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

Diff of /code/trunk/ChangeLog

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

revision 447 by ph10, Tue Sep 15 18:17:54 2009 UTC revision 1025 by ph10, Wed Aug 29 11:24:23 2012 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 8.00 ??-???-??  Version 8.32
5    ------------
6    
7    1.  Improved JIT compiler optimizations for first character search and single
8        character iterators.
9    
10    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
11        Patch by Daniel Richard G.
12    
13    3.  Single character iterator optimizations in the JIT compiler.
14    
15    4.  Improved JIT compiler optimizations for character ranges.
16    
17    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
18        Reported by Giuseppe D'Angelo.
19    
20    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
21        of a line, was being set incorrectly in cases where .* appeared inside
22        atomic brackets at the start of a pattern, or where there was a subsequent
23        *PRUNE or *SKIP.
24    
25    7.  Improved instruction cache flush for POWER/PowerPC.
26        Patch by Daniel Richard G.
27    
28    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
29        grep:
30    
31        (a) There is now no limit to the number of patterns to be matched.
32    
33        (b) An error is given if a pattern is too long.
34    
35        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
36            are now supported.
37    
38        (d) --exclude-from and --include-from (multiple use) have been added.
39    
40        (e) Exclusions and inclusions now apply to all files and directories, not
41            just to those obtained from scanning a directory recursively.
42    
43        (f) Multiple uses of -f and --file-list are now supported.
44    
45        (g) In a Windows environment, the default for -d has been changed from
46            "read" (the GNU grep default) to "skip", because otherwise the presence
47            of a directory in the file list provokes an error.
48    
49        (h) The documentation has been revised and clarified in places.
50    
51    9.  Improve the matching speed of capturing brackets.
52    
53    10. Changed the meaning of \X so that it now matches a Unicode extended
54        grapheme cluster.
55    
56    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
57        out POSIX threads when JIT support is configured.
58    
59    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
60    
61    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
62        structure could go wrong in environments where size_t is not the same size
63        as int.
64    
65    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
66        set.
67    
68    
69    
70    Version 8.31 06-July-2012
71    -------------------------
72    
73    1.  Fixing a wrong JIT test case and some compiler warnings.
74    
75    2.  Removed a bashism from the RunTest script.
76    
77    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
78        to unsigned type, result still unsigned" that was given by an MS compiler
79        on encountering the code "-sizeof(xxx)".
80    
81    4.  Partial matching support is added to the JIT compiler.
82    
83    5.  Fixed several bugs concerned with partial matching of items that consist
84        of more than one character:
85    
86        (a) /^(..)\1/ did not partially match "aba" because checking references was
87            done on an "all or nothing" basis. This also applied to repeated
88            references.
89    
90        (b) \R did not give a hard partial match if \r was found at the end of the
91            subject.
92    
93        (c) \X did not give a hard partial match after matching one or more
94            characters at the end of the subject.
95    
96        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
97            a partial match for the string "\r".
98    
99        (e) When newline was set to CRLF, the metacharacter "." did not recognize
100            a partial match for a CR character at the end of the subject string.
101    
102    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
103        running pcretest, the text "(JIT)" added to the output whenever JIT is
104        actually used to run the match.
105    
106    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
107        or /S+[+] with a digit between 1 and 7.
108    
109    8.  OP_NOT now supports any UTF character not just single-byte ones.
110    
111    9.  (*MARK) control verb is now supported by the JIT compiler.
112    
113    10. The command "./RunTest list" lists the available tests without actually
114        running any of them. (Because I keep forgetting what they all are.)
115    
116    11. Add PCRE_INFO_MAXLOOKBEHIND.
117    
118    12. Applied a (slightly modified) user-supplied patch that improves performance
119        when the heap is used for recursion (compiled with --disable-stack-for-
120        recursion). Instead of malloc and free for each heap frame each time a
121        logical recursion happens, frames are retained on a chain and re-used where
122        possible. This sometimes gives as much as 30% improvement.
123    
124    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
125        call.
126    
127    14. As documented, (*COMMIT) is now confined to within a positive assertion.
128    
129    15. It is now possible to link pcretest with libedit as an alternative to
130        libreadline.
131    
132    16. (*COMMIT) control verb is now supported by the JIT compiler.
133    
134    17. The Unicode data tables have been updated to Unicode 6.1.0.
135    
136    18. Added --file-list option to pcregrep.
137    
138    19. Added binary file support to pcregrep, including the -a, --binary-files,
139        -I, and --text options.
140    
141    20. The madvise function is renamed for posix_madvise for QNX compatibility
142        reasons. Fixed by Giuseppe D'Angelo.
143    
144    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
145    
146    22. Changed the option for creating symbolic links for 16-bit man pages from
147        -s to -sf so that re-installing does not cause issues.
148    
149    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
150    
151    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
152        matches in certain environments (the workspace was not being correctly
153        retained). Also added to pcre_dfa_exec() a simple plausibility check on
154        some of the workspace data at the beginning of a restart.
155    
156    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
157        was not doing so when it should - probably a typo introduced by SVN 528
158        (change 8.10/14).
159    
160    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
161        \w+ when the character tables indicated that \x{c4} was a word character.
162        There were several related cases, all because the tests for doing a table
163        lookup were testing for characters less than 127 instead of 255.
164    
165    27. If a pattern contains capturing parentheses that are not used in a match,
166        their slots in the ovector are set to -1. For those that are higher than
167        any matched groups, this happens at the end of processing. In the case when
168        there were back references that the ovector was too small to contain
169        (causing temporary malloc'd memory to be used during matching), and the
170        highest capturing number was not used, memory off the end of the ovector
171        was incorrectly being set to -1. (It was using the size of the temporary
172        memory instead of the true size.)
173    
174    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
175        ovector size, it uses memory at the end of the block that it has got.
176    
177    29. Check for an overlong MARK name and give an error at compile time. The
178        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
179    
180    30. JIT compiler update.
181    
182    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
183        Rill for the patch.
184    
185    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
186    
187    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
188    
189    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
190        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
191        was enabled.
192    
193    35. Improve JIT code generation for greedy plus quantifier.
194    
195    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
196        1 to "aa" instead of to an empty string. The bug affected repeated groups
197        that could potentially match an empty string.
198    
199    37. Optimizing single character iterators in JIT.
200    
201    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
202        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
203        codepoints that are too big for the mode are faulted, and in a UTF mode,
204        disallowed codepoints are also faulted.
205    
206    39. If PCRE was compiled with UTF support, in three places in the DFA
207        matcher there was code that should only have been obeyed in UTF mode, but
208        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
209        processing when bytes with values greater than 127 were present. In 16-bit
210        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
211        both cases the values are those that cannot be the first data item in a UTF
212        character. The three items that might have provoked this were recursions,
213        possessively repeated groups, and atomic groups.
