/[pcre]/code/trunk/ChangeLog
ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 116 by ph10, Fri Mar 9 15:23:02 2007 UTC revision 970 by zherczeg, Sun May 27 05:28:19 2012 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 7.1 05-Mar-07  Version 8.31
5    -----------------------------
6    
7    1.  Fixing a wrong JIT test case and some compiler warnings.
8    
9    2.  Removed a bashism from the RunTest script.
10    
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)".
14    
15    4.  Partial matching support is added to the JIT compiler.
16    
17    5.  Fixed several bugs concerned with partial matching of items that consist
18        of more than one character:
19    
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.
23    
24        (b) \R did not give a hard partial match if \r was found at the end of the
25            subject.
26    
27        (c) \X did not give a hard partial match after matching one or more
28            characters at the end of the subject.
29    
30        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
31            a partial match for the string "\r".
32    
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.
35    
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.
39    
40    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
41        or /S+[+] with a digit between 1 and 7.
42    
43    8.  OP_NOT now supports any UTF character not just single-byte ones.
44    
45    9.  (*MARK) control verb is now supported by the JIT compiler.
46    
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.)
49    
50    11. Add PCRE_INFO_MAXLOOKBEHIND.
51    
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.
57    
58    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
59        call.
60    
61    14. As documented, (*COMMIT) is now confined to within a positive assertion.
62    
63    15. It is now possible to link pcretest with libedit as an alternative to
64        libreadline.
65    
66    16. (*COMMIT) control verb is now supported by the JIT compiler.
67    
68    17. The Unicode data tables have been updated to Unicode 6.1.0.
69    
70    18. Added --file-list option to pcregrep.
71    
72    19. Added binary file support to pcregrep, including the -a, --binary-files,
73        -I, and --text options.
74    
75    20. The madvise function is renamed for posix_madvise for QNX compatibility
76        reasons. Fixed by Giuseppe D'Angelo.
77    
78    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
79    
80    22. Changed the option for creating symbolic links for 16-bit man pages from
81        -s to -sf so that re-installing does not cause issues.
82    
83    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
84    
85    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
86        matches in certain environments (the workspace was not being correctly
87        retained). Also added to pcre_dfa_exec() a simple plausibility check on
88        some of the workspace data at the beginning of a restart.
89    
90    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
91        was not doing so when it should - probably a typo introduced by SVN 528
92        (change 8.10/14).
93    
94    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
95        \w+ when the character tables indicated that \x{c4} was a word character.
96        There were several related cases, all because the tests for doing a table
97        lookup were testing for characters less than 127 instead of 255.
98    
99    27. If a pattern contains capturing parentheses that are not used in a match,
100        their slots in the ovector are set to -1. For those that are higher than
101        any matched groups, this happens at the end of processing. In the case when
102        there were back references that the ovector was too small to contain
103        (causing temporary malloc'd memory to be used during matching), and the
104        highest capturing number was not used, memory off the end of the ovector
105        was incorrectly being set to -1. (It was using the size of the temporary
106        memory instead of the true size.)
107    
108    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
109        ovector size, it uses memory at the end of the block that it has got.
110    
111    29. Check for an overlong MARK name and give an error at compile time. The
112        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
113    
114    30. JIT compiler update.
115    
116    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
117        Rill for the patch.
118    
119    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
120    
121    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
122    
123    
124    Version 8.30 04-February-2012
125    -----------------------------
126    
127    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
128        name is defined in ctype.h.
129    
130    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
131        only in quite long subpatterns.
132    
133    3.  Removed the function pcre_info(), which has been obsolete and deprecated
134        since it was replaced by pcre_fullinfo() in February 2000.
135    
136    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
137        match a (*MARK), and the match failed at the start of the subject, a
138        reference to memory before the start of the subject could occur. This bug
139        was introduced by fix 17 of release 8.21.
140    
141    5.  A reference to an unset group with zero minimum repetition was giving
142        totally wrong answers (in non-JavaScript-compatibility mode). For example,
143        /(another)?(\1?)test/ matched against "hello world test". This bug was
144        introduced in release 8.13.
145    
146    6.  Add support for 16-bit character strings (a large amount of work involving
147        many changes and refactorings).
148    
149    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
150        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
151        from a file.
152    
153    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
154        rounding is not applied in this particular case).
155    
156    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
157        if they appear, or are escaped, in patterns.
158    
159    10. Get rid of a number of -Wunused-but-set-variable warnings.
160    
161    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
162        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
163        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
164        also returns the mark "x". This bug applied to capturing parentheses,
165        non-capturing parentheses, and atomic parentheses. It also applied to some
166        assertions.
167    
168    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
169        information out of configure.ac instead of relying on pcre.h.generic, which
170        is not stored in the repository.
171    
172    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
173        -lreadline.
174    
175    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
176    
177    15. Applied Graycode's patch to put the top-level frame on the stack rather
178        than the heap when not using the stack for recursion. This gives a
179        performance improvement in many cases when recursion is not deep.
180    
181    16. Experimental code added to "pcretest -C" to output the stack frame size.
182    
183    
184    Version 8.21 12-Dec-2011
185    ------------------------
186    
187    1.  Updating the JIT compiler.
188    
189    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
190        are added as well.
191    
192    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
193        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
194        calling _pcre_jit_exec. Some extra comments are added.
195    
196    4.  (*MARK) settings inside atomic groups that do not contain any capturing
197        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
198        was introduced by change 18 for 8.20.
199    
200    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
201        ECMA-262 standard.
202    
203    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
204        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
205        This bug was probably introduced by change 9 of 8.13.
206    
207    7.  While fixing 6 above, I noticed that a number of other items were being
208        incorrectly rejected as "not fixed length". This arose partly because newer
209        opcodes had not been added to the fixed-length checking code. I have (a)
210        corrected the bug and added tests for these items, and (b) arranged for an
211        error to occur if an unknown opcode is encountered while checking for fixed
212        length instead of just assuming "not fixed length". The items that were
213        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
214        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
215        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
216    
217    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
218        being incorrectly compiled and would have given unpredicatble results.
219    
220    9.  A possessively repeated subpattern with minimum repeat count greater than
221        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
222        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
223        the first (A) could occur when it should not.
224    
225    10. Add a cast and remove a redundant test from the code.
226    
227    11. JIT should use pcre_malloc/pcre_free for allocation.
228    
229    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
230        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
231        is anything other than /usr, -L is still shown).
232    
233    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
234    
235    14. Perl does not support \N without a following name in a [] class; PCRE now
236        also gives an error.
237    
238    15. If a forward reference was repeated with an upper limit of around 2000,
239        it caused the error "internal error: overran compiling workspace". The
240        maximum number of forward references (including repeats) was limited by the
241        internal workspace, and dependent on the LINK_SIZE. The code has been
242        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
243        the default depends on LINK_SIZE. There is a new upper limit (for safety)
244        of around 200,000 forward references. While doing this, I also speeded up
245        the filling in of repeated forward references.
246    
247    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
248        incorrectly expecting the subject to contain another "a" after the start.
249    
250    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
251        in the match, the SKIP should be ignored. This was not happening; instead
252        the SKIP was being treated as NOMATCH. For patterns such as
253        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
254        tested.
255    
256    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
257        now much more compatible with Perl, in particular in cases where the result
258        is a non-match for a non-anchored pattern. For example, if
259        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
260        "m", where previously it did not return a name. A side effect of this
261        change is that for partial matches, the last encountered mark name is
262        returned, as for non matches. A number of tests that were previously not
263        Perl-compatible have been moved into the Perl-compatible test files. The
264        refactoring has had the pleasing side effect of removing one argument from
265        the match() function, thus reducing its stack requirements.
266    
267    19. If the /S+ option was used in pcretest to study a pattern using JIT,
268        subsequent uses of /S (without +) incorrectly behaved like /S+.
269    
270    21. Retrieve executable code size support for the JIT compiler and fixing
271        some warnings.
272    
273    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
274        not work when the shorter character appeared right at the end of the
275        subject string.
276    
277    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
278        systems.
279    
280    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
281        output it when the /M option is used in pcretest.
282    
283    25. The CheckMan script was not being included in the distribution. Also, added
284        an explicit "perl" to run Perl scripts from the PrepareRelease script
285        because this is reportedly needed in Windows.
286    
287    26. If study data was being save in a file and studying had not found a set of
288        "starts with" bytes for the pattern, the data written to the file (though
289        never used) was taken from uninitialized memory and so caused valgrind to
290        complain.
291    
292    27. Updated RunTest.bat as provided by Sheri Pierce.
293    
294    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
295    
296    29. Computation of memory usage for the table of capturing group names was
297        giving an unnecessarily large value.
298    
299    
300    Version 8.20 21-Oct-2011
301    ------------------------
302    
303    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
304        a POSIX class. After further experiments with Perl, which convinced me that
305        Perl has bugs and confusions, a closing square bracket is no longer allowed
306        in a POSIX name. This bug also affected patterns with classes that started
307        with full stops.
308    
309    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
310        captured substring, but while checking the failing first alternative,
311        substring 1 is temporarily captured. If the output vector supplied to
312        pcre_exec() was not big enough for this capture, the yield of the function
313        was still zero ("insufficient space for captured substrings"). This cannot
314        be totally fixed without adding another stack variable, which seems a lot
315        of expense for a edge case. However, I have improved the situation in cases
316        such as /(a)(b)x|abc/ matched against "abc", where the return code
317        indicates that fewer than the maximum number of slots in the ovector have
318        been set.
319    
320    3.  Related to (2) above: when there are more back references in a pattern than
321        slots in the output vector, pcre_exec() uses temporary memory during
322        matching, and copies in the captures as far as possible afterwards. It was
323        using the entire output vector, but this conflicts with the specification
324        that only 2/3 is used for passing back captured substrings. Now it uses
325        only the first 2/3, for compatibility. This is, of course, another edge
326        case.
