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

Diff of /code/trunk/ChangeLog

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

revision 361 by ph10, Thu Jul 10 16:03:28 2008 UTC revision 1389 by ph10, Tue Nov 5 18:05:29 2013 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 8.0 02 Jul-08  Version 8.34 xx-xxxx-201x
5    --------------------------
6    
7    1.  Add pcre[16|32]_jit_free_unused_memory to forcibly free unused JIT
8        executable memory. Patch inspired by Carsten Klein.
9    
10    2.  ./configure --enable-coverage defined SUPPORT_GCOV in config.h, although
11        this macro is never tested and has no effect, because the work to support
12        coverage involves only compiling and linking options and special targets in
13        the Makefile. The comment in config.h implied that defining the macro would
14        enable coverage support, which is totally false. There was also support for
15        setting this macro in the CMake files (my fault, I just copied it from
16        configure). SUPPORT_GCOV has now been removed.
17    
18    3.  Make a small performance improvement in strlen16() and strlen32() in
19        pcretest.
20    
21    4.  Change 36 for 8.33 left some unreachable statements in pcre_exec.c,
22        detected by the Solaris compiler (gcc doesn't seem to be able to diagnose
23        these cases). There was also one in pcretest.c.
24    
25    5.  Cleaned up a "may be uninitialized" compiler warning in pcre_exec.c.
26    
27    6.  In UTF mode, the code for checking whether a group could match an empty
28        string (which is used for indefinitely repeated groups to allow for
29        breaking an infinite loop) was broken when the group contained a repeated
30        negated single-character class with a character that occupied more than one
31        data item and had a minimum repetition of zero (for example, [^\x{100}]* in
32        UTF-8 mode). The effect was undefined: the group might or might not be
33        deemed as matching an empty string, or the program might have crashed.
34    
35    7.  The code for checking whether a group could match an empty string was not
36        recognizing that \h, \H, \v, \V, and \R must match a character.
37    
38    8.  Implemented PCRE_INFO_MATCH_EMPTY, which yields 1 if the pattern can match
39        an empty string. If it can, pcretest shows this in its information output.
40    
41    9.  Fixed two related bugs that applied to Unicode extended grapheme clusters
42        that were repeated with a maximizing qualifier (e.g. \X* or \X{2,5}) when
43        matched by pcre_exec() without using JIT:
44    
45        (a) If the rest of the pattern did not match after a maximal run of
46            grapheme clusters, the code for backing up to try with fewer of them
47            did not always back up over a full grapheme when characters that do not
48            have the modifier quality were involved, e.g. Hangul syllables.
49    
50        (b) If the match point in a subject started with modifier character, and
51            there was no match, the code could incorrectly back up beyond the match
52            point, and potentially beyond the first character in the subject,
53            leading to a segfault or an incorrect match result.
54    
55    10. A conditional group with an assertion condition could lead to PCRE
56        recording an incorrect first data item for a match if no other first data
57        item was recorded. For example, the pattern (?(?=ab)ab) recorded "a" as a
58        first data item, and therefore matched "ca" after "c" instead of at the
59        start.
60    
61    11. Change 40 for 8.33 (allowing pcregrep to find empty strings) showed up a
62        bug that caused the command "echo a | ./pcregrep -M '|a'" to loop.
63    
64    12. The source of pcregrep now includes z/OS-specific code so that it can be
65        compiled for z/OS as part of the special z/OS distribution.
66    
67    13. Added the -T and -TM options to pcretest.
68    
69    14. The code in pcre_compile.c for creating the table of named capturing groups
70        has been refactored. Instead of creating the table dynamically during the
71        actual compiling pass, the information is remembered during the pre-compile
72        pass (on the stack unless there are more than 20 named groups, in which
73        case malloc() is used) and the whole table is created before the actual
74        compile happens. This has simplified the code (it is now nearly 150 lines
75        shorter) and prepared the way for better handling of references to groups
76        with duplicate names.
77    
78    15. A back reference to a named subpattern when there is more than one of the
79        same name now checks them in the order in which they appear in the pattern.
80        The first one that is set is used for the reference. Previously only the
81        first one was inspected. This change makes PCRE more compatible with Perl.
82    
83    16. Unicode character properties were updated from Unicode 6.3.0.
84    
85    17. The compile-time code for auto-possessification has been refactored, based
86        on a patch by Zoltan Herczeg. It now happens after instead of during
87        compilation. The code is cleaner, and more cases are handled. The option
88        PCRE_NO_AUTO_POSSESSIFY is added for testing purposes, and the -O and /O
89        options in pcretest are provided to set it.
90    
91    18. The character VT has been added to the set of characters that match \s and
92        are generally treated as white space, following this same change in Perl
93        5.18. There is now no difference between "Perl space" and "POSIX space".
94    
95    19. The code for checking named groups as conditions, either for being set or
96        for being recursed, has been refactored (this is related to 14 and 15
97        above). Processing unduplicated named groups should now be as fast at
98        numerical groups, and processing duplicated groups should be faster than
99        before.
100    
101    20. Two patches to the CMake build system, by Alexander Barkov:
102    
103          (1) Replace the "source" command by "." in CMakeLists.txt because
104              "source" is a bash-ism.