214    
215    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
216        and pcregrep, because some OS require shared objects to be explicitly
217        passed to ld, causing the link step to fail if they are not.
218    
219    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
220        mode, patterns that started with \h* or \R* might be incorrectly matched.
221    
222    
223    Version 8.30 04-February-2012
224    -----------------------------
225    
226    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
227        name is defined in ctype.h.
228    
229    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
230        only in quite long subpatterns.
231    
232    3.  Removed the function pcre_info(), which has been obsolete and deprecated
233        since it was replaced by pcre_fullinfo() in February 2000.
234    
235    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
236        match a (*MARK), and the match failed at the start of the subject, a
237        reference to memory before the start of the subject could occur. This bug
238        was introduced by fix 17 of release 8.21.
239    
240    5.  A reference to an unset group with zero minimum repetition was giving
241        totally wrong answers (in non-JavaScript-compatibility mode). For example,
242        /(another)?(\1?)test/ matched against "hello world test". This bug was
243        introduced in release 8.13.
244    
245    6.  Add support for 16-bit character strings (a large amount of work involving
246        many changes and refactorings).
247    
248    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
249        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
250        from a file.
251    
252    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
253        rounding is not applied in this particular case).
254    
255    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
256        if they appear, or are escaped, in patterns.
257    
258    10. Get rid of a number of -Wunused-but-set-variable warnings.
259    
260    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
261        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
262        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
263        also returns the mark "x". This bug applied to capturing parentheses,
264        non-capturing parentheses, and atomic parentheses. It also applied to some
265        assertions.
266    
267    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
268        information out of configure.ac instead of relying on pcre.h.generic, which
269        is not stored in the repository.
270    
271    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
272        -lreadline.
273    
274    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
275    
276    15. Applied Graycode's patch to put the top-level frame on the stack rather
277        than the heap when not using the stack for recursion. This gives a
278        performance improvement in many cases when recursion is not deep.
279    
280    16. Experimental code added to "pcretest -C" to output the stack frame size.
281    
282    
283    Version 8.21 12-Dec-2011
284    ------------------------
285    
286    1.  Updating the JIT compiler.
287    
288    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
289        are added as well.
290    
291    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
292        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
293        calling _pcre_jit_exec. Some extra comments are added.
294    
295    4.  (*MARK) settings inside atomic groups that do not contain any capturing
296        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
297        was introduced by change 18 for 8.20.
298    
299    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
300        ECMA-262 standard.
301    
302    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
303        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
304        This bug was probably introduced by change 9 of 8.13.
305    
306    7.  While fixing 6 above, I noticed that a number of other items were being
307        incorrectly rejected as "not fixed length". This arose partly because newer
308        opcodes had not been added to the fixed-length checking code. I have (a)
309        corrected the bug and added tests for these items, and (b) arranged for an
310        error to occur if an unknown opcode is encountered while checking for fixed
311        length instead of just assuming "not fixed length". The items that were
312        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
313        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
314        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
315    
316    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
317        being incorrectly compiled and would have given unpredicatble results.
318    
319    9.  A possessively repeated subpattern with minimum repeat count greater than
320        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
321        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
322        the first (A) could occur when it should not.
323    
324    10. Add a cast and remove a redundant test from the code.
325    
326    11. JIT should use pcre_malloc/pcre_free for allocation.
327    
328    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
329        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
330        is anything other than /usr, -L is still shown).
331    
332    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
333    
334    14. Perl does not support \N without a following name in a [] class; PCRE now
335        also gives an error.
336    
337    15. If a forward reference was repeated with an upper limit of around 2000,
338        it caused the error "internal error: overran compiling workspace". The
339        maximum number of forward references (including repeats) was limited by the
340        internal workspace, and dependent on the LINK_SIZE. The code has been
341        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
342        the default depends on LINK_SIZE. There is a new upper limit (for safety)
343        of around 200,000 forward references. While doing this, I also speeded up
344        the filling in of repeated forward references.
345    
346    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
347        incorrectly expecting the subject to contain another "a" after the start.
348    
349    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
350        in the match, the SKIP should be ignored. This was not happening; instead
351        the SKIP was being treated as NOMATCH. For patterns such as
352        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
353        tested.
354    
355    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
356        now much more compatible with Perl, in particular in cases where the result
357        is a non-match for a non-anchored pattern. For example, if
358        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
359        "m", where previously it did not return a name. A side effect of this
360        change is that for partial matches, the last encountered mark name is
361        returned, as for non matches. A number of tests that were previously not
362        Perl-compatible have been moved into the Perl-compatible test files. The
363        refactoring has had the pleasing side effect of removing one argument from
364        the match() function, thus reducing its stack requirements.
365    
366    19. If the /S+ option was used in pcretest to study a pattern using JIT,
367        subsequent uses of /S (without +) incorrectly behaved like /S+.
368    
369    21. Retrieve executable code size support for the JIT compiler and fixing
370        some warnings.
371    
372    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
373        not work when the shorter character appeared right at the end of the
374        subject string.
375    
376    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
377        systems.
378    
379    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
380        output it when the /M option is used in pcretest.
381    
382    25. The CheckMan script was not being included in the distribution. Also, added
383        an explicit "perl" to run Perl scripts from the PrepareRelease script
384        because this is reportedly needed in Windows.
385    
386    26. If study data was being save in a file and studying had not found a set of
387        "starts with" bytes for the pattern, the data written to the file (though
388        never used) was taken from uninitialized memory and so caused valgrind to
389        complain.
390    
391    27. Updated RunTest.bat as provided by Sheri Pierce.
392    
393    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
394    
395    29. Computation of memory usage for the table of capturing group names was
396        giving an unnecessarily large value.
397    
398    
399    Version 8.20 21-Oct-2011
400    ------------------------
401    
402    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
403        a POSIX class. After further experiments with Perl, which convinced me that
404        Perl has bugs and confusions, a closing square bracket is no longer allowed
405        in a POSIX name. This bug also affected patterns with classes that started
406        with full stops.
407    
408    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
409        captured substring, but while checking the failing first alternative,
410        substring 1 is temporarily captured. If the output vector supplied to
411        pcre_exec() was not big enough for this capture, the yield of the function
412        was still zero ("insufficient space for captured substrings"). This cannot
413        be totally fixed without adding another stack variable, which seems a lot
414        of expense for a edge case. However, I have improved the situation in cases
415        such as /(a)(b)x|abc/ matched against "abc", where the return code
416        indicates that fewer than the maximum number of slots in the ovector have
417        been set.
418    
419    3.  Related to (2) above: when there are more back references in a pattern than
420        slots in the output vector, pcre_exec() uses temporary memory during
421        matching, and copies in the captures as far as possible afterwards. It was
422        using the entire output vector, but this conflicts with the specification
423        that only 2/3 is used for passing back captured substrings. Now it uses
424        only the first 2/3, for compatibility. This is, of course, another edge
425        case.