327    
328    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
329        main code base, and can be used by building with --enable-jit. When this is
330        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
331        runtime --no-jit option is given.
332    
333    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
334        ovector, the return from the function was zero, implying that there were
335        other matches that did not fit. The correct "exactly full" value is now
336        returned.
337    
338    6.  If a subpattern that was called recursively or as a subroutine contained
339        (*PRUNE) or any other control that caused it to give a non-standard return,
340        invalid errors such as "Error -26 (nested recursion at the same subject
341        position)" or even infinite loops could occur.
342    
343    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
344        computing the minimum length on reaching *ACCEPT, and so ended up with the
345        wrong value of 1 rather than 0. Further investigation indicates that
346        computing a minimum subject length in the presence of *ACCEPT is difficult
347        (think back references, subroutine calls), and so I have changed the code
348        so that no minimum is registered for a pattern that contains *ACCEPT.
349    
350    8.  If (*THEN) was present in the first (true) branch of a conditional group,
351        it was not handled as intended. [But see 16 below.]
352    
353    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
354        Sheri Pierce.
355    
356    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
357        the first byte in a match must be "a".
358    
359    11. Change 17 for 8.13 increased the recursion depth for patterns like
360        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
361        pattern contains any instances of (*THEN). If it does not, the old
362        optimizations are restored. It would be nice to do this on a per-group
363        basis, but at the moment that is not feasible.
364    
365    12. In some environments, the output of pcretest -C is CRLF terminated. This
366        broke RunTest's code that checks for the link size. A single white space
367        character after the value is now allowed for.
368    
369    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
370        For "fr", it uses the Windows-specific input and output files.
371    
372    14. If (*THEN) appeared in a group that was called recursively or as a
373        subroutine, it did not work as intended. [But see next item.]
374    
375    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
376        pattern fragments (but not containing any | characters). If A and B are
377        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
378        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
379        D. In other words, Perl considers parentheses that do not contain any |
380        characters to be part of a surrounding alternative, whereas PCRE was
381        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
382        differently. PCRE now behaves in the same way as Perl, except in the case
383        of subroutine/recursion calls such as (?1) which have in any case always
384        been different (but PCRE had them first :-).
385    
386    16. Related to 15 above: Perl does not treat the | in a conditional group as
387        creating alternatives. Such a group is treated in the same way as an
388        ordinary group without any | characters when processing (*THEN). PCRE has
389        been changed to match Perl's behaviour.
390    
391    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
392        RunGrepTest script failed.
393    
394    18. Change 22 for version 13 caused atomic groups to use more stack. This is
395        inevitable for groups that contain captures, but it can lead to a lot of
396        stack use in large patterns. The old behaviour has been restored for atomic
397        groups that do not contain any capturing parentheses.
398    
399    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
400        suppress the check for a minimum subject length at run time. (If it was
401        given to pcre_exec() or pcre_dfa_exec() it did work.)
402    
403    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
404        fail to work when decoding hex characters in data strings in EBCDIC
405        environments.
406    
407    21. It appears that in at least one Mac OS environment, the isxdigit() function
408        is implemented as a macro that evaluates to its argument more than once,
409        contravening the C 90 Standard (I haven't checked a later standard). There
410        was an instance in pcretest which caused it to go wrong when processing
411        \x{...} escapes in subject strings. The has been rewritten to avoid using
412        things like p++ in the argument of isxdigit().
413    
414    
415    Version 8.13 16-Aug-2011
416    ------------------------
417    
418    1.  The Unicode data tables have been updated to Unicode 6.0.0.
419    
420    2.  Two minor typos in pcre_internal.h have been fixed.
421    
422    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
423        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
424        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
425    
426    4.  There were a number of related bugs in the code for matching backrefences
427        caselessly in UTF-8 mode when codes for the characters concerned were
428        different numbers of bytes. For example, U+023A and U+2C65 are an upper
429        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
430        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
431        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
432        2-byte code at the end of the subject (it thought there wasn't enough data
433        left).
434    
435    5.  Comprehensive information about what went wrong is now returned by
436        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
437        as the output vector has at least 2 elements. The offset of the start of
438        the failing character and a reason code are placed in the vector.
439    
440    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
441        now returned is for the first byte of the failing character, instead of the
442        last byte inspected. This is an incompatible change, but I hope it is small
443        enough not to be a problem. It makes the returned offset consistent with
444        pcre_exec() and pcre_dfa_exec().
445    
446    7.  pcretest now gives a text phrase as well as the error number when
447        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
448        failure, the offset and reason code are output.
449    
450    8.  When \R was used with a maximizing quantifier it failed to skip backwards
451        over a \r\n pair if the subsequent match failed. Instead, it just skipped
452        back over a single character (\n). This seems wrong (because it treated the
453        two characters as a single entity when going forwards), conflicts with the
454        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
455        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
456        has been changed.
457    
458    9.  Some internal refactoring has changed the processing so that the handling
459        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
460        time (the PCRE_DOTALL option was changed this way some time ago: version
461        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
462        which was always a bit of a fudge. It also means that there is one less
463        argument for the match() function, which reduces its stack requirements
464        slightly. This change also fixes an incompatibility with Perl: the pattern
465        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
466    
467    10. More internal refactoring has drastically reduced the number of recursive
468        calls to match() for possessively repeated groups such as (abc)++ when
469        using pcre_exec().
470    
471    11. While implementing 10, a number of bugs in the handling of groups were
472        discovered and fixed:
473    
474        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
475        (a|)*(?1) gave a compile-time internal error.
476        ((a|)+)+  did not notice that the outer group could match an empty string.
477        (^a|^)+   was not marked as anchored.
478        (.*a|.*)+ was not marked as matching at start or after a newline.
479    
480    12. Yet more internal refactoring has removed another argument from the match()
481        function. Special calls to this function are now indicated by setting a
482        value in a variable in the "match data" data block.
483    
484    13. Be more explicit in pcre_study() instead of relying on "default" for
485        opcodes that mean there is no starting character; this means that when new
486        ones are added and accidentally left out of pcre_study(), testing should
487        pick them up.
488    
489    14. The -s option of pcretest has been documented for ages as being an old
490        synonym of -m (show memory usage). I have changed it to mean "force study
491        for every regex", that is, assume /S for every regex. This is similar to -i
492        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
493        using it. It makes it easier to run collections of tests with and without
494        study enabled, and thereby test pcre_study() more easily. All the standard
495        tests are now run with and without -s (but some patterns can be marked as
496        "never study" - see 20 below).
497    
498    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
499        restoration of the capturing data to the outer values was not happening
500        correctly.
501    
502    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
503        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
504        pattern had matched an empty string, and so incorrectly returned a no
505        match.
506    
507    17. There was optimizing code for the last branch of non-capturing parentheses,
508        and also for the obeyed branch of a conditional subexpression, which used
509        tail recursion to cut down on stack usage. Unfortunately, now that there is
510        the possibility of (*THEN) occurring in these branches, tail recursion is
511        no longer possible because the return has to be checked for (*THEN). These
512        two optimizations have therefore been removed. [But see 8.20/11 above.]
513    
514    18. If a pattern containing \R was studied, it was assumed that \R always
515        matched two bytes, thus causing the minimum subject length to be
516        incorrectly computed because \R can also match just one byte.
517    
518    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
519        was incorrectly computed.
520    
521    20. If /S is present twice on a test pattern in pcretest input, it now
522        *disables* studying, thereby overriding the use of -s on the command line
523        (see 14 above). This is necessary for one or two tests to keep the output
524        identical in both cases.
525    
526    21. When (*ACCEPT) was used in an assertion that matched an empty string and
527        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
528    
529    22. When an atomic group that contained a capturing parenthesis was
530        successfully matched, but the branch in which it appeared failed, the
531        capturing was not being forgotten if a higher numbered group was later
532        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
533        group 1 to "a", when in fact it should be unset. This applied to multi-
534        branched capturing and non-capturing groups, repeated or not, and also to
535        positive assertions (capturing in negative assertions does not happen
536        in PCRE) and also to nested atomic groups.
537    
538    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
539        subject after a captured substring, to make it easier to tell which of a
540        number of identical substrings has been captured.
541    
542    24. The way atomic groups are processed by pcre_exec() has been changed so that
543        if they are repeated, backtracking one repetition now resets captured
544        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
545        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
546        "aaa". Previously, it would have been "a". As part of this code
547        refactoring, the way recursive calls are handled has also been changed.
548    
549    25. If an assertion condition captured any substrings, they were not passed
550        back unless some other capturing happened later. For example, if
551        (?(?=(a))a) was matched against "a", no capturing was returned.
552    
553    26. When studying a pattern that contained subroutine calls or assertions,
554        the code for finding the minimum length of a possible match was handling
555        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
556        group 1 called group 2 while simultaneously a separate group 2 called group
557        1). A stack overflow occurred in this case. I have fixed this by limiting
558        the recursion depth to 10.
559    
560    27. Updated RunTest.bat in the distribution to the version supplied by Tom
561        Fortmann. This supports explicit test numbers on the command line, and has
562        argument validation and error reporting.
563    
564    28. An instance of \X with an unlimited repeat could fail if at any point the
565        first character it looked at was a mark character.
566    
567    29. Some minor code refactoring concerning Unicode properties and scripts
568        should reduce the stack requirement of match() slightly.
569    
570    30. Added the '=' option to pcretest to check the setting of unused capturing
571        slots at the end of the pattern, which are documented as being -1, but are
572        not included in the return count.
573    
574    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
575        compiled something random. Now it gives a compile-time error (as does
576        Perl).
577    
578    32. A *MARK encountered during the processing of a positive assertion is now
579        recorded and passed back (compatible with Perl).