105    
106          (2) Add missing HAVE_STDINT_H and HAVE_INTTYPES_H to config-cmake.h.in;
107              without these the CMake build does not work on Solaris.
108    
109    21. Perl has changed its handling of \8 and \9. If there is no previously
110        encountered capturing group of those numbers, they are treated as the
111        literal characters 8 and 9 instead of a binary zero followed by the
112        literals. PCRE now does the same.
113    
114    22. Following Perl, added \o{} to specify codepoints in octal, making it
115        possible to specify values greater than 0777 and also making them
116        unambiguous.
117    
118    23. Perl now gives an error for missing closing braces after \x{... instead of
119        treating the string as literal. PCRE now does the same.
120    
121    24. RunTest used to grumble if an inappropriate test was selected explicitly,
122        but just skip it when running all tests. This make it awkward to run ranges
123        of tests when one of them was inappropriate. Now it just skips any
124        inappropriate tests, as it always did when running all tests.
125    
126    25. If PCRE_AUTO_CALLOUT and PCRE_UCP were set for a pattern that contained
127        character types such as \d or \w, too many callouts were inserted, and the
128        data that they returned was rubbish.
129    
130    26. In UCP mode, \s was not matching two of the characters that Perl matches,
131        namely NEL (U+0085) and MONGOLIAN VOWEL SEPARATOR (U+180E), though they
132        were matched by \h. The code has now been refactored so that the lists of
133        the horizontal and vertical whitespace characters used for \h and \v (which
134        are defined only in one place) are now also used for \s.
135    
136    27. Add JIT support for the 64 bit TileGX architecture.
137        Patch by Jiong Wang (Tilera Corporation).
138    
139    28. Possessive quantifiers for classes (both explicit and automatically
140        generated) now use special opcodes instead of wrapping in ONCE brackets.
141    
142    29. Whereas an item such as A{4}+ ignored the possessivenes of the quantifier
143        (because it's meaningless), this was not happening when PCRE_CASELESS was
144        set. Not wrong, but inefficient.
145    
146    30. Updated perltest.pl to add /u (force Unicode mode) when /W (use Unicode
147        properties for \w, \d, etc) is present in a test regex. Otherwise if the
148        test contains no characters greater than 255, Perl doesn't realise it
149        should be using Unicode semantics.
150    
151    31. Upgraded the handling of the POSIX classes [:graph:], [:print:], and
152        [:punct:] when PCRE_UCP is set so as to include the same characters as Perl
153        does in Unicode mode.
154    
155    32. Added the "forbid" facility to pcretest so that putting tests into the
156        wrong test files can sometimes be quickly detected.
157    
158    33. There is now a limit (default 250) on the depth of nesting of parentheses.
159        This limit is imposed to control the amount of system stack used at compile
160        time. It can be changed at build time by --with-parens-nest-limit=xxx or
161        the equivalent in CMake.
162    
163    
164    Version 8.33 28-May-2013
165    ------------------------
166    
167    1.  Added 'U' to some constants that are compared to unsigned integers, to
168        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
169        variables that are added to signed variables, to ensure the result is
170        signed and can be negated.
171    
172    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
173        CMake config files.
174    
175    3.  Revise the creation of config.h.generic so that all boolean macros are
176        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
177        overriding via -D on the command line possible.
178    
179    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
180        to unsigned int is reported to make a quite noticeable speed difference in
181        a specific Windows environment. Testing on Linux did also appear to show
182        some benefit (and it is clearly not harmful). Also fixed the definition of
183        Xop which should be unsigned.
184    
185    5.  Related to (4), changing the definition of the intermediate variable cc
186        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
187        improvements.
188    
189    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
190        unnecessary spaces.
191    
192    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
193        and later.
194    
195    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
196    
197    9.  Optimizing fast_forward_start_bits in JIT.
198    
199    10. Adding support for callouts in JIT, and fixing some issues revealed
200        during this work. Namely:
201    
202        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
203    
204        (b) Minimum length was not checked before the matching is started.
205    
206    11. The value of capture_last that is passed to callouts was incorrect in some
207        cases when there was a capture on one path that was subsequently abandoned
208        after a backtrack. Also, the capture_last value is now reset after a
209        recursion, since all captures are also reset in this case.
210    
211    12. The interpreter no longer returns the "too many substrings" error in the
212        case when an overflowing capture is in a branch that is subsequently
213        abandoned after a backtrack.
214    
215    13. In the pathological case when an offset vector of size 2 is used, pcretest
216        now prints out the matched string after a yield of 0 or 1.
217    
218    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
219        Only supported by the JIT compiler at the moment.
220    
221    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
222    
223    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
224        offset of the starting point of the matching process, provided the offsets
225        vector is large enough.
226    
227    17. The \A escape now records a lookbehind value of 1, though its execution
228        does not actually inspect the previous character. This is to ensure that,
229        in partial multi-segment matching, at least one character from the old
230        segment is retained when a new segment is processed. Otherwise, if there
231        are no lookbehinds in the pattern, \A might match incorrectly at the start
232        of a new segment.
233    
234    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
235    
236    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
237        gives some modest performance improvement in 8-bit mode.
