      bit of new cunning has reduced the workspace needed for groups with
2021      alternatives. The 1000-alternative test pattern now uses 12 bytes of      alternatives. The 1000-alternative test pattern now uses 12 bytes of
2022      workspace instead of running out of the 4096 that are available.      workspace instead of running out of the 4096 that are available.
2024  10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.  10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2026  11. Applied patch from Google to remove an optimization that didn't quite work.  11. Applied patch from Google to remove an optimization that didn't quite work.
2027      The report of the bug said:      The report of the bug said:
2029        pcrecpp::RE("a*").FullMatch("aaa") matches, while        pcrecpp::RE("a*").FullMatch("aaa") matches, while
2030        pcrecpp::RE("a*?").FullMatch("aaa") does not, and        pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2031        pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.        pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2033    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2034        it matched the wrong number of bytes.
2037  Version 7.1 24-Apr-07  Version 7.1 24-Apr-07

Removed from v.181  
changed lines
  Added in v.947

  ViewVC Help
Powered by ViewVC 1.1.5