580    
581    33. If --only-matching or --colour was set on a pcregrep call whose pattern
582        had alternative anchored branches, the search for a second match in a line
583        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
584        matched the line "0102" twice. The same bug affected patterns that started
585        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
586        twice.
587    
588    34. Previously, PCRE did not allow quantification of assertions. However, Perl
589        does, and because of capturing effects, quantifying parenthesized
590        assertions may at times be useful. Quantifiers are now allowed for
591        parenthesized assertions.
592    
593    35. A minor code tidy in pcre_compile() when checking options for \R usage.
594    
595    36. \g was being checked for fancy things in a character class, when it should
596        just be a literal "g".
597    
598    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
599        appearance of a nested POSIX class supersedes an apparent external class.
600        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
601        unescaped square brackets may also appear as part of class names. For
602        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
603        more like Perl. (But see 8.20/1 above.)
604    
605    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
606        was because it thought it was \N{name}, which is not supported).
607    
608    39. Add minix to OS list not supporting the -S option in pcretest.
609    
610    40. PCRE tries to detect cases of infinite recursion at compile time, but it
611        cannot analyze patterns in sufficient detail to catch mutual recursions
612        such as ((?1))((?2)). There is now a runtime test that gives an error if a
613        subgroup is called recursively as a subpattern for a second time at the
614        same position in the subject string. In previous releases this might have
615        been caught by the recursion limit, or it might have run out of stack.
616    
617    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
618        happen only once. PCRE was, however incorrectly giving a compile time error
619        "recursive call could loop indefinitely" because it cannot analyze the
620        pattern in sufficient detail. The compile time test no longer happens when
621        PCRE is compiling a conditional subpattern, but actual runaway loops are
622        now caught at runtime (see 40 above).
623    
624    42. It seems that Perl allows any characters other than a closing parenthesis
625        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
626        has been changed to be the same.
627    
628    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
629        as not to get warnings when autogen.sh is called. Also changed
630        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
631    
632    44. To help people who use pcregrep to scan files containing exceedingly long
633        lines, the following changes have been made:
634    
635        (a) The default value of the buffer size parameter has been increased from
636            8K to 20K. (The actual buffer used is three times this size.)
637    
638        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
639            PCRE is built.
640    
641        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
642            to be set at run time.
643    
644        (d) Numerical values in pcregrep options can be followed by K or M, for
645            example --buffer-size=50K.
646    
647        (e) If a line being scanned overflows pcregrep's buffer, an error is now
648            given and the return code is set to 2.
649    
650    45. Add a pointer to the latest mark to the callout data block.
651    
652    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
653        partial match of an empty string instead of no match. This was specific to
654        the use of ".".
655    
656    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
657        complete match instead of a partial match. This bug was dependent on both
658        the PCRE_UTF8 and PCRE_DOTALL options being set.
659    
660    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
661        starting byte set, because \b was not being ignored.
662    
663    
664    Version 8.12 15-Jan-2011
665    ------------------------
666    
667    1.  Fixed some typos in the markup of the man pages, and wrote a script that
668        checks for such things as part of the documentation building process.
669    
670    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
671        --match-limit and --recursion-limit options (added for 8.11). In
672        particular, this made one of the standard tests fail. (The integer value
673        went into the wrong half of a long int.)
674    
675    3.  If the --colour option was given to pcregrep with -v (invert match), it
676        did strange things, either producing crazy output, or crashing. It should,
677        of course, ignore a request for colour when reporting lines that do not
678        match.
679    
680    4.  Another pcregrep bug caused similar problems if --colour was specified with
681        -M (multiline) and the pattern match finished with a line ending.
682    
683    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
684        matched in multiline mode, the following line was shown as part of the
685        match. This seems wrong, so I have changed it.
686    
687    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
688        the check for further matches in the same line (so they could be coloured)
689        to overrun the end of the current line. If another match was found, it was
690        incorrectly shown (and then shown again when found in the next line).
691    
692    7.  If pcregrep was compiled under Windows, there was a reference to the
693        function pcregrep_exit() before it was defined. I am assuming this was
694        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
695        reported by a user. I've moved the definition above the reference.
696    
697    
698    Version 8.11 10-Dec-2010
699    ------------------------
700    
701    1.  (*THEN) was not working properly if there were untried alternatives prior
702        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
703        backtracked to try for "b" instead of moving to the next alternative branch
704        at the same level (in this case, to look for "c"). The Perl documentation
705        is clear that when (*THEN) is backtracked onto, it goes to the "next
706        alternative in the innermost enclosing group".
707    
708    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
709        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
710        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
711        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
712        (*THEN).
713    
714    3.  If \s appeared in a character class, it removed the VT character from
715        the class, even if it had been included by some previous item, for example
716        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
717        of \s, but is part of the POSIX "space" class.)
718    
719    4.  A partial match never returns an empty string (because you can always
720        match an empty string at the end of the subject); however the checking for
721        an empty string was starting at the "start of match" point. This has been
722        changed to the "earliest inspected character" point, because the returned
723        data for a partial match starts at this character. This means that, for
724        example, /(?<=abc)def/ gives a partial match for the subject "abc"
725        (previously it gave "no match").
726    
727    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
728        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
729        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
730        has an implication that the given string is incomplete (because a partial
731        match is preferred over a full match). For this reason, these items now
732        give a partial match in this situation. [Aside: previously, the one case
733        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
734        match rather than a full match, which was wrong by the old rules, but is
735        now correct.]
736    
737    6.  There was a bug in the handling of #-introduced comments, recognized when
738        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
739        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
740        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
741        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
742        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
743        places in pcre_compile().
744    
745    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
746        comments when looking ahead for named forward references to subpatterns,
747        the only newline sequence it recognized was NL. It now handles newlines
748        according to the set newline convention.
749    
750    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
751        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
752        cater for a lack of strerror(). These oversights have been fixed.
753    
754    9.  Added --match-limit and --recursion-limit to pcregrep.
755    
756    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
757    
758    11. When the -o option was used, pcregrep was setting a return code of 1, even
759        when matches were found, and --line-buffered was not being honoured.
760    
761    12. Added an optional parentheses number to the -o and --only-matching options
762        of pcregrep.
763    
764    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
765        can match an empty string. The code to do it in pcretest and pcredemo
766        needed fixing:
767    
768        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
769            only one byte after an empty string match just before CRLF (this case
770            just got forgotten; "any" and "anycrlf" were OK).
771    
772        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
773            mode when an empty string match preceded an ASCII character followed by
774            a non-ASCII character. (The code for advancing by one character rather
775            than one byte was nonsense.)
776    
777        (c) The pcredemo.c sample program did not have any code at all to handle
778            the cases when CRLF is a valid newline sequence.
779    
780    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
781        as a starting offset was within the subject string. There is now a new
782        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
783        negative or greater than the length of the string. In order to test this,
784        pcretest is extended to allow the setting of negative starting offsets.
785    
786    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
787        starting offset points to the beginning of a UTF-8 character was
788        unnecessarily clumsy. I tidied it up.
789    
790    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
791        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
792    
793    17. Nobody had reported that the --include_dir option, which was added in
794        release 7.7 should have been called --include-dir (hyphen, not underscore)
795        for compatibility with GNU grep. I have changed it to --include-dir, but
796        left --include_dir as an undocumented synonym, and the same for
797        --exclude-dir, though that is not available in GNU grep, at least as of
798        release 2.5.4.
799    
800    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
801        characters from a string of bytes) have been redefined so as not to use
802        loops, in order to improve performance in some environments. At the same
803        time, I abstracted some of the common code into auxiliary macros to save
804        repetition (this should not affect the compiled code).
805    
806    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
807        compile-time error is now given if \c is not followed by an ASCII
808        character, that is, a byte less than 128. (In EBCDIC mode, the code is
809        different, and any byte value is allowed.)
810    
811    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
812        START_OPTIMIZE option, which is now allowed at compile time - but just
813        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
814        to pcregrep and other applications that have no direct access to PCRE
815        options. The new /Y option in pcretest sets this option when calling
816        pcre_compile().
817    
818    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
819        back references. Groups containing recursive back references were forced to
820        be atomic by that change, but in the case of named groups, the amount of
821        memory required was incorrectly computed, leading to "Failed: internal
822        error: code overflow". This has been fixed.
823    
824    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
825        pcretest.c, to avoid build problems in some Borland environments.
826    
827    
828    Version 8.10 25-Jun-2010
829    ------------------------
830    
831    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
832        THEN.
833    
834    2.  (*ACCEPT) was not working when inside an atomic group.
835    
836    3.  Inside a character class, \B is treated as a literal by default, but
837        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
838        causes the error). The code is unchanged, but I tidied the documentation.
839    
840    4.  Inside a character class, PCRE always treated \R and \X as literals,
841        whereas Perl faults them if its -w option is set. I have changed PCRE so
842        that it faults them when PCRE_EXTRA is set.
843    
844    5.  Added support for \N, which always matches any character other than
845        newline. (It is the same as "." when PCRE_DOTALL is not set.)
846    
847    6.  When compiling pcregrep with newer versions of gcc which may have
848        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
849        declared with attribute warn_unused_result" were given. Just casting the
850        result to (void) does not stop the warnings; a more elaborate fudge is
851        needed. I've used a macro to implement this.
852    
853    7.  Minor change to pcretest.c to avoid a compiler warning.
854    
855    8.  Added four artifical Unicode properties to help with an option to make
856        \s etc use properties (see next item). The new properties are: Xan
857        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
858    
859    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
860        use Unicode properties. (*UCP) at the start of a pattern can be used to set
861        this option. Modified pcretest to add /W to test this facility. Added
862        REG_UCP to make it available via the POSIX interface.
863    
864    10. Added --line-buffered to pcregrep.
865    
866    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
867        studied, and the match started with a letter with a code point greater than
868        127 whose first byte was different to the first byte of the other case of
869        the letter, the other case of this starting letter was not recognized
870        (#976).