238    
239    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
240        be expressed in certain programming languages using Universal Character
241        Names.
242    
243    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
244        which points out that "non characters" are not "characters that may not
245        appear in Unicode strings" but rather "characters that are reserved for
246        internal use and have only local meaning".
247    
248    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
249        there was a conditional group that depended on an assertion, if the
250        assertion was false, the callout that immediately followed the alternation
251        in the condition was skipped when pcre_exec() was used for matching.
252    
253    23. Allow an explicit callout to be inserted before an assertion that is the
254        condition for a conditional group, for compatibility with automatic
255        callouts, which always insert a callout at this point.
256    
257    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
258        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
259    
260    25. (*PRUNE) is now supported by the JIT compiler.
261    
262    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
263    
264    27. Fix the case where there are two or more SKIPs with arguments that may be
265        ignored.
266    
267    28. (*SKIP) is now supported by the JIT compiler.
268    
269    29. (*THEN) is now supported by the JIT compiler.
270    
271    30. Update RunTest with additional test selector options.
272    
273    31. The way PCRE handles backtracking verbs has been changed in two ways.
274    
275        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
276        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
277        backtracking. In some cases this makes it more Perl-compatible, but Perl's
278        rather obscure rules do not always do the same thing.
279    
280        (2) Previously, backtracking verbs were confined within assertions. This is
281        no longer the case for positive assertions, except for (*ACCEPT). Again,
282        this sometimes improves Perl compatibility, and sometimes does not.
283    
284    32. A number of tests that were in test 2 because Perl did things differently
285        have been moved to test 1, because either Perl or PCRE has changed, and
286        these tests are now compatible.
287    
288    32. Backtracking control verbs are now handled in the same way in JIT and
289        interpreter.
290    
291    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
292        contained a forward subroutine reference caused a compile error.
293    
294    34. Auto-detect and optimize limited repetitions in JIT.
295    
296    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
297        blocking (*UTF) etc.
298    
299    36. In the interpreter, maximizing pattern repetitions for characters and
300        character types now use tail recursion, which reduces stack usage.
301    
302    37. The value of the max lookbehind was not correctly preserved if a compiled
303        and saved regex was reloaded on a host of different endianness.
304    
305    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
306        of the compiled pattern block, expand the flags field from 16 to 32 bits
307        because it was almost full.
308    
309    39. Try madvise first before posix_madvise.
310    
311    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
312        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
313        original change locked out all matches of empty strings. This has been
314        changed so that one match of an empty string per line is recognized.
315        Subsequent searches on the same line (for colouring or for --only-matching,
316        for example) do not recognize empty strings.
317    
318    41. Applied a user patch to fix a number of spelling mistakes in comments.
319    
320    42. Data lines longer than 65536 caused pcretest to crash.
321    
322    43. Clarified the data type for length and startoffset arguments for pcre_exec
323        and pcre_dfa_exec in the function-specific man pages, where they were
324        explicitly stated to be in bytes, never having been updated. I also added
325        some clarification to the pcreapi man page.
326    
327    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
328        a segmentation fault.
329    
330    
331    Version 8.32 30-November-2012
332    -----------------------------
333    
334    1.  Improved JIT compiler optimizations for first character search and single
335        character iterators.
336    
337    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
338        Patch by Daniel Richard G.
339    
340    3.  Single character iterator optimizations in the JIT compiler.
341    
342    4.  Improved JIT compiler optimizations for character ranges.
343    
344    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
345        Reported by Giuseppe D'Angelo.
346    
347    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
348        of a line, was being set incorrectly in cases where .* appeared inside
349        atomic brackets at the start of a pattern, or where there was a subsequent
350        *PRUNE or *SKIP.
351    
352    7.  Improved instruction cache flush for POWER/PowerPC.
353        Patch by Daniel Richard G.
354    
355    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
356        grep:
357    
358        (a) There is now no limit to the number of patterns to be matched.
359    
360        (b) An error is given if a pattern is too long.
361    
362        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
363            are now supported.
364    
365        (d) --exclude-from and --include-from (multiple use) have been added.
366    
367        (e) Exclusions and inclusions now apply to all files and directories, not
368            just to those obtained from scanning a directory recursively.
369    
370        (f) Multiple uses of -f and --file-list are now supported.
371    
372        (g) In a Windows environment, the default for -d has been changed from
373            "read" (the GNU grep default) to "skip", because otherwise the presence
374            of a directory in the file list provokes an error.
375    
376        (h) The documentation has been revised and clarified in places.
377    
378    9.  Improve the matching speed of capturing brackets.
379    
380    10. Changed the meaning of \X so that it now matches a Unicode extended
381        grapheme cluster.
382    
383    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
384        out POSIX threads when JIT support is configured.
385    
386    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
387    
388    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
389        structure could go wrong in environments where size_t is not the same size
390        as int.
391    
392    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
393        set.
394    
395    15. The EBCDIC support had decayed; later updates to the code had included
396        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
397        general tidy up of EBCDIC-related issues, and the documentation was also
398        not quite right. There is now a test that can be run on ASCII systems to
399        check some of the EBCDIC-related things (but is it not a full test).
400    
401    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
402        in a small tidy to the code.