426    
427    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
428        main code base, and can be used by building with --enable-jit. When this is
429        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
430        runtime --no-jit option is given.
431    
432    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
433        ovector, the return from the function was zero, implying that there were
434        other matches that did not fit. The correct "exactly full" value is now
435        returned.
436    
437    6.  If a subpattern that was called recursively or as a subroutine contained
438        (*PRUNE) or any other control that caused it to give a non-standard return,
439        invalid errors such as "Error -26 (nested recursion at the same subject
440        position)" or even infinite loops could occur.
441    
442    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
443        computing the minimum length on reaching *ACCEPT, and so ended up with the
444        wrong value of 1 rather than 0. Further investigation indicates that
445        computing a minimum subject length in the presence of *ACCEPT is difficult
446        (think back references, subroutine calls), and so I have changed the code
447        so that no minimum is registered for a pattern that contains *ACCEPT.
448    
449    8.  If (*THEN) was present in the first (true) branch of a conditional group,
450        it was not handled as intended. [But see 16 below.]
451    
452    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
453        Sheri Pierce.
454    
455    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
456        the first byte in a match must be "a".
457    
458    11. Change 17 for 8.13 increased the recursion depth for patterns like
459        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
460        pattern contains any instances of (*THEN). If it does not, the old
461        optimizations are restored. It would be nice to do this on a per-group
462        basis, but at the moment that is not feasible.
463    
464    12. In some environments, the output of pcretest -C is CRLF terminated. This
465        broke RunTest's code that checks for the link size. A single white space
466        character after the value is now allowed for.
467    
468    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
469        For "fr", it uses the Windows-specific input and output files.
470    
471    14. If (*THEN) appeared in a group that was called recursively or as a
472        subroutine, it did not work as intended. [But see next item.]
473    
474    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
475        pattern fragments (but not containing any | characters). If A and B are
476        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
477        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
478        D. In other words, Perl considers parentheses that do not contain any |
479        characters to be part of a surrounding alternative, whereas PCRE was
480        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
481        differently. PCRE now behaves in the same way as Perl, except in the case
482        of subroutine/recursion calls such as (?1) which have in any case always
483        been different (but PCRE had them first :-).
484    
485    16. Related to 15 above: Perl does not treat the | in a conditional group as
486        creating alternatives. Such a group is treated in the same way as an
487        ordinary group without any | characters when processing (*THEN). PCRE has
488        been changed to match Perl's behaviour.
489    
490    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
491        RunGrepTest script failed.
492    
493    18. Change 22 for version 13 caused atomic groups to use more stack. This is
494        inevitable for groups that contain captures, but it can lead to a lot of
495        stack use in large patterns. The old behaviour has been restored for atomic
496        groups that do not contain any capturing parentheses.
497    
498    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
499        suppress the check for a minimum subject length at run time. (If it was
500        given to pcre_exec() or pcre_dfa_exec() it did work.)
501    
502    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
503        fail to work when decoding hex characters in data strings in EBCDIC
504        environments.
505    
506    21. It appears that in at least one Mac OS environment, the isxdigit() function
507        is implemented as a macro that evaluates to its argument more than once,
508        contravening the C 90 Standard (I haven't checked a later standard). There
509        was an instance in pcretest which caused it to go wrong when processing
510        \x{...} escapes in subject strings. The has been rewritten to avoid using
511        things like p++ in the argument of isxdigit().
512    
513    
514    Version 8.13 16-Aug-2011
515    ------------------------
516    
517    1.  The Unicode data tables have been updated to Unicode 6.0.0.
518    
519    2.  Two minor typos in pcre_internal.h have been fixed.
520    
521    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
522        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
523        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
524    
525    4.  There were a number of related bugs in the code for matching backrefences
526        caselessly in UTF-8 mode when codes for the characters concerned were
527        different numbers of bytes. For example, U+023A and U+2C65 are an upper
528        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
529        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
530        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
531        2-byte code at the end of the subject (it thought there wasn't enough data
532        left).
533    
534    5.  Comprehensive information about what went wrong is now returned by
535        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
536        as the output vector has at least 2 elements. The offset of the start of
537        the failing character and a reason code are placed in the vector.
538    
539    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
540        now returned is for the first byte of the failing character, instead of the
541        last byte inspected. This is an incompatible change, but I hope it is small
542        enough not to be a problem. It makes the returned offset consistent with
543        pcre_exec() and pcre_dfa_exec().
544    
545    7.  pcretest now gives a text phrase as well as the error number when
546        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
547        failure, the offset and reason code are output.
548    
549    8.  When \R was used with a maximizing quantifier it failed to skip backwards
550        over a \r\n pair if the subsequent match failed. Instead, it just skipped
551        back over a single character (\n). This seems wrong (because it treated the
552        two characters as a single entity when going forwards), conflicts with the
553        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
554        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
555        has been changed.
556    
557    9.  Some internal refactoring has changed the processing so that the handling
558        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
559        time (the PCRE_DOTALL option was changed this way some time ago: version
560        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
561        which was always a bit of a fudge. It also means that there is one less
562        argument for the match() function, which reduces its stack requirements
563        slightly. This change also fixes an incompatibility with Perl: the pattern
564        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
565    
566    10. More internal refactoring has drastically reduced the number of recursive
567        calls to match() for possessively repeated groups such as (abc)++ when
568        using pcre_exec().
569    
570    11. While implementing 10, a number of bugs in the handling of groups were
571        discovered and fixed:
572    
573        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
574        (a|)*(?1) gave a compile-time internal error.
575        ((a|)+)+  did not notice that the outer group could match an empty string.
576        (^a|^)+   was not marked as anchored.
577        (.*a|.*)+ was not marked as matching at start or after a newline.
578    
579    12. Yet more internal refactoring has removed another argument from the match()
580        function. Special calls to this function are now indicated by setting a
581        value in a variable in the "match data" data block.
582    
583    13. Be more explicit in pcre_study() instead of relying on "default" for
584        opcodes that mean there is no starting character; this means that when new
585        ones are added and accidentally left out of pcre_study(), testing should
586        pick them up.
587    
588    14. The -s option of pcretest has been documented for ages as being an old
589        synonym of -m (show memory usage). I have changed it to mean "force study
590        for every regex", that is, assume /S for every regex. This is similar to -i
591        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
592        using it. It makes it easier to run collections of tests with and without
593        study enabled, and thereby test pcre_study() more easily. All the standard
594        tests are now run with and without -s (but some patterns can be marked as
595        "never study" - see 20 below).
596    
597    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
598        restoration of the capturing data to the outer values was not happening
599        correctly.