871    
872    12. If a pattern that was studied started with a repeated Unicode property
873        test, for example, \p{Nd}+, there was the theoretical possibility of
874        setting up an incorrect bitmap of starting bytes, but fortunately it could
875        not have actually happened in practice until change 8 above was made (it
876        added property types that matched character-matching opcodes).
877    
878    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
879        possible starting bytes for non-anchored patterns.
880    
881    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
882        \R, and also a number of cases that involve Unicode properties, both
883        explicit and implicit when PCRE_UCP is set.
884    
885    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
886        input, it could crash or give wrong results if characters with values
887        greater than 0xc0 were present in the subject string. (Detail: it assumed
888        UTF-8 input when processing these items.)
889    
890    16. Added a lot of (int) casts to avoid compiler warnings in systems where
891        size_t is 64-bit (#991).
892    
893    17. Added a check for running out of memory when PCRE is compiled with
894        --disable-stack-for-recursion (#990).
895    
896    18. If the last data line in a file for pcretest does not have a newline on
897        the end, a newline was missing in the output.
898    
899    19. The default pcre_chartables.c file recognizes only ASCII characters (values
900        less than 128) in its various bitmaps. However, there is a facility for
901        generating tables according to the current locale when PCRE is compiled. It
902        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
903        characters, are recognized by isspace() and therefore were getting set in
904        these tables, and indeed these tables seem to approximate to ISO 8859. This
905        caused a problem in UTF-8 mode when pcre_study() was used to create a list
906        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
907        which of course cannot start UTF-8 characters. I have changed the code so
908        that only real ASCII characters (less than 128) and the correct starting
909        bytes for UTF-8 encodings are set for characters greater than 127 when in
910        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
911        altogether.)
912    
913    20. Added the /T option to pcretest so as to be able to run tests with non-
914        standard character tables, thus making it possible to include the tests
915        used for 19 above in the standard set of tests.
916    
917    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
918        reference to a subpattern the other side of a comment that contains an
919        opening parenthesis caused either an internal compiling error, or a
920        reference to the wrong subpattern.
921    
922    
923    Version 8.02 19-Mar-2010
924    ------------------------
925    
926    1.  The Unicode data tables have been updated to Unicode 5.2.0.
927    
928    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
929        configured.
930    
931    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
932        original author of that file, following a query about its status.
933    
934    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
935        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
936    
937    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
938        quantifier applied to a forward-referencing subroutine call, could compile
939        incorrect code or give the error "internal error: previously-checked
940        referenced subpattern not found".
941    
942    6.  Both MS Visual Studio and Symbian OS have problems with initializing
943        variables to point to external functions. For these systems, therefore,
944        pcre_malloc etc. are now initialized to local functions that call the
945        relevant global functions.
946    
947    7.  There were two entries missing in the vectors called coptable and poptable
948        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
949        I've fixed the data, and added a kludgy way of testing at compile time that
950        the lengths are correct (equal to the number of opcodes).
951    
952    8.  Following on from 7, I added a similar kludge to check the length of the
953        eint vector in pcreposix.c.
954    
955    9.  Error texts for pcre_compile() are held as one long string to avoid too
956        much relocation at load time. To find a text, the string is searched,
957        counting zeros. There was no check for running off the end of the string,
958        which could happen if a new error number was added without updating the
959        string.
960    
961    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
962    
963    11. \K was not working if it appeared in an atomic group or in a group that
964        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
965        \K is "not well defined" if used in an assertion. PCRE now accepts it if
966        the assertion is positive, but not if it is negative.
967    
968    12. Change 11 fortuitously reduced the size of the stack frame used in the
969        "match()" function of pcre_exec.c by one pointer. Forthcoming
970        implementation of support for (*MARK) will need an extra pointer on the
971        stack; I have reserved it now, so that the stack frame size does not
972        decrease.
973    
974    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
975        item in branch that calls a recursion is a subroutine call - as in the
976        second branch in the above example - was incorrectly given the compile-
977        time error "recursive call could loop indefinitely" because pcre_compile()
978        was not correctly checking the subroutine for matching a non-empty string.
979    
980    14. The checks for overrunning compiling workspace could trigger after an
981        overrun had occurred. This is a "should never occur" error, but it can be
982        triggered by pathological patterns such as hundreds of nested parentheses.
983        The checks now trigger 100 bytes before the end of the workspace.
984    
985    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
986    
987    
988    Version 8.01 19-Jan-2010
989    ------------------------
990    
991    1.  If a pattern contained a conditional subpattern with only one branch (in
992        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
993        computed the wrong minimum data length (which is of course zero for such
994        subpatterns). This could cause incorrect "no match" results.
995    
996    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
997        the pattern is reset in the first branch, pcre_compile() failed with
998        "internal error: code overflow at offset...". This happened only when
999        the reset was to the original external option setting. (An optimization
1000        abstracts leading options settings into an external setting, which was the
1001        cause of this.)
1002    
1003    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1004        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1005        assertion pattern did not match (meaning that the assertion was true), it
1006        was incorrectly treated as false if the SKIP had been reached during the
1007        matching. This also applied to assertions used as conditions.
1008    
1009    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1010        assertion subpattern, including such a pattern used as a condition,
1011        unpredictable results occurred, instead of the error return
1012        PCRE_ERROR_DFA_UITEM.
1013    
1014    5.  The C++ GlobalReplace function was not working like Perl for the special
1015        situation when an empty string is matched. It now does the fancy magic
1016        stuff that is necessary.
1017    
1018    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1019        removed. (These were left over from very, very early versions of PCRE.)
1020    
1021    7.  Some cosmetic changes to the code to make life easier when compiling it
1022        as part of something else:
1023    
1024        (a) Change DEBUG to PCRE_DEBUG.
1025    
1026        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1027            called "current" as "current_branch", to prevent a collision with the
1028            Linux macro when compiled as a kernel module.
1029    
1030        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1031            prevent a collision with the Linux macro when compiled as a kernel
1032            module.
1033    
1034    8.  In pcre_compile() there are some checks for integer overflows that used to
1035        cast potentially large values to (double). This has been changed to that
1036        when building, a check for int64_t is made, and if it is found, it is used
1037        instead, thus avoiding the use of floating point arithmetic. (There is no
1038        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1039        double.
1040    
1041    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1042        2005 (difference between two addresses compared to an unsigned value).
1043    
1044    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1045        custom one, because of the following reported problem in Windows:
1046    
1047          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1048              under Win32.
1049          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1050              therefore missing the function definition.
1051          - The compiler thus generates a "C" signature for the test function.
1052          - The linker fails to find the "C" function.
1053          - PCRE fails to configure if asked to do so against libbz2.
1054    
1055    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1056        messages were output:
1057    
1058          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1059          rerunning libtoolize, to keep the correct libtool macros in-tree.
1060          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1061    
1062        I have done both of these things.
1063    
1064    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1065        most of the time, it *can* run out if it is given a pattern that contains a
1066        runaway infinite recursion. I updated the discussion in the pcrestack man
1067        page.
1068    
1069    13. Now that we have gone to the x.xx style of version numbers, the minor
1070        version may start with zero. Using 08 or 09 is a bad idea because users
1071        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1072        interpreted as invalid octal numbers. I've updated the previous comment in
1073        configure.ac, and also added a check that gives an error if 08 or 09 are
1074        used.
1075    
1076    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1077        causing partial matching to fail when the end of the subject matched \W
1078        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1079    
1080    15. There were some discrepancies between the declarations in pcre_internal.h
1081        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1082        their definitions. The declarations used "const uschar *" and the
1083        definitions used USPTR. Even though USPTR is normally defined as "const
1084        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1085        reported that: "This difference in casting confuses some C++ compilers, for
1086        example, SunCC recognizes above declarations as different functions and
1087        generates broken code for hbpcre." I have changed the declarations to use
1088        USPTR.
1089    
1090    16. GNU libtool is named differently on some systems. The autogen.sh script now
1091        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1092        (FreeBSD).
1093    
1094    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1095        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1096        comment: "Figure out how to create a longlong from a string: strtoll and
1097        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1098        instance, but it only takes 2 args instead of 3!"
1099    
1100    18. A subtle bug concerned with back references has been fixed by a change of
1101        specification, with a corresponding code fix. A pattern such as
1102        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1103        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1104        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1105        same bug. Such groups have to be quantified to be useful, or contained
1106        inside another quantified group. (If there's no repetition, the reference
1107        can never match.) The problem arises because, having left the group and
1108        moved on to the rest of the pattern, a later failure that backtracks into
1109        the group uses the captured value from the final iteration of the group
1110        rather than the correct earlier one. I have fixed this in PCRE by forcing
1111        any group that contains a reference to itself to be an atomic group; that
1112        is, there cannot be any backtracking into it once it has completed. This is
1113        similar to recursive and subroutine calls.
1114    
1115    
1116    Version 8.00 19-Oct-09
1117    ----------------------
1118    
1119    1.  The table for translating pcre_compile() error codes into POSIX error codes
1120        was out-of-date, and there was no check on the pcre_compile() error code
1121        being within the table. This could lead to an OK return being given in
1122        error.
1123    
1124    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1125        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1126        in a Windows environment.
1127    
1128    3.  The pcregrep --count option prints the count for each file even when it is
1129        zero, as does GNU grep. However, pcregrep was also printing all files when
1130        --files-with-matches was added. Now, when both options are given, it prints
1131        counts only for those files that have at least one match. (GNU grep just
1132        prints the file name in this circumstance, but including the count seems
1133        more useful - otherwise, why use --count?) Also ensured that the
1134        combination -clh just lists non-zero counts, with no names.
1135    
1136    4.  The long form of the pcregrep -F option was incorrectly implemented as
1137        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1138        but it seems right to fix it, and I didn't think it was worth preserving
1139        the old behaviour.