403    
404    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
405    
406    18. If the --only-matching (-o) option in pcregrep is specified multiple
407        times, each one causes appropriate output. For example, -o1 -o2 outputs the
408        substrings matched by the 1st and 2nd capturing parentheses. A separating
409        string can be specified by --om-separator (default empty).
410    
411    19. Improving the first n character searches.
412    
413    20. Turn case lists for horizontal and vertical white space into macros so that
414        they are defined only once.
415    
416    21. This set of changes together give more compatible Unicode case-folding
417        behaviour for characters that have more than one other case when UCP
418        support is available.
419    
420        (a) The Unicode property table now has offsets into a new table of sets of
421            three or more characters that are case-equivalent. The MultiStage2.py
422            script that generates these tables (the pcre_ucd.c file) now scans
423            CaseFolding.txt instead of UnicodeData.txt for character case
424            information.
425    
426        (b) The code for adding characters or ranges of characters to a character
427            class has been abstracted into a generalized function that also handles
428            case-independence. In UTF-mode with UCP support, this uses the new data
429            to handle characters with more than one other case.
430    
431        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
432            whose other case is greater than 256 are now correctly matched
433            caselessly. Previously, the high codepoint matched the low one, but not
434            vice versa.
435    
436        (d) The processing of \h, \H, \v, and \ in character classes now makes use
437            of the new class addition function, using character lists defined as
438            macros alongside the case definitions of 20 above.
439    
440        (e) Caseless back references now work with characters that have more than
441            one other case.
442    
443        (f) General caseless matching of characters with more than one other case
444            is supported.
445    
446    22. Unicode character properties were updated from Unicode 6.2.0
447    
448    23. Improved CMake support under Windows. Patch by Daniel Richard G.
449    
450    24. Add support for 32-bit character strings, and UTF-32
451    
452    25. Major JIT compiler update (code refactoring and bugfixing).
453        Experimental Sparc 32 support is added.
454    
455    26. Applied a modified version of Daniel Richard G's patch to create
456        pcre.h.generic and config.h.generic by "make" instead of in the
457        PrepareRelease script.
458    
459    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
460        pcre_compile.c when checking for a zero character.
461    
462    28. Introducing a native interface for JIT. Through this interface, the compiled
463        machine code can be directly executed. The purpose of this interface is to
464        provide fast pattern matching, so several sanity checks are not performed.
465        However, feature tests are still performed. The new interface provides
466        1.4x speedup compared to the old one.
467    
468    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
469        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
470        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
471    
472    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
473        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
474        the "old" RFC 2279). Instead, it ended up passing a negative length to
475        pcre_exec().
476    
477    31. Add support for GCC's visibility feature to hide internal functions.
478    
479    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
480        "unknown -C option" after outputting 0 or 1.
481    
482    33. There is now support for generating a code coverage report for the test
483        suite in environments where gcc is the compiler and lcov is installed. This
484        is mainly for the benefit of the developers.
485    
486    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
487        unaddressable using valgrind annotations, allowing valgrind to detect
488        invalid memory accesses. This is mainly for the benefit of the developers.
489    
490    25. (*UTF) can now be used to start a pattern in any of the three libraries.
491    
492    26. Give configure error if --enable-cpp but no C++ compiler found.
493    
494    
495    Version 8.31 06-July-2012
496    -------------------------
497    
498    1.  Fixing a wrong JIT test case and some compiler warnings.
499    
500    2.  Removed a bashism from the RunTest script.
501    
502    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
503        to unsigned type, result still unsigned" that was given by an MS compiler
504        on encountering the code "-sizeof(xxx)".
505    
506    4.  Partial matching support is added to the JIT compiler.
507    
508    5.  Fixed several bugs concerned with partial matching of items that consist
509        of more than one character:
510    
511        (a) /^(..)\1/ did not partially match "aba" because checking references was
512            done on an "all or nothing" basis. This also applied to repeated
513            references.
514    
515        (b) \R did not give a hard partial match if \r was found at the end of the
516            subject.
517    
518        (c) \X did not give a hard partial match after matching one or more
519            characters at the end of the subject.
520    
521        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
522            a partial match for the string "\r".
523    
524        (e) When newline was set to CRLF, the metacharacter "." did not recognize
525            a partial match for a CR character at the end of the subject string.
526    
527    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
528        running pcretest, the text "(JIT)" added to the output whenever JIT is
529        actually used to run the match.
530    
531    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
532        or /S+[+] with a digit between 1 and 7.
533    
534    8.  OP_NOT now supports any UTF character not just single-byte ones.
535    
536    9.  (*MARK) control verb is now supported by the JIT compiler.
537    
538    10. The command "./RunTest list" lists the available tests without actually
539        running any of them. (Because I keep forgetting what they all are.)
540    
541    11. Add PCRE_INFO_MAXLOOKBEHIND.
542    
543    12. Applied a (slightly modified) user-supplied patch that improves performance
544        when the heap is used for recursion (compiled with --disable-stack-for-
545        recursion). Instead of malloc and free for each heap frame each time a
546        logical recursion happens, frames are retained on a chain and re-used where
547        possible. This sometimes gives as much as 30% improvement.
548    
549    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
550        call.
551    
552    14. As documented, (*COMMIT) is now confined to within a positive assertion.
553    
554    15. It is now possible to link pcretest with libedit as an alternative to
555        libreadline.