600    
601    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
602        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
603        pattern had matched an empty string, and so incorrectly returned a no
604        match.
605    
606    17. There was optimizing code for the last branch of non-capturing parentheses,
607        and also for the obeyed branch of a conditional subexpression, which used
608        tail recursion to cut down on stack usage. Unfortunately, now that there is
609        the possibility of (*THEN) occurring in these branches, tail recursion is
610        no longer possible because the return has to be checked for (*THEN). These
611        two optimizations have therefore been removed. [But see 8.20/11 above.]
612    
613    18. If a pattern containing \R was studied, it was assumed that \R always
614        matched two bytes, thus causing the minimum subject length to be
615        incorrectly computed because \R can also match just one byte.
616    
617    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
618        was incorrectly computed.
619    
620    20. If /S is present twice on a test pattern in pcretest input, it now
621        *disables* studying, thereby overriding the use of -s on the command line
622        (see 14 above). This is necessary for one or two tests to keep the output
623        identical in both cases.
624    
625    21. When (*ACCEPT) was used in an assertion that matched an empty string and
626        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
627    
628    22. When an atomic group that contained a capturing parenthesis was
629        successfully matched, but the branch in which it appeared failed, the
630        capturing was not being forgotten if a higher numbered group was later
631        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
632        group 1 to "a", when in fact it should be unset. This applied to multi-
633        branched capturing and non-capturing groups, repeated or not, and also to
634        positive assertions (capturing in negative assertions does not happen
635        in PCRE) and also to nested atomic groups.
636    
637    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
638        subject after a captured substring, to make it easier to tell which of a
639        number of identical substrings has been captured.
640    
641    24. The way atomic groups are processed by pcre_exec() has been changed so that
642        if they are repeated, backtracking one repetition now resets captured
643        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
644        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
645        "aaa". Previously, it would have been "a". As part of this code
646        refactoring, the way recursive calls are handled has also been changed.
647    
648    25. If an assertion condition captured any substrings, they were not passed
649        back unless some other capturing happened later. For example, if
650        (?(?=(a))a) was matched against "a", no capturing was returned.
651    
652    26. When studying a pattern that contained subroutine calls or assertions,
653        the code for finding the minimum length of a possible match was handling
654        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
655        group 1 called group 2 while simultaneously a separate group 2 called group
656        1). A stack overflow occurred in this case. I have fixed this by limiting
657        the recursion depth to 10.
658    
659    27. Updated RunTest.bat in the distribution to the version supplied by Tom
660        Fortmann. This supports explicit test numbers on the command line, and has
661        argument validation and error reporting.
662    
663    28. An instance of \X with an unlimited repeat could fail if at any point the
664        first character it looked at was a mark character.
665    
666    29. Some minor code refactoring concerning Unicode properties and scripts
667        should reduce the stack requirement of match() slightly.
668    
669    30. Added the '=' option to pcretest to check the setting of unused capturing
670        slots at the end of the pattern, which are documented as being -1, but are
671        not included in the return count.
672    
673    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
674        compiled something random. Now it gives a compile-time error (as does
675        Perl).
676    
677    32. A *MARK encountered during the processing of a positive assertion is now
678        recorded and passed back (compatible with Perl).
679    
680    33. If --only-matching or --colour was set on a pcregrep call whose pattern
681        had alternative anchored branches, the search for a second match in a line
682        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
683        matched the line "0102" twice. The same bug affected patterns that started
684        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
685        twice.
686    
687    34. Previously, PCRE did not allow quantification of assertions. However, Perl
688        does, and because of capturing effects, quantifying parenthesized
689        assertions may at times be useful. Quantifiers are now allowed for
690        parenthesized assertions.
691    
692    35. A minor code tidy in pcre_compile() when checking options for \R usage.
693    
694    36. \g was being checked for fancy things in a character class, when it should
695        just be a literal "g".
696    
697    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
698        appearance of a nested POSIX class supersedes an apparent external class.
699        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
700        unescaped square brackets may also appear as part of class names. For
701        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
702        more like Perl. (But see 8.20/1 above.)
703    
704    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
705        was because it thought it was \N{name}, which is not supported).
706    
707    39. Add minix to OS list not supporting the -S option in pcretest.
708    
709    40. PCRE tries to detect cases of infinite recursion at compile time, but it
710        cannot analyze patterns in sufficient detail to catch mutual recursions
711        such as ((?1))((?2)). There is now a runtime test that gives an error if a
712        subgroup is called recursively as a subpattern for a second time at the
713        same position in the subject string. In previous releases this might have
714        been caught by the recursion limit, or it might have run out of stack.
715    
716    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
717        happen only once. PCRE was, however incorrectly giving a compile time error
718        "recursive call could loop indefinitely" because it cannot analyze the
719        pattern in sufficient detail. The compile time test no longer happens when
720        PCRE is compiling a conditional subpattern, but actual runaway loops are
721        now caught at runtime (see 40 above).
722    
723    42. It seems that Perl allows any characters other than a closing parenthesis
724        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
725        has been changed to be the same.
726    
727    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
728        as not to get warnings when autogen.sh is called. Also changed
729        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
730    
731    44. To help people who use pcregrep to scan files containing exceedingly long
732        lines, the following changes have been made:
733    
734        (a) The default value of the buffer size parameter has been increased from
735            8K to 20K. (The actual buffer used is three times this size.)
736    
737        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
738            PCRE is built.
739    
740        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
741            to be set at run time.
742    
743        (d) Numerical values in pcregrep options can be followed by K or M, for
744            example --buffer-size=50K.
745    
746        (e) If a line being scanned overflows pcregrep's buffer, an error is now
747            given and the return code is set to 2.
748    
749    45. Add a pointer to the latest mark to the callout data block.
750    
751    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
752        partial match of an empty string instead of no match. This was specific to
753        the use of ".".
754    
755    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
756        complete match instead of a partial match. This bug was dependent on both
757        the PCRE_UTF8 and PCRE_DOTALL options being set.
758    
759    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
760        starting byte set, because \b was not being ignored.
761    
762    
763    Version 8.12 15-Jan-2011
764    ------------------------
765    
766    1.  Fixed some typos in the markup of the man pages, and wrote a script that
767        checks for such things as part of the documentation building process.
768    
769    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
770        --match-limit and --recursion-limit options (added for 8.11). In
771        particular, this made one of the standard tests fail. (The integer value
772        went into the wrong half of a long int.)
773    
774    3.  If the --colour option was given to pcregrep with -v (invert match), it
775        did strange things, either producing crazy output, or crashing. It should,
776        of course, ignore a request for colour when reporting lines that do not
777        match.
778    
779    4.  Another pcregrep bug caused similar problems if --colour was specified with
780        -M (multiline) and the pattern match finished with a line ending.