1140    
1141    5.  The command line items --regex=pattern and --regexp=pattern were not
1142        recognized by pcregrep, which required --regex pattern or --regexp pattern
1143        (with a space rather than an '='). The man page documented the '=' forms,
1144        which are compatible with GNU grep; these now work.
1145    
1146    6.  No libpcreposix.pc file was created for pkg-config; there was just
1147        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1148    
1149    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1150        when UCP support is not needed, by modifying the Python script that
1151        generates it from Unicode data files. This should not matter if the module
1152        is correctly used as a library, but I received one complaint about 50K of
1153        unwanted data. My guess is that the person linked everything into his
1154        program rather than using a library. Anyway, it does no harm.
1155    
1156    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1157        was a minimum greater than 1 for a wide character in a possessive
1158        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1159        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1160        character. Chaos in the form of incorrect output or a compiling loop could
1161        result.
1162    
1163    9.  The restrictions on what a pattern can contain when partial matching is
1164        requested for pcre_exec() have been removed. All patterns can now be
1165        partially matched by this function. In addition, if there are at least two
1166        slots in the offset vector, the offset of the earliest inspected character
1167        for the match and the offset of the end of the subject are set in them when
1168        PCRE_ERROR_PARTIAL is returned.
1169    
1170    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1171        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1172        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1173        and may be more useful for multi-segment matching.
1174    
1175    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1176        used to be given if ever the end of the subject was reached; now it is
1177        given only if matching could not proceed because another character was
1178        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1179        string "Z", which now yields "no match" instead of "partial match". In the
1180        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1181        final character ended with (*FAIL).
1182    
1183    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1184        if the pattern had a "must contain" character that was already found in the
1185        earlier partial match, unless partial matching was again requested. For
1186        example, with the pattern /dog.(body)?/, the "must contain" character is
1187        "g". If the first part-match was for the string "dog", restarting with
1188        "sbody" failed. This bug has been fixed.
1189    
1190    13. The string returned by pcre_dfa_exec() after a partial match has been
1191        changed so that it starts at the first inspected character rather than the
1192        first character of the match. This makes a difference only if the pattern
1193        starts with a lookbehind assertion or \b or \B (\K is not supported by
1194        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1195        matching functions compatible, and I think it's the right thing to do.
1196    
1197    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1198        so that the demonstration program is easily available in environments where
1199        PCRE has not been installed from source.
1200    
1201    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1202        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1203        library.
1204    
1205    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1206        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1207        is not the first non-POSIX option to be added. Clearly some people find
1208        these options useful.
1209    
1210    17. If a caller to the POSIX matching function regexec() passes a non-zero
1211        value for nmatch with a NULL value for pmatch, the value of
1212        nmatch is forced to zero.
1213    
1214    18. RunGrepTest did not have a test for the availability of the -u option of
1215        the diff command, as RunTest does. It now checks in the same way as
1216        RunTest, and also checks for the -b option.
1217    
1218    19. If an odd number of negated classes containing just a single character
1219        interposed, within parentheses, between a forward reference to a named
1220        subpattern and the definition of the subpattern, compilation crashed with
1221        an internal error, complaining that it could not find the referenced
1222        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1223        [The bug was that it was starting one character too far in when skipping
1224        over the character class, thus treating the ] as data rather than
1225        terminating the class. This meant it could skip too much.]
1226    
1227    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1228        /g option in pcretest when the pattern contains \K, which makes it possible
1229        to have an empty string match not at the start, even when the pattern is
1230        anchored. Updated pcretest and pcredemo to use this option.
1231    
1232    21. If the maximum number of capturing subpatterns in a recursion was greater
1233        than the maximum at the outer level, the higher number was returned, but
1234        with unset values at the outer level. The correct (outer level) value is
1235        now given.
1236    
1237    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1238        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1239        make it do so. The string so far is captured, making this feature
1240        compatible with Perl.
1241    
1242    23. The tests have been re-organized, adding tests 11 and 12, to make it
1243        possible to check the Perl 5.10 features against Perl 5.10.
1244    
1245    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1246        pattern matches a fixed length string. PCRE did not allow this; now it
1247        does. Neither allows recursion.
1248    
1249    25. I finally figured out how to implement a request to provide the minimum
1250        length of subject string that was needed in order to match a given pattern.
1251        (It was back references and recursion that I had previously got hung up
1252        on.) This code has now been added to pcre_study(); it finds a lower bound
1253        to the length of subject needed. It is not necessarily the greatest lower
1254        bound, but using it to avoid searching strings that are too short does give
1255        some useful speed-ups. The value is available to calling programs via
1256        pcre_fullinfo().
1257    
1258    26. While implementing 25, I discovered to my embarrassment that pcretest had
1259        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1260        study optimizations had never been tested with that matching function.
1261        Oops. What is worse, even when it was passed study data, there was a bug in
1262        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1263        were also very few tests of studied patterns with pcre_dfa_exec().
1264    
1265    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1266        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1267        on the other side of the coin, they are no longer allowed to have different
1268        names, because these cannot be distinguished in PCRE, and this has caused
1269        confusion. (This is a difference from Perl.)
1270    
1271    28. When duplicate subpattern names are present (necessarily with different
1272        numbers, as required by 27 above), and a test is made by name in a
1273        conditional pattern, either for a subpattern having been matched, or for
1274        recursion in such a pattern, all the associated numbered subpatterns are
1275        tested, and the overall condition is true if the condition is true for any
1276        one of them. This is the way Perl works, and is also more like the way
1277        testing by number works.
1278    
1279    
1280    Version 7.9 11-Apr-09
1281    ---------------------
1282    
1283    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1284        (pcretest), all targets were linked against these libraries. This included
1285        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1286        libraries. This caused unwanted dependencies to be created. This problem
1287        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1288        pcretest is linked with readline.
1289    
1290    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1291        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1292        moved outside it again, because FALSE and TRUE are already defined in AIX,
1293        but BOOL is not.
1294    
1295    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1296        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1297    
1298    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1299        hyphens) following file names and line numbers when outputting matching
1300        lines. This is not true; no spaces are inserted. I have also clarified the
1301        wording for the --colour (or --color) option.
1302    
1303    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1304        was not coloured; this is different to GNU grep, so I have changed it to be
1305        the same.
1306    
1307    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1308        each matching line was coloured. Now it goes on to look for further matches
1309        of any of the test patterns, which is the same behaviour as GNU grep.
1310    
1311    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1312        doesn't make sense to accept an empty string match in pcregrep, so I have
1313        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1314        seems to be how GNU grep behaves.
1315    
1316    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1317        start or after a newline", because the conditional assertion was not being
1318        correctly handled. The rule now is that both the assertion and what follows
1319        in the first alternative must satisfy the test.
1320    
1321    9.  If auto-callout was enabled in a pattern with a conditional group whose
1322        condition was an assertion, PCRE could crash during matching, both with
1323        pcre_exec() and pcre_dfa_exec().
1324    
1325    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1326        used for matching.
1327    
1328    11. Unicode property support in character classes was not working for
1329        characters (bytes) greater than 127 when not in UTF-8 mode.
1330    
1331    12. Added the -M command line option to pcretest.
1332    
1333    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1334    
1335    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1336    
1337    16. Added comments and documentation about mis-use of no_arg in the C++
1338        wrapper.
1339    
1340    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1341        from Martin Jerabek that uses macro names for all relevant character and
1342        string constants.
1343    
1344    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1345        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1346        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1347        these, but not everybody uses configure.
1348    
1349    19. A conditional group that had only one branch was not being correctly
1350        recognized as an item that could match an empty string. This meant that an
1351        enclosing group might also not be so recognized, causing infinite looping
1352        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1353        with the subject "ab", where knowledge that the repeated group can match
1354        nothing is needed in order to break the loop.
1355    
1356    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1357        exec(), but without supplying a callout function, matching went wrong.
1358    
1359    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1360        leak if the size of the offset vector was greater than 30. When the vector
1361        is smaller, the saved offsets during recursion go onto a local stack
1362        vector, but for larger vectors malloc() is used. It was failing to free
1363        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1364        error, in fact).
1365    
1366    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1367        heapframe that is used only when UTF-8 support is enabled. This caused no
1368        problem, but was untidy.
1369    
1370    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1371        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1372        included within another project.
1373    
1374    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1375        slightly modified by me:
1376    
1377          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1378              not building pcregrep.
1379    
1380          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1381              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1382    
1383    25. Forward references, both numeric and by name, in patterns that made use of
1384        duplicate group numbers, could behave incorrectly or give incorrect errors,
1385        because when scanning forward to find the reference group, PCRE was not
1386        taking into account the duplicate group numbers. A pattern such as
1387        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1388    
1389    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1390        the feature of a custom pointer more persuasive (as requested by a user).
1391    
1392    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1393        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1394        pre-defined.
1395    
1396    28. Added support for (*UTF8) at the start of a pattern.
1397    
1398    29. Arrange for flags added by the "release type" setting in CMake to be shown
1399        in the configuration summary.
1400    
1401    
1402    Version 7.8 05-Sep-08
1403    ---------------------
1404    
1405    1.  Replaced UCP searching code with optimized version as implemented for Ad
1406        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1407        stage table and inline lookup instead of a function, giving speed ups of 2
1408        to 5 times on some simple patterns that I tested. Permission was given to
1409        distribute the MultiStage2.py script that generates the tables (it's not in
1410        the tarball, but is in the Subversion repository).
1411    
1412    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1413        scripts.
1414    
1415    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1416        a group with a zero qualifier. The result of the study could be incorrect,
1417        or the function might crash, depending on the pattern.
1418    
1419    4.  Caseless matching was not working for non-ASCII characters in back
1420        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1421        It now works when Unicode Property Support is available.
1422    
1423    5.  In pcretest, an escape such as \x{de} in the data was always generating
1424        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1425        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1426        truncation.