556    
557    16. (*COMMIT) control verb is now supported by the JIT compiler.
558    
559    17. The Unicode data tables have been updated to Unicode 6.1.0.
560    
561    18. Added --file-list option to pcregrep.
562    
563    19. Added binary file support to pcregrep, including the -a, --binary-files,
564        -I, and --text options.
565    
566    20. The madvise function is renamed for posix_madvise for QNX compatibility
567        reasons. Fixed by Giuseppe D'Angelo.
568    
569    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
570    
571    22. Changed the option for creating symbolic links for 16-bit man pages from
572        -s to -sf so that re-installing does not cause issues.
573    
574    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
575    
576    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
577        matches in certain environments (the workspace was not being correctly
578        retained). Also added to pcre_dfa_exec() a simple plausibility check on
579        some of the workspace data at the beginning of a restart.
580    
581    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
582        was not doing so when it should - probably a typo introduced by SVN 528
583        (change 8.10/14).
584    
585    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
586        \w+ when the character tables indicated that \x{c4} was a word character.
587        There were several related cases, all because the tests for doing a table
588        lookup were testing for characters less than 127 instead of 255.
589    
590    27. If a pattern contains capturing parentheses that are not used in a match,
591        their slots in the ovector are set to -1. For those that are higher than
592        any matched groups, this happens at the end of processing. In the case when
593        there were back references that the ovector was too small to contain
594        (causing temporary malloc'd memory to be used during matching), and the
595        highest capturing number was not used, memory off the end of the ovector
596        was incorrectly being set to -1. (It was using the size of the temporary
597        memory instead of the true size.)
598    
599    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
600        ovector size, it uses memory at the end of the block that it has got.
601    
602    29. Check for an overlong MARK name and give an error at compile time. The
603        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
604    
605    30. JIT compiler update.
606    
607    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
608        Rill for the patch.
609    
610    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
611    
612    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
613    
614    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
615        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
616        was enabled.
617    
618    35. Improve JIT code generation for greedy plus quantifier.
619    
620    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
621        1 to "aa" instead of to an empty string. The bug affected repeated groups
622        that could potentially match an empty string.
623    
624    37. Optimizing single character iterators in JIT.
625    
626    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
627        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
628        codepoints that are too big for the mode are faulted, and in a UTF mode,
629        disallowed codepoints are also faulted.
630    
631    39. If PCRE was compiled with UTF support, in three places in the DFA
632        matcher there was code that should only have been obeyed in UTF mode, but
633        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
634        processing when bytes with values greater than 127 were present. In 16-bit
635        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
636        both cases the values are those that cannot be the first data item in a UTF
637        character. The three items that might have provoked this were recursions,
638        possessively repeated groups, and atomic groups.
639    
640    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
641        and pcregrep, because some OS require shared objects to be explicitly
642        passed to ld, causing the link step to fail if they are not.
643    
644    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
645        mode, patterns that started with \h* or \R* might be incorrectly matched.
646    
647    
648    Version 8.30 04-February-2012
649    -----------------------------
650    
651    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
652        name is defined in ctype.h.
653    
654    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
655        only in quite long subpatterns.
656    
657    3.  Removed the function pcre_info(), which has been obsolete and deprecated
658        since it was replaced by pcre_fullinfo() in February 2000.
659    
660    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
661        match a (*MARK), and the match failed at the start of the subject, a
662        reference to memory before the start of the subject could occur. This bug
663        was introduced by fix 17 of release 8.21.
664    
665    5.  A reference to an unset group with zero minimum repetition was giving
666        totally wrong answers (in non-JavaScript-compatibility mode). For example,
667        /(another)?(\1?)test/ matched against "hello world test". This bug was
668        introduced in release 8.13.
669    
670    6.  Add support for 16-bit character strings (a large amount of work involving
671        many changes and refactorings).
672    
673    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
674        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
675        from a file.
676    
677    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
678        rounding is not applied in this particular case).
679    
680    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
681        if they appear, or are escaped, in patterns.
682    
683    10. Get rid of a number of -Wunused-but-set-variable warnings.
684    
685    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
686        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
687        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
688        also returns the mark "x". This bug applied to capturing parentheses,
689        non-capturing parentheses, and atomic parentheses. It also applied to some
690        assertions.
691    
692    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
693        information out of configure.ac instead of relying on pcre.h.generic, which
694        is not stored in the repository.
695    
696    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
697        -lreadline.
698    
699    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
700    
701    15. Applied Graycode's patch to put the top-level frame on the stack rather
702        than the heap when not using the stack for recursion. This gives a
703        performance improvement in many cases when recursion is not deep.
704    
705    16. Experimental code added to "pcretest -C" to output the stack frame size.
706    
707    
708    Version 8.21 12-Dec-2011
709    ------------------------
710    
711    1.  Updating the JIT compiler.
712    
713    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
714        are added as well.
715    
716    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
717        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
718        calling _pcre_jit_exec. Some extra comments are added.
719    
720    4.  (*MARK) settings inside atomic groups that do not contain any capturing
721        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
722        was introduced by change 18 for 8.20.