781    
782    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
783        matched in multiline mode, the following line was shown as part of the
784        match. This seems wrong, so I have changed it.
785    
786    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
787        the check for further matches in the same line (so they could be coloured)
788        to overrun the end of the current line. If another match was found, it was
789        incorrectly shown (and then shown again when found in the next line).
790    
791    7.  If pcregrep was compiled under Windows, there was a reference to the
792        function pcregrep_exit() before it was defined. I am assuming this was
793        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
794        reported by a user. I've moved the definition above the reference.
795    
796    
797    Version 8.11 10-Dec-2010
798    ------------------------
799    
800    1.  (*THEN) was not working properly if there were untried alternatives prior
801        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
802        backtracked to try for "b" instead of moving to the next alternative branch
803        at the same level (in this case, to look for "c"). The Perl documentation
804        is clear that when (*THEN) is backtracked onto, it goes to the "next
805        alternative in the innermost enclosing group".
806    
807    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
808        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
809        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
810        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
811        (*THEN).
812    
813    3.  If \s appeared in a character class, it removed the VT character from
814        the class, even if it had been included by some previous item, for example
815        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
816        of \s, but is part of the POSIX "space" class.)
817    
818    4.  A partial match never returns an empty string (because you can always
819        match an empty string at the end of the subject); however the checking for
820        an empty string was starting at the "start of match" point. This has been
821        changed to the "earliest inspected character" point, because the returned
822        data for a partial match starts at this character. This means that, for
823        example, /(?<=abc)def/ gives a partial match for the subject "abc"
824        (previously it gave "no match").
825    
826    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
827        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
828        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
829        has an implication that the given string is incomplete (because a partial
830        match is preferred over a full match). For this reason, these items now
831        give a partial match in this situation. [Aside: previously, the one case
832        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
833        match rather than a full match, which was wrong by the old rules, but is
834        now correct.]
835    
836    6.  There was a bug in the handling of #-introduced comments, recognized when
837        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
838        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
839        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
840        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
841        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
842        places in pcre_compile().
843    
844    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
845        comments when looking ahead for named forward references to subpatterns,
846        the only newline sequence it recognized was NL. It now handles newlines
847        according to the set newline convention.
848    
849    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
850        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
851        cater for a lack of strerror(). These oversights have been fixed.
852    
853    9.  Added --match-limit and --recursion-limit to pcregrep.
854    
855    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
856    
857    11. When the -o option was used, pcregrep was setting a return code of 1, even
858        when matches were found, and --line-buffered was not being honoured.
859    
860    12. Added an optional parentheses number to the -o and --only-matching options
861        of pcregrep.
862    
863    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
864        can match an empty string. The code to do it in pcretest and pcredemo
865        needed fixing:
866    
867        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
868            only one byte after an empty string match just before CRLF (this case
869            just got forgotten; "any" and "anycrlf" were OK).
870    
871        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
872            mode when an empty string match preceded an ASCII character followed by
873            a non-ASCII character. (The code for advancing by one character rather
874            than one byte was nonsense.)
875    
876        (c) The pcredemo.c sample program did not have any code at all to handle
877            the cases when CRLF is a valid newline sequence.
878    
879    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
880        as a starting offset was within the subject string. There is now a new
881        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
882        negative or greater than the length of the string. In order to test this,
883        pcretest is extended to allow the setting of negative starting offsets.
884    
885    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
886        starting offset points to the beginning of a UTF-8 character was
887        unnecessarily clumsy. I tidied it up.
888    
889    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
890        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
891    
892    17. Nobody had reported that the --include_dir option, which was added in
893        release 7.7 should have been called --include-dir (hyphen, not underscore)
894        for compatibility with GNU grep. I have changed it to --include-dir, but
895        left --include_dir as an undocumented synonym, and the same for
896        --exclude-dir, though that is not available in GNU grep, at least as of
897        release 2.5.4.
898    
899    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
900        characters from a string of bytes) have been redefined so as not to use
901        loops, in order to improve performance in some environments. At the same
902        time, I abstracted some of the common code into auxiliary macros to save
903        repetition (this should not affect the compiled code).
904    
905    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
906        compile-time error is now given if \c is not followed by an ASCII
907        character, that is, a byte less than 128. (In EBCDIC mode, the code is
908        different, and any byte value is allowed.)
909    
910    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
911        START_OPTIMIZE option, which is now allowed at compile time - but just
912        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
913        to pcregrep and other applications that have no direct access to PCRE
914        options. The new /Y option in pcretest sets this option when calling
915        pcre_compile().
916    
917    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
918        back references. Groups containing recursive back references were forced to
919        be atomic by that change, but in the case of named groups, the amount of
920        memory required was incorrectly computed, leading to "Failed: internal
921        error: code overflow". This has been fixed.
922    
923    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
924        pcretest.c, to avoid build problems in some Borland environments.
925    
926    
927    Version 8.10 25-Jun-2010
928    ------------------------
929    
930    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
931        THEN.
932    
933    2.  (*ACCEPT) was not working when inside an atomic group.
934    
935    3.  Inside a character class, \B is treated as a literal by default, but
936        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
937        causes the error). The code is unchanged, but I tidied the documentation.
938    
939    4.  Inside a character class, PCRE always treated \R and \X as literals,
940        whereas Perl faults them if its -w option is set. I have changed PCRE so
941        that it faults them when PCRE_EXTRA is set.
942    
943    5.  Added support for \N, which always matches any character other than
944        newline. (It is the same as "." when PCRE_DOTALL is not set.)
945    
946    6.  When compiling pcregrep with newer versions of gcc which may have
947        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
948        declared with attribute warn_unused_result" were given. Just casting the
949        result to (void) does not stop the warnings; a more elaborate fudge is
950        needed. I've used a macro to implement this.
951    
952    7.  Minor change to pcretest.c to avoid a compiler warning.
953    
954    8.  Added four artifical Unicode properties to help with an option to make
955        \s etc use properties (see next item). The new properties are: Xan
956        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
957    
958    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
959        use Unicode properties. (*UCP) at the start of a pattern can be used to set
960        this option. Modified pcretest to add /W to test this facility. Added
961        REG_UCP to make it available via the POSIX interface.
962    
963    10. Added --line-buffered to pcregrep.
964    
965    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
966        studied, and the match started with a letter with a code point greater than
967        127 whose first byte was different to the first byte of the other case of
968        the letter, the other case of this starting letter was not recognized
969        (#976).
970    
971    12. If a pattern that was studied started with a repeated Unicode property
972        test, for example, \p{Nd}+, there was the theoretical possibility of
973        setting up an incorrect bitmap of starting bytes, but fortunately it could
974        not have actually happened in practice until change 8 above was made (it
975        added property types that matched character-matching opcodes).
976    
977    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
978        possible starting bytes for non-anchored patterns.