1427    
1428    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1429    
1430    7.  Added two (int) casts to pcregrep when printing the difference of two
1431        pointers, in case they are 64-bit values.
1432    
1433    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1434        test 2 if it fails.
1435    
1436    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1437        and a #define of that name to empty if it is not externally set. This is to
1438        allow users of MSVC to set it if necessary.
1439    
1440    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1441        the convenience functions in the pcre_get.c source file.
1442    
1443    11. An option change at the start of a pattern that had top-level alternatives
1444        could cause overwriting and/or a crash. This command provoked a crash in
1445        some environments:
1446    
1447          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1448    
1449        This potential security problem was recorded as CVE-2008-2371.
1450    
1451    12. For a pattern where the match had to start at the beginning or immediately
1452        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1453        pcre_dfa_exec() could read past the end of the passed subject if there was
1454        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1455        pcretest so that it places the subject at the end of its malloc-ed buffer.
1456    
1457    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1458        exec() might read past the end of the data buffer in UTF-8 mode.
1459    
1460    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1461        the data contained the byte 0x85 as part of a UTF-8 character within its
1462        first line. This applied both to normal and DFA matching.
1463    
1464    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1465        /^[^d]*?$/8 failed to match "abc".
1466    
1467    16. Added a missing copyright notice to pcrecpp_internal.h.
1468    
1469    17. Make it more clear in the documentation that values returned from
1470        pcre_exec() in ovector are byte offsets, not character counts.
1471    
1472    18. Tidied a few places to stop certain compilers from issuing warnings.
1473    
1474    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1475        supplied by Stefan Weber. I made a further small update for 7.8 because
1476        there is a change of source arrangements: the pcre_searchfuncs.c module is
1477        replaced by pcre_ucd.c.
1478    
1479    
1480    Version 7.7 07-May-08
1481    ---------------------
1482    
1483    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1484        a string to a long long, pretend we don't even have a long long." This is
1485        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1486    
1487    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1488        pre-7.6 versions, which defined a global no_arg variable instead of putting
1489        it in the RE class. (See also #8 below.)
1490    
1491    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1492        Lopes.
1493    
1494    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1495    
1496        (1) The include/exclude patterns were being applied to the whole pathnames
1497            of files, instead of just to the final components.
1498    
1499        (2) If there was more than one level of directory, the subdirectories were
1500            skipped unless they satisfied the include/exclude conditions. This is
1501            inconsistent with GNU grep (and could even be seen as contrary to the
1502            pcregrep specification - which I improved to make it absolutely clear).
1503            The action now is always to scan all levels of directory, and just
1504            apply the include/exclude patterns to regular files.
1505    
1506    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1507        --exclude_dir in the tests to avoid scanning .svn directories.
1508    
1509    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1510        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1511        doesn't support NULs in patterns.
1512    
1513    7.  Added some missing "const"s to declarations of static tables in
1514        pcre_compile.c and pcre_dfa_exec.c.
1515    
1516    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1517        caused by fix #2  above. (Subsequently also a second patch to fix the
1518        first patch. And a third patch - this was a messy problem.)
1519    
1520    9.  Applied Craig's patch to remove the use of push_back().
1521    
1522    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1523        matching function regexec().
1524    
1525    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1526        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1527        references. PCRE supports relative numbers with this syntax (I don't think
1528        Oniguruma does).
1529    
1530    12. Previously, a group with a zero repeat such as (...){0} was completely
1531        omitted from the compiled regex. However, this means that if the group
1532        was called as a subroutine from elsewhere in the pattern, things went wrong
1533        (an internal error was given). Such groups are now left in the compiled
1534        pattern, with a new opcode that causes them to be skipped at execution
1535        time.
1536    
1537    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1538        to the way PCRE behaves:
1539    
1540        (a) A lone ] character is dis-allowed (Perl treats it as data).
1541    
1542        (b) A back reference to an unmatched subpattern matches an empty string
1543            (Perl fails the current match path).
1544    
1545        (c) A data ] in a character class must be notated as \] because if the
1546            first data character in a class is ], it defines an empty class. (In
1547            Perl it is not possible to have an empty class.) The empty class []
1548            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1549            The negative empty class [^] matches any one character, independently
1550            of the DOTALL setting.
1551    
1552    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1553        non-existent subpattern following a character class starting with ']' and
1554        containing () gave an internal compiling error instead of "reference to
1555        non-existent subpattern". Fortunately, when the pattern did exist, the
1556        compiled code was correct. (When scanning forwards to check for the
1557        existencd of the subpattern, it was treating the data ']' as terminating
1558        the class, so got the count wrong. When actually compiling, the reference
1559        was subsequently set up correctly.)
1560    
1561    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1562        it was being rejected as not supported by pcre_dfa_exec(), even though
1563        other assertions are supported. I have made pcre_dfa_exec() support
1564        (*FAIL).
1565    
1566    16. The implementation of 13c above involved the invention of a new opcode,
1567        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1568        cannot be changed at match time, I realized I could make a small
1569        improvement to matching performance by compiling OP_ALLANY instead of
1570        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1571        on the OP_ANY path.
1572    
1573    17. Compiling pcretest on Windows with readline support failed without the
1574        following two fixes: (1) Make the unistd.h include conditional on
1575        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1576    
1577    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1578        ncurses library to be included for pcretest when ReadLine support is
1579        requested, but also to allow for it to be overridden. This patch came from
1580        Daniel Bergström.
1581    
1582    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1583        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1584        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1585        spotting this.
1586    
1587    
1588    Version 7.6 28-Jan-08
1589    ---------------------
1590    
1591    1.  A character class containing a very large number of characters with
1592        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1593        overflow.
1594    
1595    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1596        HAVE_LONG_LONG is not defined.
1597    
1598    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1599        bring them up to date and include new features. This patch includes:
1600    
1601        - Fixed PH's badly added libz and libbz2 support.
1602        - Fixed a problem with static linking.
1603        - Added pcredemo. [But later removed - see 7 below.]
1604        - Fixed dftables problem and added an option.
1605        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1606            HAVE_LONG_LONG.
1607        - Added readline support for pcretest.
1608        - Added an listing of the option settings after cmake has run.
1609    
1610    4.  A user submitted a patch to Makefile that makes it easy to create
1611        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1612        Makefile.am that cause it to include this special target, without
1613        affecting anything else. Note that the same mingw target plus all
1614        the other distribution libraries and programs are now supported
1615        when configuring with CMake (see 6 below) instead of with
1616        Configure/Make.
1617    
1618    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1619        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1620        exported in the Windows port". It has not yet been confirmed that the patch
1621        solves the problem, but it does no harm.
1622    
1623    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1624        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1625        with CMake, and also correct the comment about stack recursion.
1626    
1627    7.  Remove the automatic building of pcredemo from the ./configure system and
1628        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1629        of a program that users should build themselves after PCRE is installed, so
1630        building it automatically is not really right. What is more, it gave
1631        trouble in some build environments.
1632    
1633    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1634    
1635    
1636    Version 7.5 10-Jan-08
1637    ---------------------
1638    
1639    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1640        values in parens when parsing an RE using the C++ wrapper."
1641    
1642    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1643        Characters greater than 255 were excluded from the class instead of being
1644        included.
1645    
1646    3.  The same bug as (2) above applied to negated POSIX classes such as
1647        [:^space:].
1648    
1649    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1650        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1651        I have changed it.
1652    
1653    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1654        first named subpattern) and a construct such as (?&a) would reference the
1655        first named subpattern whose name started with "a" (in other words, the
1656        length check was missing). Both these problems are fixed. "Subpattern name
1657        expected" is now given for (?&) (a zero-length name), and this patch also
1658        makes it give the same error for \k'' (previously it complained that that
1659        was a reference to a non-existent subpattern).
1660    
1661    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1662        this is right because (?- can be followed by option settings as well as by
1663        digits. I have, however, made the messages clearer.
1664    
1665    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1666        than the number used in the conditional) now cause a compile-time error.
1667        This is actually not compatible with Perl, which accepts such patterns, but
1668        treats the conditional as always being FALSE (as PCRE used to), but it
1669        seems to me that giving a diagnostic is better.
1670    
1671    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1672        and messages.
1673    
1674    9.  Fix two occurrences of "backslash" in comments that should have been
1675        "backspace".
1676    
1677    10. Remove two redundant lines of code that can never be obeyed (their function
1678        was moved elsewhere).
1679    
1680    11. The program that makes PCRE's Unicode character property table had a bug
1681        which caused it to generate incorrect table entries for sequences of
1682        characters that have the same character type, but are in different scripts.
1683        It amalgamated them into a single range, with the script of the first of
1684        them. In other words, some characters were in the wrong script. There were
1685        thirteen such cases, affecting characters in the following ranges:
1686    
1687          U+002b0 - U+002c1
1688          U+0060c - U+0060d
1689          U+0061e - U+00612
1690          U+0064b - U+0065e
1691          U+0074d - U+0076d
1692          U+01800 - U+01805
1693          U+01d00 - U+01d77
1694          U+01d9b - U+01dbf
1695          U+0200b - U+0200f
1696          U+030fc - U+030fe
1697          U+03260 - U+0327f
1698          U+0fb46 - U+0fbb1
1699          U+10450 - U+1049d
1700    
1701    12. The -o option (show only the matching part of a line) for pcregrep was not
1702        compatible with GNU grep in that, if there was more than one match in a
1703        line, it showed only the first of them. It now behaves in the same way as
1704        GNU grep.
1705    
1706    13. If the -o and -v options were combined for pcregrep, it printed a blank
1707        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1708        does the same. The return code can be used to tell if there were any
1709        non-matching lines.
1710    
1711    14. Added --file-offsets and --line-offsets to pcregrep.
1712    
1713    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1714        infinitely looping recursion. The bug was that positive lookaheads were not
1715        being skipped when checking for a possible empty match (negative lookaheads
1716        and both kinds of lookbehind were skipped).