723    
724    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
725        ECMA-262 standard.
726    
727    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
728        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
729        This bug was probably introduced by change 9 of 8.13.
730    
731    7.  While fixing 6 above, I noticed that a number of other items were being
732        incorrectly rejected as "not fixed length". This arose partly because newer
733        opcodes had not been added to the fixed-length checking code. I have (a)
734        corrected the bug and added tests for these items, and (b) arranged for an
735        error to occur if an unknown opcode is encountered while checking for fixed
736        length instead of just assuming "not fixed length". The items that were
737        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
738        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
739        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
740    
741    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
742        being incorrectly compiled and would have given unpredicatble results.
743    
744    9.  A possessively repeated subpattern with minimum repeat count greater than
745        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
746        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
747        the first (A) could occur when it should not.
748    
749    10. Add a cast and remove a redundant test from the code.
750    
751    11. JIT should use pcre_malloc/pcre_free for allocation.
752    
753    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
754        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
755        is anything other than /usr, -L is still shown).
756    
757    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
758    
759    14. Perl does not support \N without a following name in a [] class; PCRE now
760        also gives an error.
761    
762    15. If a forward reference was repeated with an upper limit of around 2000,
763        it caused the error "internal error: overran compiling workspace". The
764        maximum number of forward references (including repeats) was limited by the
765        internal workspace, and dependent on the LINK_SIZE. The code has been
766        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
767        the default depends on LINK_SIZE. There is a new upper limit (for safety)
768        of around 200,000 forward references. While doing this, I also speeded up
769        the filling in of repeated forward references.
770    
771    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
772        incorrectly expecting the subject to contain another "a" after the start.
773    
774    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
775        in the match, the SKIP should be ignored. This was not happening; instead
776        the SKIP was being treated as NOMATCH. For patterns such as
777        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
778        tested.
779    
780    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
781        now much more compatible with Perl, in particular in cases where the result
782        is a non-match for a non-anchored pattern. For example, if
783        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
784        "m", where previously it did not return a name. A side effect of this
785        change is that for partial matches, the last encountered mark name is
786        returned, as for non matches. A number of tests that were previously not
787        Perl-compatible have been moved into the Perl-compatible test files. The
788        refactoring has had the pleasing side effect of removing one argument from
789        the match() function, thus reducing its stack requirements.
790    
791    19. If the /S+ option was used in pcretest to study a pattern using JIT,
792        subsequent uses of /S (without +) incorrectly behaved like /S+.
793    
794    21. Retrieve executable code size support for the JIT compiler and fixing
795        some warnings.
796    
797    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
798        not work when the shorter character appeared right at the end of the
799        subject string.
800    
801    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
802        systems.
803    
804    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
805        output it when the /M option is used in pcretest.
806    
807    25. The CheckMan script was not being included in the distribution. Also, added
808        an explicit "perl" to run Perl scripts from the PrepareRelease script
809        because this is reportedly needed in Windows.
810    
811    26. If study data was being save in a file and studying had not found a set of
812        "starts with" bytes for the pattern, the data written to the file (though
813        never used) was taken from uninitialized memory and so caused valgrind to
814        complain.
815    
816    27. Updated RunTest.bat as provided by Sheri Pierce.
817    
818    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
819    
820    29. Computation of memory usage for the table of capturing group names was
821        giving an unnecessarily large value.
822    
823    
824    Version 8.20 21-Oct-2011
825    ------------------------
826    
827    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
828        a POSIX class. After further experiments with Perl, which convinced me that
829        Perl has bugs and confusions, a closing square bracket is no longer allowed
830        in a POSIX name. This bug also affected patterns with classes that started
831        with full stops.
832    
833    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
834        captured substring, but while checking the failing first alternative,
835        substring 1 is temporarily captured. If the output vector supplied to
836        pcre_exec() was not big enough for this capture, the yield of the function
837        was still zero ("insufficient space for captured substrings"). This cannot
838        be totally fixed without adding another stack variable, which seems a lot
839        of expense for a edge case. However, I have improved the situation in cases
840        such as /(a)(b)x|abc/ matched against "abc", where the return code
841        indicates that fewer than the maximum number of slots in the ovector have
842        been set.
843    
844    3.  Related to (2) above: when there are more back references in a pattern than
845        slots in the output vector, pcre_exec() uses temporary memory during
846        matching, and copies in the captures as far as possible afterwards. It was
847        using the entire output vector, but this conflicts with the specification
848        that only 2/3 is used for passing back captured substrings. Now it uses
849        only the first 2/3, for compatibility. This is, of course, another edge
850        case.
851    
852    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
853        main code base, and can be used by building with --enable-jit. When this is
854        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
855        runtime --no-jit option is given.
856    
857    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
858        ovector, the return from the function was zero, implying that there were
859        other matches that did not fit. The correct "exactly full" value is now
860        returned.
861    
862    6.  If a subpattern that was called recursively or as a subroutine contained
863        (*PRUNE) or any other control that caused it to give a non-standard return,
864        invalid errors such as "Error -26 (nested recursion at the same subject
865        position)" or even infinite loops could occur.
866    
867    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
868        computing the minimum length on reaching *ACCEPT, and so ended up with the
869        wrong value of 1 rather than 0. Further investigation indicates that
870        computing a minimum subject length in the presence of *ACCEPT is difficult
871        (think back references, subroutine calls), and so I have changed the code
872        so that no minimum is registered for a pattern that contains *ACCEPT.