979    
980    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
981        \R, and also a number of cases that involve Unicode properties, both
982        explicit and implicit when PCRE_UCP is set.
983    
984    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
985        input, it could crash or give wrong results if characters with values
986        greater than 0xc0 were present in the subject string. (Detail: it assumed
987        UTF-8 input when processing these items.)
988    
989    16. Added a lot of (int) casts to avoid compiler warnings in systems where
990        size_t is 64-bit (#991).
991    
992    17. Added a check for running out of memory when PCRE is compiled with
993        --disable-stack-for-recursion (#990).
994    
995    18. If the last data line in a file for pcretest does not have a newline on
996        the end, a newline was missing in the output.
997    
998    19. The default pcre_chartables.c file recognizes only ASCII characters (values
999        less than 128) in its various bitmaps. However, there is a facility for
1000        generating tables according to the current locale when PCRE is compiled. It
1001        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1002        characters, are recognized by isspace() and therefore were getting set in
1003        these tables, and indeed these tables seem to approximate to ISO 8859. This
1004        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1005        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1006        which of course cannot start UTF-8 characters. I have changed the code so
1007        that only real ASCII characters (less than 128) and the correct starting
1008        bytes for UTF-8 encodings are set for characters greater than 127 when in
1009        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1010        altogether.)
1011    
1012    20. Added the /T option to pcretest so as to be able to run tests with non-
1013        standard character tables, thus making it possible to include the tests
1014        used for 19 above in the standard set of tests.
1015    
1016    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1017        reference to a subpattern the other side of a comment that contains an
1018        opening parenthesis caused either an internal compiling error, or a
1019        reference to the wrong subpattern.
1020    
1021    
1022    Version 8.02 19-Mar-2010
1023    ------------------------
1024    
1025    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1026    
1027    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1028        configured.
1029    
1030    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1031        original author of that file, following a query about its status.
1032    
1033    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1034        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1035    
1036    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1037        quantifier applied to a forward-referencing subroutine call, could compile
1038        incorrect code or give the error "internal error: previously-checked
1039        referenced subpattern not found".
1040    
1041    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1042        variables to point to external functions. For these systems, therefore,
1043        pcre_malloc etc. are now initialized to local functions that call the
1044        relevant global functions.
1045    
1046    7.  There were two entries missing in the vectors called coptable and poptable
1047        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1048        I've fixed the data, and added a kludgy way of testing at compile time that
1049        the lengths are correct (equal to the number of opcodes).
1050    
1051    8.  Following on from 7, I added a similar kludge to check the length of the
1052        eint vector in pcreposix.c.
1053    
1054    9.  Error texts for pcre_compile() are held as one long string to avoid too
1055        much relocation at load time. To find a text, the string is searched,
1056        counting zeros. There was no check for running off the end of the string,
1057        which could happen if a new error number was added without updating the
1058        string.
1059    
1060    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1061    
1062    11. \K was not working if it appeared in an atomic group or in a group that
1063        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1064        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1065        the assertion is positive, but not if it is negative.
1066    
1067    12. Change 11 fortuitously reduced the size of the stack frame used in the
1068        "match()" function of pcre_exec.c by one pointer. Forthcoming
1069        implementation of support for (*MARK) will need an extra pointer on the
1070        stack; I have reserved it now, so that the stack frame size does not
1071        decrease.
1072    
1073    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1074        item in branch that calls a recursion is a subroutine call - as in the
1075        second branch in the above example - was incorrectly given the compile-
1076        time error "recursive call could loop indefinitely" because pcre_compile()
1077        was not correctly checking the subroutine for matching a non-empty string.
1078    
1079    14. The checks for overrunning compiling workspace could trigger after an
1080        overrun had occurred. This is a "should never occur" error, but it can be
1081        triggered by pathological patterns such as hundreds of nested parentheses.
1082        The checks now trigger 100 bytes before the end of the workspace.
1083    
1084    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1085    
1086    
1087    Version 8.01 19-Jan-2010
1088    ------------------------
1089    
1090    1.  If a pattern contained a conditional subpattern with only one branch (in
1091        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1092        computed the wrong minimum data length (which is of course zero for such
1093        subpatterns). This could cause incorrect "no match" results.
1094    
1095    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1096        the pattern is reset in the first branch, pcre_compile() failed with
1097        "internal error: code overflow at offset...". This happened only when
1098        the reset was to the original external option setting. (An optimization
1099        abstracts leading options settings into an external setting, which was the
1100        cause of this.)
1101    
1102    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1103        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1104        assertion pattern did not match (meaning that the assertion was true), it
1105        was incorrectly treated as false if the SKIP had been reached during the
1106        matching. This also applied to assertions used as conditions.
1107    
1108    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1109        assertion subpattern, including such a pattern used as a condition,
1110        unpredictable results occurred, instead of the error return
1111        PCRE_ERROR_DFA_UITEM.
1112    
1113    5.  The C++ GlobalReplace function was not working like Perl for the special
1114        situation when an empty string is matched. It now does the fancy magic
1115        stuff that is necessary.
1116    
1117    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1118        removed. (These were left over from very, very early versions of PCRE.)
1119    
1120    7.  Some cosmetic changes to the code to make life easier when compiling it
1121        as part of something else:
1122    
1123        (a) Change DEBUG to PCRE_DEBUG.
1124    
1125        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1126            called "current" as "current_branch", to prevent a collision with the
1127            Linux macro when compiled as a kernel module.
1128    
1129        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1130            prevent a collision with the Linux macro when compiled as a kernel
1131            module.
1132    
1133    8.  In pcre_compile() there are some checks for integer overflows that used to
1134        cast potentially large values to (double). This has been changed to that
1135        when building, a check for int64_t is made, and if it is found, it is used
1136        instead, thus avoiding the use of floating point arithmetic. (There is no
1137        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1138        double.
1139    
1140    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1141        2005 (difference between two addresses compared to an unsigned value).
1142    
1143    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1144        custom one, because of the following reported problem in Windows:
1145    
1146          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1147              under Win32.
1148          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1149              therefore missing the function definition.
1150          - The compiler thus generates a "C" signature for the test function.
1151          - The linker fails to find the "C" function.
1152          - PCRE fails to configure if asked to do so against libbz2.
1153    
1154    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1155        messages were output:
1156    
1157          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1158          rerunning libtoolize, to keep the correct libtool macros in-tree.
1159          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1160    
1161        I have done both of these things.
1162    
1163    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1164        most of the time, it *can* run out if it is given a pattern that contains a
1165        runaway infinite recursion. I updated the discussion in the pcrestack man
1166        page.
1167    
1168    13. Now that we have gone to the x.xx style of version numbers, the minor
1169        version may start with zero. Using 08 or 09 is a bad idea because users
1170        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1171        interpreted as invalid octal numbers. I've updated the previous comment in
1172        configure.ac, and also added a check that gives an error if 08 or 09 are
1173        used.