1717    
1718    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1719        inclusion of <windows.h> to before rather than after the definition of
1720        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1721    
1722    17. Specifying a possessive quantifier with a specific limit for a Unicode
1723        character property caused pcre_compile() to compile bad code, which led at
1724        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1725        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1726        caused the error; without that there was no problem.
1727    
1728    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1729    
1730    19. Added --enable-pcretest-libreadline.
1731    
1732    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1733        RE::GlobalReplace(). As a result, the number of replacements returned was
1734        double what it should be. I removed one of the increments, but Craig sent a
1735        later patch that removed the other one (the right fix) and added unit tests
1736        that check the return values (which was not done before).
1737    
1738    21. Several CMake things:
1739    
1740        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1741            the names libpcre and libpcreposix, not just pcre and pcreposix.
1742    
1743        (2) The above change means that pcretest and pcregrep are now correctly
1744            linked with the newly-built libraries, not previously installed ones.
1745    
1746        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1747    
1748    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1749        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1750        UTF-8 newline character). The key issue is that the pattern starts .*;
1751        this means that the match must be either at the beginning, or after a
1752        newline. The bug was in the code for advancing after a failed match and
1753        checking that the new position followed a newline. It was not taking
1754        account of UTF-8 characters correctly.
1755    
1756    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1757        character classes. PCRE was not treating the sequence [:...:] as a
1758        character class unless the ... were all letters. Perl, however, seems to
1759        allow any characters between [: and :], though of course it rejects as
1760        unknown any "names" that contain non-letters, because all the known class
1761        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1762        for example, whereas PCRE did not - it did not recognize a POSIX character
1763        class. This seemed a bit dangerous, so the code has been changed to be
1764        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1765        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1766        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1767        Perl does, and where it didn't before.
1768    
1769    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1770        Windows environments %n is disabled by default.
1771    
1772    
1773    Version 7.4 21-Sep-07
1774    ---------------------
1775    
1776    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1777        means that a class such as [\s] counted as "explicit reference to CR or
1778        LF". That isn't really right - the whole point of the change was to try to
1779        help when there was an actual mention of one of the two characters. So now
1780        the change happens only if \r or \n (or a literal CR or LF) character is
1781        encountered.
1782    
1783    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1784        of both had grown to the point where there were only 3 bits left.
1785        Fortunately, there was spare space in the data structure, and so I have
1786        moved the internal flags into a new 16-bit field to free up more option
1787        bits.
1788    
1789    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1790        but did not set the internal JCHANGED flag - either of these is enough to
1791        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1792        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1793        start sets both bits.
1794    
1795    4.  Added options (at build time, compile time, exec time) to change \R from
1796        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1797    
1798    5.  doc/pcresyntax.html was missing from the distribution.
1799    
1800    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1801        compatibility, even though it is no longer used.
1802    
1803    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1804        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1805        windows.h file is present (where different names are used). [This was
1806        reversed later after testing - see 16 below.]
1807    
1808    8.  Changed all #include <config.h> to #include "config.h". There were also
1809        some further <pcre.h> cases that I changed to "pcre.h".
1810    
1811    9.  When pcregrep was used with the --colour option, it missed the line ending
1812        sequence off the lines that it output.
1813    
1814    10. It was pointed out to me that arrays of string pointers cause lots of
1815        relocations when a shared library is dynamically loaded. A technique of
1816        using a single long string with a table of offsets can drastically reduce
1817        these. I have refactored PCRE in four places to do this. The result is
1818        dramatic:
1819    
1820          Originally:                          290
1821          After changing UCP table:            187
1822          After changing error message table:   43
1823          After changing table of "verbs"       36
1824          After changing table of Posix names   22
1825    
1826        Thanks to the folks working on Gregex for glib for this insight.
1827    
1828    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1829        unicode-properties was also set.
1830    
1831    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1832    
1833    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1834        checked only for CRLF.
1835    
1836    14. Added casts to pcretest.c to avoid compiler warnings.
1837    
1838    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1839    
1840    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1841        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1842        entirely. This removes changes made in 7 above.
1843    
1844    17. The CMake files have been updated, and there is now more information about
1845        building with CMake in the NON-UNIX-USE document.
1846    
1847    
1848    Version 7.3 28-Aug-07
1849    ---------------------
1850    
1851     1. In the rejigging of the build system that eventually resulted in 7.1, the
1852        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1853        brackets there is not right, since it causes compilers to look for an
1854        installed pcre.h, not the version that is in the source that is being
1855        compiled (which of course may be different). I have changed it back to:
1856    
1857          #include "pcre.h"
1858    
1859        I have a vague recollection that the change was concerned with compiling in
1860        different directories, but in the new build system, that is taken care of
1861        by the VPATH setting the Makefile.
1862    
1863     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1864        when the subject happened to end in the byte 0x85 (e.g. if the last
1865        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1866        characters but of course it shouldn't be taken as a newline when it is part
1867        of another character. The bug was that, for an unlimited repeat of . in
1868        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1869        characters when looking for a newline.
1870    
1871     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1872    
1873     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1874        in debug output.
1875    
1876     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1877        long printing in the pcrecpp unittest when running under MinGW.
1878    
1879     6. ESC_K was left out of the EBCDIC table.
1880    
1881     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1882        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1883        limit also applies to "virtual nesting" when a pattern is recursive, and in
1884        this case 1000 isn't so big. I have been able to remove this limit at the
1885        expense of backing off one optimization in certain circumstances. Normally,
1886        when pcre_exec() would call its internal match() function recursively and
1887        immediately return the result unconditionally, it uses a "tail recursion"
1888        feature to save stack. However, when a subpattern that can match an empty
1889        string has an unlimited repetition quantifier, it no longer makes this
1890        optimization. That gives it a stack frame in which to save the data for
1891        checking that an empty string has been matched. Previously this was taken
1892        from the 1000-entry workspace that had been reserved. So now there is no
1893        explicit limit, but more stack is used.
1894    
1895     8. Applied Daniel's patches to solve problems with the import/export magic
1896        syntax that is required for Windows, and which was going wrong for the
1897        pcreposix and pcrecpp parts of the library. These were overlooked when this
1898        problem was solved for the main library.
1899    
1900     9. There were some crude static tests to avoid integer overflow when computing
1901        the size of patterns that contain repeated groups with explicit upper
1902        limits. As the maximum quantifier is 65535, the maximum group length was
1903        set at 30,000 so that the product of these two numbers did not overflow a
1904        32-bit integer. However, it turns out that people want to use groups that
1905        are longer than 30,000 bytes (though not repeat them that many times).
1906        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1907        made it possible to implement the integer overflow checks in a much more
1908        dynamic way, which I have now done. The artificial limitation on group
1909        length has been removed - we now have only the limit on the total length of
1910        the compiled pattern, which depends on the LINK_SIZE setting.
1911    
1912    10. Fixed a bug in the documentation for get/copy named substring when
1913        duplicate names are permitted. If none of the named substrings are set, the
1914        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1915        empty string.
1916    
1917    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1918        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1919        because the ] is interpreted as the first data character and the
1920        terminating ] is not found. PCRE has been made compatible with Perl in this
1921        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1922        cause memory overwriting.
1923    
1924    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1925        string has been matched (to stop an infinite loop). It was not recognizing
1926        a conditional subpattern that could match an empty string if that
1927        subpattern was within another subpattern. For example, it looped when
1928        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1929        condition was not nested. This bug has been fixed.
1930    
1931    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1932        past the start of the subject in the presence of bytes with the top bit
1933        set, for example "\x8aBCD".
1934    
1935    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1936        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1937    
1938    14. Optimized (?!) to (*FAIL).
1939    
1940    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1941        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1942        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1943        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1944        does: it's just the validity check that is more restrictive.
1945    
1946    16. Inserted checks for integer overflows during escape sequence (backslash)
1947        processing, and also fixed erroneous offset values for syntax errors during
1948        backslash processing.
1949    
1950    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1951        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1952    
1953    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1954        caused an overrun.
1955    
1956    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1957        something other than just ASCII characters) inside a group that had an
1958        unlimited repeat caused a loop at compile time (while checking to see
1959        whether the group could match an empty string).
1960    
1961    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1962        [\P{Any}] did so. (Error in the code for printing property names.)
1963    
1964    21. An orphan \E inside a character class could cause a crash.
1965    
1966    22. A repeated capturing bracket such as (A)? could cause a wild memory
1967        reference during compilation.
1968    
1969    23. There are several functions in pcre_compile() that scan along a compiled
1970        expression for various reasons (e.g. to see if it's fixed length for look
1971        behind). There were bugs in these functions when a repeated \p or \P was
1972        present in the pattern. These operators have additional parameters compared
1973        with \d, etc, and these were not being taken into account when moving along
1974        the compiled data. Specifically:
1975    
1976        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1977            length.
1978    
1979        (b) An item such as \pL+ within a repeated group could cause crashes or
1980            loops.
1981    
1982        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1983            "reference to non-existent subpattern" error.
1984    
1985        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1986    
1987    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1988        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1989    
1990    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1991    
1992    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1993        character were causing crashes (broken optimization).
1994    
1995    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1996        \p or \P) caused a compile-time loop.
1997    
1998    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1999        break. For example, the unstudied pattern [\r\n]A does not match the string
2000        "\r\nA" because change 7.0/46 below moves the current point on by two
2001        characters after failing to match at the start. However, the pattern \nA
2002        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2003        the same is true. There doesn't seem any very clean way out of this, but
2004        what I have chosen to do makes the common cases work: PCRE now takes note
2005        of whether there can be an explicit match for \r or \n anywhere in the
2006        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2007        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2008        pattern has explicit CR or LF references.
2009    
2010    29. Added (*CR) etc for changing newline setting at start of pattern.