873    
874    8.  If (*THEN) was present in the first (true) branch of a conditional group,
875        it was not handled as intended. [But see 16 below.]
876    
877    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
878        Sheri Pierce.
879    
880    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
881        the first byte in a match must be "a".
882    
883    11. Change 17 for 8.13 increased the recursion depth for patterns like
884        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
885        pattern contains any instances of (*THEN). If it does not, the old
886        optimizations are restored. It would be nice to do this on a per-group
887        basis, but at the moment that is not feasible.
888    
889    12. In some environments, the output of pcretest -C is CRLF terminated. This
890        broke RunTest's code that checks for the link size. A single white space
891        character after the value is now allowed for.
892    
893    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
894        For "fr", it uses the Windows-specific input and output files.
895    
896    14. If (*THEN) appeared in a group that was called recursively or as a
897        subroutine, it did not work as intended. [But see next item.]
898    
899    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
900        pattern fragments (but not containing any | characters). If A and B are
901        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
902        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
903        D. In other words, Perl considers parentheses that do not contain any |
904        characters to be part of a surrounding alternative, whereas PCRE was
905        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
906        differently. PCRE now behaves in the same way as Perl, except in the case
907        of subroutine/recursion calls such as (?1) which have in any case always
908        been different (but PCRE had them first :-).
909    
910    16. Related to 15 above: Perl does not treat the | in a conditional group as
911        creating alternatives. Such a group is treated in the same way as an
912        ordinary group without any | characters when processing (*THEN). PCRE has
913        been changed to match Perl's behaviour.
914    
915    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
916        RunGrepTest script failed.
917    
918    18. Change 22 for version 13 caused atomic groups to use more stack. This is
919        inevitable for groups that contain captures, but it can lead to a lot of
920        stack use in large patterns. The old behaviour has been restored for atomic
921        groups that do not contain any capturing parentheses.
922    
923    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
924        suppress the check for a minimum subject length at run time. (If it was
925        given to pcre_exec() or pcre_dfa_exec() it did work.)
926    
927    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
928        fail to work when decoding hex characters in data strings in EBCDIC
929        environments.
930    
931    21. It appears that in at least one Mac OS environment, the isxdigit() function
932        is implemented as a macro that evaluates to its argument more than once,
933        contravening the C 90 Standard (I haven't checked a later standard). There
934        was an instance in pcretest which caused it to go wrong when processing
935        \x{...} escapes in subject strings. The has been rewritten to avoid using
936        things like p++ in the argument of isxdigit().
937    
938    
939    Version 8.13 16-Aug-2011
940    ------------------------
941    
942    1.  The Unicode data tables have been updated to Unicode 6.0.0.
943    
944    2.  Two minor typos in pcre_internal.h have been fixed.
945    
946    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
947        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
948        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
949    
950    4.  There were a number of related bugs in the code for matching backrefences
951        caselessly in UTF-8 mode when codes for the characters concerned were
952        different numbers of bytes. For example, U+023A and U+2C65 are an upper
953        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
954        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
955        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
956        2-byte code at the end of the subject (it thought there wasn't enough data
957        left).
958    
959    5.  Comprehensive information about what went wrong is now returned by
960        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
961        as the output vector has at least 2 elements. The offset of the start of
962        the failing character and a reason code are placed in the vector.
963    
964    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
965        now returned is for the first byte of the failing character, instead of the
966        last byte inspected. This is an incompatible change, but I hope it is small
967        enough not to be a problem. It makes the returned offset consistent with
968        pcre_exec() and pcre_dfa_exec().
969    
970    7.  pcretest now gives a text phrase as well as the error number when
971        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
972        failure, the offset and reason code are output.
973    
974    8.  When \R was used with a maximizing quantifier it failed to skip backwards
975        over a \r\n pair if the subsequent match failed. Instead, it just skipped
976        back over a single character (\n). This seems wrong (because it treated the
977        two characters as a single entity when going forwards), conflicts with the
978        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
979        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
980        has been changed.
981    
982    9.  Some internal refactoring has changed the processing so that the handling
983        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
984        time (the PCRE_DOTALL option was changed this way some time ago: version
985        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
986        which was always a bit of a fudge. It also means that there is one less
987        argument for the match() function, which reduces its stack requirements
988        slightly. This change also fixes an incompatibility with Perl: the pattern
989        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
990    
991    10. More internal refactoring has drastically reduced the number of recursive
992        calls to match() for possessively repeated groups such as (abc)++ when
993        using pcre_exec().
994    
995    11. While implementing 10, a number of bugs in the handling of groups were
996        discovered and fixed:
997    
998        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
999        (a|)*(?1) gave a compile-time internal error.
1000        ((a|)+)+  did not notice that the outer group could match an empty string.
1001        (^a|^)+   was not marked as anchored.
1002        (.*a|.*)+ was not marked as matching at start or after a newline.
1003    
1004    12. Yet more internal refactoring has removed another argument from the match()
1005        function. Special calls to this function are now indicated by setting a
1006        value in a variable in the "match data" data block.