1174    
1175    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1176        causing partial matching to fail when the end of the subject matched \W
1177        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1178    
1179    15. There were some discrepancies between the declarations in pcre_internal.h
1180        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1181        their definitions. The declarations used "const uschar *" and the
1182        definitions used USPTR. Even though USPTR is normally defined as "const
1183        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1184        reported that: "This difference in casting confuses some C++ compilers, for
1185        example, SunCC recognizes above declarations as different functions and
1186        generates broken code for hbpcre." I have changed the declarations to use
1187        USPTR.
1188    
1189    16. GNU libtool is named differently on some systems. The autogen.sh script now
1190        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1191        (FreeBSD).
1192    
1193    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1194        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1195        comment: "Figure out how to create a longlong from a string: strtoll and
1196        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1197        instance, but it only takes 2 args instead of 3!"
1198    
1199    18. A subtle bug concerned with back references has been fixed by a change of
1200        specification, with a corresponding code fix. A pattern such as
1201        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1202        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1203        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1204        same bug. Such groups have to be quantified to be useful, or contained
1205        inside another quantified group. (If there's no repetition, the reference
1206        can never match.) The problem arises because, having left the group and
1207        moved on to the rest of the pattern, a later failure that backtracks into
1208        the group uses the captured value from the final iteration of the group
1209        rather than the correct earlier one. I have fixed this in PCRE by forcing
1210        any group that contains a reference to itself to be an atomic group; that
1211        is, there cannot be any backtracking into it once it has completed. This is
1212        similar to recursive and subroutine calls.
1213    
1214    
1215    Version 8.00 19-Oct-09
1216  ----------------------  ----------------------
1217    
1218  1.  The table for translating pcre_compile() error codes into POSIX error codes  1.  The table for translating pcre_compile() error codes into POSIX error codes
1219      was out-of-date, and there was no check on the pcre_compile() error code      was out-of-date, and there was no check on the pcre_compile() error code
1220      being within the table. This could lead to an OK return being given in      being within the table. This could lead to an OK return being given in
1221      error.      error.
1222    
1223  2.  Changed the call to open a subject file in pcregrep from fopen(pathname,  2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1224      "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests      "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1225      in a Windows environment.      in a Windows environment.
1226    
1227  3.  The pcregrep --count option prints the count for each file even when it is  3.  The pcregrep --count option prints the count for each file even when it is
1228      zero, as does GNU grep. However, pcregrep was also printing all files when      zero, as does GNU grep. However, pcregrep was also printing all files when
1229      --files-with-matches was added. Now, when both options are given, it prints      --files-with-matches was added. Now, when both options are given, it prints
1230      counts only for those files that have at least one match. (GNU grep just      counts only for those files that have at least one match. (GNU grep just
1231      prints the file name in this circumstance, but including the count seems      prints the file name in this circumstance, but including the count seems
1232      more useful - otherwise, why use --count?) Also ensured that the      more useful - otherwise, why use --count?) Also ensured that the
1233      combination -clh just lists non-zero counts, with no names.      combination -clh just lists non-zero counts, with no names.
1234    
1235  4.  The long form of the pcregrep -F option was incorrectly implemented as  4.  The long form of the pcregrep -F option was incorrectly implemented as
1236      --fixed_strings instead of --fixed-strings. This is an incompatible change,      --fixed_strings instead of --fixed-strings. This is an incompatible change,
1237      but it seems right to fix it, and I didn't think it was worth preserving      but it seems right to fix it, and I didn't think it was worth preserving
1238      the old behaviour.      the old behaviour.
1239    
1240  5.  The command line items --regex=pattern and --regexp=pattern were not  5.  The command line items --regex=pattern and --regexp=pattern were not
1241      recognized by pcregrep, which required --regex pattern or --regexp pattern      recognized by pcregrep, which required --regex pattern or --regexp pattern
1242      (with a space rather than an '='). The man page documented the '=' forms,      (with a space rather than an '='). The man page documented the '=' forms,
1243      which are compatible with GNU grep; these now work.      which are compatible with GNU grep; these now work.
1244    
1245  6.  No libpcreposix.pc file was created for pkg-config; there was just  6.  No libpcreposix.pc file was created for pkg-config; there was just
1246      libpcre.pc and libpcrecpp.pc. The omission has been rectified.      libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1247    
1248  7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size  7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1249      when UCP support is not needed, by modifying the Python script that      when UCP support is not needed, by modifying the Python script that
1250      generates it from Unicode data files. This should not matter if the module      generates it from Unicode data files. This should not matter if the module
1251      is correctly used as a library, but I received one complaint about 50K of      is correctly used as a library, but I received one complaint about 50K of
1252      unwanted data. My guess is that the person linked everything into his      unwanted data. My guess is that the person linked everything into his
1253      program rather than using a library. Anyway, it does no harm.      program rather than using a library. Anyway, it does no harm.
1254    
1255  8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger  8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1256      was a minimum greater than 1 for a wide character in a possessive      was a minimum greater than 1 for a wide character in a possessive
1257      repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8      repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1258      which had an unlimited repeat of a nested, fixed maximum repeat of a wide      which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1259      character. Chaos in the form of incorrect output or a compiling loop could      character. Chaos in the form of incorrect output or a compiling loop could
1260      result.      result.
1261    
1262  9.  The restrictions on what a pattern can contain when partial matching is  9.  The restrictions on what a pattern can contain when partial matching is
1263      requested for pcre_exec() have been removed. All patterns can now be      requested for pcre_exec() have been removed. All patterns can now be
1264      partially matched by this function. In addition, if there are at least two      partially matched by this function. In addition, if there are at least two
1265      slots in the offset vector, the offset of the earliest inspected character      slots in the offset vector, the offset of the earliest inspected character
1266      for the match and the offset of the end of the subject are set in them when      for the match and the offset of the end of the subject are set in them when
1267      PCRE_ERROR_PARTIAL is returned.      PCRE_ERROR_PARTIAL is returned.
1268    
1269  10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is  10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1270      synonymous with PCRE_PARTIAL, for backwards compatibility, and      synonymous with PCRE_PARTIAL, for backwards compatibility, and
1271      PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,      PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1272      and may be more useful for multi-segment matching, especially with      and may be more useful for multi-segment matching.
1273      pcre_exec().  
1274    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1275  11. Partial matching with pcre_exec() is now more intuitive. A partial match      used to be given if ever the end of the subject was reached; now it is
1276      used to be given if ever the end of the subject was reached; now it is      given only if matching could not proceed because another character was
1277      given only if matching could not proceed because another character was      needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1278      needed. This makes a difference in some odd cases such as Z(*FAIL) with the      string "Z", which now yields "no match" instead of "partial match". In the
1279      string "Z", which now yields "no match" instead of "partial match". In the      case of pcre_dfa_exec(), "no match" is given if every matching path for the
1280      case of pcre_dfa_exec(), "no match" is given if every matching path for the      final character ended with (*FAIL).