2011    
2012    
2013    Version 7.2 19-Jun-07
2014    ---------------------
2015    
2016     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
2017        which is apparently normally available under Windows.
2018    
2019     2. Re-jig the pcregrep tests with different newline settings in an attempt
2020        to make them independent of the local environment's newline setting.
2021    
2022     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
2023    
2024     4. Some of the "internals" tests were previously cut out when the link size
2025        was not 2, because the output contained actual offsets. The recent new
2026        "Z" feature of pcretest means that these can be cut out, making the tests
2027        usable with all link sizes.
2028    
2029     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
2030        stack recursion. This gives a massive performance boost under BSD, but just
2031        a small improvement under Linux. However, it saves one field in the frame
2032        in all cases.
2033    
2034     6. Added more features from the forthcoming Perl 5.10:
2035    
2036        (a) (?-n) (where n is a string of digits) is a relative subroutine or
2037            recursion call. It refers to the nth most recently opened parentheses.
2038    
2039        (b) (?+n) is also a relative subroutine call; it refers to the nth next
2040            to be opened parentheses.
2041    
2042        (c) Conditions that refer to capturing parentheses can be specified
2043            relatively, for example, (?(-2)... or (?(+3)...
2044    
2045        (d) \K resets the start of the current match so that everything before
2046            is not part of it.
2047    
2048        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2049    
2050        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2051            reference syntax.
2052    
2053        (g) (?| introduces a group in which the numbering of parentheses in each
2054            alternative starts with the same number.
2055    
2056        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2057    
2058     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2059        PCRE_INFO_JCHANGED.
2060    
2061     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2062        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2063        for detecting groups that can match an empty string.
2064    
2065     9. A pattern with a very large number of alternatives (more than several
2066        hundred) was running out of internal workspace during the pre-compile
2067        phase, where pcre_compile() figures out how much memory will be needed. A
2068        bit of new cunning has reduced the workspace needed for groups with
2069        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2070        workspace instead of running out of the 4096 that are available.
2071    
2072    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2073    
2074    11. Applied patch from Google to remove an optimization that didn't quite work.
2075        The report of the bug said:
2076    
2077          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2078          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2079          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2080    
2081    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2082        it matched the wrong number of bytes.
2083    
2084    
2085    Version 7.1 24-Apr-07
2086  ---------------------  ---------------------
2087    
2088   1. Applied Bob Rossi and Daniel G's patches to convert the build system to one   1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
2089      that is more "standard", making use of automake and other autotools. There      that is more "standard", making use of automake and other Autotools. There
2090      is some re-arrangement of the files and adjustment of comments consequent      is some re-arrangement of the files and adjustment of comments consequent
2091      on this.      on this.
2092    
# Line 26  Version 7.1 05-Mar-07 Line 2107  Version 7.1 05-Mar-07
2107      .br or .in.      .br or .in.
2108    
2109   4. Updated comments in configure.ac that get placed in config.h.in and also   4. Updated comments in configure.ac that get placed in config.h.in and also
2110      arranged for config.h to be included in the distribution, with the name      arranged for config.h to be included in the distribution, with the name
2111      config.h.generic, for the benefit of those who have to compile without      config.h.generic, for the benefit of those who have to compile without
2112      Autotools (compare pcre.h, which is now distributed as pcre.h.generic).      Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
2113    
2114   5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan   5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
2115      Weber: (1) pcre_internal.h was missing some function renames; (2) updated      Weber: (1) pcre_internal.h was missing some function renames; (2) updated
2116      makevp.bat for the current PCRE, using the additional files !compile.txt,      makevp.bat for the current PCRE, using the additional files
2117      !linklib.txt, and pcregexp.pas.      makevp_c.txt, makevp_l.txt, and pcregexp.pas.
2118    
2119   6. A Windows user reported a minor discrepancy with test 2, which turned out   6. A Windows user reported a minor discrepancy with test 2, which turned out
2120      to be caused by a trailing space on an input line that had got lost in his      to be caused by a trailing space on an input line that had got lost in his
# Line 60  Version 7.1 05-Mar-07 Line 2141  Version 7.1 05-Mar-07
2141      called when UTF-8 support is disabled. Otherwise there are problems with a      called when UTF-8 support is disabled. Otherwise there are problems with a
2142      shared library.      shared library.
2143    
2144  10. The emulated memmove() function in pcre_internal.h (provided for those  10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
2145      environments that have neither memmove() nor bcopy()) was defining its  
2146      arguments as char * instead of void *.      (a) It was defining its arguments as char * instead of void *.
2147    
2148        (b) It was assuming that all moves were upwards in memory; this was true
2149            a long time ago when I wrote it, but is no longer the case.
2150    
2151        The emulated memove() is provided for those environments that have neither
2152        memmove() nor bcopy(). I didn't think anyone used it these days, but that
2153        is clearly not the case, as these two bugs were recently reported.
2154    
2155  11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,  11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
2156      and Detrail to create the HTML documentation, the .txt form of the man      and Detrail to create the HTML documentation, the .txt form of the man
2157      pages, and it removes trailing spaces from listed files. It also creates      pages, and it removes trailing spaces from listed files. It also creates
2158      pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter      pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
2159      case, it wraps all the #defines with #ifndefs. This script should be run      case, it wraps all the #defines with #ifndefs. This script should be run
2160      before "make dist".      before "make dist".
2161    
2162  12. Fixed two fairly obscure bugs concerned with quantified caseless matching  12. Fixed two fairly obscure bugs concerned with quantified caseless matching
2163      with Unicode property support.      with Unicode property support.
2164    
2165      (a) For a maximizing quantifier, if the two different cases of the      (a) For a maximizing quantifier, if the two different cases of the
2166          character were of different lengths in their UTF-8 codings (there are          character were of different lengths in their UTF-8 codings (there are
2167          some cases like this - I found 11), and the matching function had to          some cases like this - I found 11), and the matching function had to
2168          back up over a mixture of the two cases, it incorrectly assumed they          back up over a mixture of the two cases, it incorrectly assumed they
2169          were both the same length.          were both the same length.
2170    
2171      (b) When PCRE was configured to use the heap rather than the stack for      (b) When PCRE was configured to use the heap rather than the stack for
2172          recursion during matching, it was not correctly preserving the data for          recursion during matching, it was not correctly preserving the data for
2173          the other case of a UTF-8 character when checking ahead for a match          the other case of a UTF-8 character when checking ahead for a match
2174          while processing a minimizing repeat. If the check also involved          while processing a minimizing repeat. If the check also involved
2175          matching a wide character, but failed, corruption could cause an          matching a wide character, but failed, corruption could cause an
2176          erroneous result when trying to check for a repeat of the original          erroneous result when trying to check for a repeat of the original
2177          character.          character.
2178    
2179  13. Some tidying changes to the testing mechanism:  13. Some tidying changes to the testing mechanism:
2180    
2181      (a) The RunTest script now detects the internal link size and whether there      (a) The RunTest script now detects the internal link size and whether there
2182          is UTF-8 and UCP support by running ./pcretest -C instead of relying on          is UTF-8 and UCP support by running ./pcretest -C instead of relying on
2183          values substituted by "configure". (The RunGrepTest script already did          values substituted by "configure". (The RunGrepTest script already did
2184          this for UTF-8.) The configure.ac script no longer substitutes the          this for UTF-8.) The configure.ac script no longer substitutes the
2185          relevant variables.          relevant variables.
2186    
2187      (b) The debugging options /B and /D in pcretest show the compiled bytecode      (b) The debugging options /B and /D in pcretest show the compiled bytecode
2188          with length and offset values. This means that the output is different          with length and offset values. This means that the output is different
2189          for different internal link sizes. Test 2 is skipped for link sizes          for different internal link sizes. Test 2 is skipped for link sizes
2190          other than 2 because of this, bypassing the problem. Unfortunately,          other than 2 because of this, bypassing the problem. Unfortunately,
2191          there was also a test in test 3 (the locale tests) that used /B and          there was also a test in test 3 (the locale tests) that used /B and
2192          failed for link sizes other than 2. Rather than cut the whole test out,          failed for link sizes other than 2. Rather than cut the whole test out,
2193          I have added a new /Z option to pcretest that replaces the length and          I have added a new /Z option to pcretest that replaces the length and
2194          offset values with spaces. This is now used to make test 3 independent          offset values with spaces. This is now used to make test 3 independent
2195          of link size.          of link size. (Test 2 will be tidied up later.)
2196    
2197    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
2198        segmentation fault instead of returning the appropriate error message.
2199    
2200    15. In multiline mode when the newline sequence was set to "any", the pattern
2201        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
2202        This doesn't seem right; it now treats the CRLF combination as the line
2203        ending, and so does not match in that case. It's only a pattern such as ^$
2204        that would hit this one: something like ^ABC$ would have failed after \r
2205        and then tried again after \r\n.
2206    
2207    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
2208        in an attempt to make files that differ only in their line terminators
2209        compare equal. This works on Linux.
2210    
2211    17. Under certain error circumstances pcregrep might try to free random memory
2212        as it exited. This is now fixed, thanks to valgrind.
2213    
2214    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
2215        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
2216        was because its rules for how to advance for /g after matching an empty
2217        string at the end of a line did not allow for this case. They now check for
2218        it specially.
2219    
2220    20. pcretest is supposed to handle patterns and data of any length, by
2221        extending its buffers when necessary. It was getting this wrong when the
2222        buffer for a data line had to be extended.
2223    
2224    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
2225        CRLF as a newline sequence.
2226    
2227    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
2228        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
2229        I have nevertheless tidied it up.
2230    
2231    23. Added some casts to kill warnings from HP-UX ia64 compiler.
2232    
2233    24. Added a man page for pcre-config.
2234    
2235    
2236  Version 7.0 19-Dec-06  Version 7.0 19-Dec-06

Legend:
Removed from v.116  
changed lines
  Added in v.970

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12