1281      final character ended with (*FAIL).  
   
1282  12. Restarting a match using pcre_dfa_exec() after a partial match did not work  12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1283      if the pattern had a "must contain" character that was already found in the      if the pattern had a "must contain" character that was already found in the
1284      earlier partial match, unless partial matching was again requested. For      earlier partial match, unless partial matching was again requested. For
1285      example, with the pattern /dog.(body)?/, the "must contain" character is      example, with the pattern /dog.(body)?/, the "must contain" character is
1286      "g". If the first part-match was for the string "dog", restarting with      "g". If the first part-match was for the string "dog", restarting with
1287      "sbody" failed. This bug has been fixed.      "sbody" failed. This bug has been fixed.
1288    
1289  13. The string returned by pcre_dfa_exec() after a partial match has been  13. The string returned by pcre_dfa_exec() after a partial match has been
1290      changed so that it starts at the first inspected character rather than the      changed so that it starts at the first inspected character rather than the
1291      first character of the match. This makes a difference only if the pattern      first character of the match. This makes a difference only if the pattern
1292      starts with a lookbehind assertion or \b or \B (\K is not supported by      starts with a lookbehind assertion or \b or \B (\K is not supported by
1293      pcre_dfa_exec()). It's an incompatible change, but it makes the two      pcre_dfa_exec()). It's an incompatible change, but it makes the two
1294      matching functions compatible, and I think it's the right thing to do.      matching functions compatible, and I think it's the right thing to do.
1295    
1296  14. Added a pcredemo man page, created automatically from the pcredemo.c file,  14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1297      so that the demonstration program is easily available in environments where      so that the demonstration program is easily available in environments where
1298      PCRE has not been installed from source.      PCRE has not been installed from source.
1299    
1300  15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,  15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1301      libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared      libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1302      library.      library.
1303    
1304  16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.  16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1305      It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it      It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1306      is not the first non-POSIX option to be added. Clearly some people find      is not the first non-POSIX option to be added. Clearly some people find
1307      these options useful.      these options useful.
1308    
1309  17. If a caller to the POSIX matching function regexec() passes a non-zero  17. If a caller to the POSIX matching function regexec() passes a non-zero
1310      value for nmatch with a NULL value for pmatch, the value of      value for nmatch with a NULL value for pmatch, the value of
1311      nmatch is forced to zero.      nmatch is forced to zero.
1312    
1313  18. RunGrepTest did not have a test for the availability of the -u option of  18. RunGrepTest did not have a test for the availability of the -u option of
1314      the diff command, as RunTest does. It now checks in the same way as      the diff command, as RunTest does. It now checks in the same way as
1315      RunTest, and also checks for the -b option.      RunTest, and also checks for the -b option.
1316    
1317  19. If an odd number of negated classes containing just a single character  19. If an odd number of negated classes containing just a single character
1318      interposed, within parentheses, between a forward reference to a named      interposed, within parentheses, between a forward reference to a named
1319      subpattern and the definition of the subpattern, compilation crashed with      subpattern and the definition of the subpattern, compilation crashed with
1320      an internal error, complaining that it could not find the referenced      an internal error, complaining that it could not find the referenced
1321      subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.      subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1322      [The bug was that it was starting one character too far in when skipping      [The bug was that it was starting one character too far in when skipping
1323      over the character class, thus treating the ] as data rather than      over the character class, thus treating the ] as data rather than
1324      terminating the class. This meant it could skip too much.]      terminating the class. This meant it could skip too much.]
1325    
1326  20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the  20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1327      /g option in pcretest when the pattern contains \K, which makes it possible      /g option in pcretest when the pattern contains \K, which makes it possible
1328      to have an empty string match not at the start, even when the pattern is      to have an empty string match not at the start, even when the pattern is
1329      anchored. Updated pcretest and pcredemo to use this option.      anchored. Updated pcretest and pcredemo to use this option.
1330    
1331  21. If the maximum number of capturing subpatterns in a recursion was greater  21. If the maximum number of capturing subpatterns in a recursion was greater
1332      than the maximum at the outer level, the higher number was returned, but      than the maximum at the outer level, the higher number was returned, but
1333      with unset values at the outer level. The correct (outer level) value is      with unset values at the outer level. The correct (outer level) value is
1334      now given.      now given.
1335    
1336  22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of  22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1337      PCRE did not set those parentheses (unlike Perl). I have now found a way to      PCRE did not set those parentheses (unlike Perl). I have now found a way to
1338      make it do so. The string so far is captured, making this feature      make it do so. The string so far is captured, making this feature
1339      compatible with Perl.      compatible with Perl.
1340    
1341    23. The tests have been re-organized, adding tests 11 and 12, to make it
1342        possible to check the Perl 5.10 features against Perl 5.10.
1343    
1344    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1345        pattern matches a fixed length string. PCRE did not allow this; now it
1346        does. Neither allows recursion.
1347    
1348    25. I finally figured out how to implement a request to provide the minimum
1349        length of subject string that was needed in order to match a given pattern.
1350        (It was back references and recursion that I had previously got hung up
1351        on.) This code has now been added to pcre_study(); it finds a lower bound
1352        to the length of subject needed. It is not necessarily the greatest lower
1353        bound, but using it to avoid searching strings that are too short does give
1354        some useful speed-ups. The value is available to calling programs via
1355        pcre_fullinfo().
1356    
1357    26. While implementing 25, I discovered to my embarrassment that pcretest had
1358        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1359        study optimizations had never been tested with that matching function.
1360        Oops. What is worse, even when it was passed study data, there was a bug in
1361        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1362        were also very few tests of studied patterns with pcre_dfa_exec().
1363    
1364    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1365        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1366        on the other side of the coin, they are no longer allowed to have different
1367        names, because these cannot be distinguished in PCRE, and this has caused
1368        confusion. (This is a difference from Perl.)
1369    
1370    28. When duplicate subpattern names are present (necessarily with different
1371        numbers, as required by 27 above), and a test is made by name in a
1372        conditional pattern, either for a subpattern having been matched, or for
1373        recursion in such a pattern, all the associated numbered subpatterns are
1374        tested, and the overall condition is true if the condition is true for any
1375        one of them. This is the way Perl works, and is also more like the way
1376        testing by number works.
1377    
1378    
1379  Version 7.9 11-Apr-09  Version 7.9 11-Apr-09
1380  ---------------------  ---------------------

Legend:
Removed from v.447  
changed lines
  Added in v.1025

  ViewVC Help
Powered by ViewVC 1.1.5