1007    
1008    13. Be more explicit in pcre_study() instead of relying on "default" for
1009        opcodes that mean there is no starting character; this means that when new
1010        ones are added and accidentally left out of pcre_study(), testing should
1011        pick them up.
1012    
1013    14. The -s option of pcretest has been documented for ages as being an old
1014        synonym of -m (show memory usage). I have changed it to mean "force study
1015        for every regex", that is, assume /S for every regex. This is similar to -i
1016        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
1017        using it. It makes it easier to run collections of tests with and without
1018        study enabled, and thereby test pcre_study() more easily. All the standard
1019        tests are now run with and without -s (but some patterns can be marked as
1020        "never study" - see 20 below).
1021    
1022    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
1023        restoration of the capturing data to the outer values was not happening
1024        correctly.
1025    
1026    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
1027        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
1028        pattern had matched an empty string, and so incorrectly returned a no
1029        match.
1030    
1031    17. There was optimizing code for the last branch of non-capturing parentheses,
1032        and also for the obeyed branch of a conditional subexpression, which used
1033        tail recursion to cut down on stack usage. Unfortunately, now that there is
1034        the possibility of (*THEN) occurring in these branches, tail recursion is
1035        no longer possible because the return has to be checked for (*THEN). These
1036        two optimizations have therefore been removed. [But see 8.20/11 above.]
1037    
1038    18. If a pattern containing \R was studied, it was assumed that \R always
1039        matched two bytes, thus causing the minimum subject length to be
1040        incorrectly computed because \R can also match just one byte.
1041    
1042    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
1043        was incorrectly computed.
1044    
1045    20. If /S is present twice on a test pattern in pcretest input, it now
1046        *disables* studying, thereby overriding the use of -s on the command line
1047        (see 14 above). This is necessary for one or two tests to keep the output
1048        identical in both cases.
1049    
1050    21. When (*ACCEPT) was used in an assertion that matched an empty string and
1051        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
1052    
1053    22. When an atomic group that contained a capturing parenthesis was
1054        successfully matched, but the branch in which it appeared failed, the
1055        capturing was not being forgotten if a higher numbered group was later
1056        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
1057        group 1 to "a", when in fact it should be unset. This applied to multi-
1058        branched capturing and non-capturing groups, repeated or not, and also to
1059        positive assertions (capturing in negative assertions does not happen
1060        in PCRE) and also to nested atomic groups.
1061    
1062    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
1063        subject after a captured substring, to make it easier to tell which of a
1064        number of identical substrings has been captured.
1065    
1066    24. The way atomic groups are processed by pcre_exec() has been changed so that
1067        if they are repeated, backtracking one repetition now resets captured
1068        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
1069        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
1070        "aaa". Previously, it would have been "a". As part of this code
1071        refactoring, the way recursive calls are handled has also been changed.
1072    
1073    25. If an assertion condition captured any substrings, they were not passed
1074        back unless some other capturing happened later. For example, if
1075        (?(?=(a))a) was matched against "a", no capturing was returned.
1076    
1077    26. When studying a pattern that contained subroutine calls or assertions,
1078        the code for finding the minimum length of a possible match was handling
1079        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
1080        group 1 called group 2 while simultaneously a separate group 2 called group
1081        1). A stack overflow occurred in this case. I have fixed this by limiting
1082        the recursion depth to 10.
1083    
1084    27. Updated RunTest.bat in the distribution to the version supplied by Tom
1085        Fortmann. This supports explicit test numbers on the command line, and has
1086        argument validation and error reporting.
1087    
1088    28. An instance of \X with an unlimited repeat could fail if at any point the
1089        first character it looked at was a mark character.
1090    
1091    29. Some minor code refactoring concerning Unicode properties and scripts
1092        should reduce the stack requirement of match() slightly.
1093    
1094    30. Added the '=' option to pcretest to check the setting of unused capturing
1095        slots at the end of the pattern, which are documented as being -1, but are
1096        not included in the return count.
1097    
1098    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
1099        compiled something random. Now it gives a compile-time error (as does
1100        Perl).
1101    
1102    32. A *MARK encountered during the processing of a positive assertion is now
1103        recorded and passed back (compatible with Perl).
1104    
1105    33. If --only-matching or --colour was set on a pcregrep call whose pattern
1106        had alternative anchored branches, the search for a second match in a line
1107        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
1108        matched the line "0102" twice. The same bug affected patterns that started
1109        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
1110        twice.
1111    
1112    34. Previously, PCRE did not allow quantification of assertions. However, Perl
1113        does, and because of capturing effects, quantifying parenthesized
1114        assertions may at times be useful. Quantifiers are now allowed for
1115        parenthesized assertions.
1116    
1117    35. A minor code tidy in pcre_compile() when checking options for \R usage.
1118    
1119    36. \g was being checked for fancy things in a character class, when it should
1120        just be a literal "g".
1121    
1122    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
1123        appearance of a nested POSIX class supersedes an apparent external class.
1124        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
1125        unescaped square brackets may also appear as part of class names. For
1126        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
1127        more like Perl. (But see 8.20/1 above.)
1128    
1129    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
1130        was because it thought it was \N{name}, which is not supported).
1131    
1132    39. Add minix to OS list not supporting the -S option in pcretest.
1133    
1134    40. PCRE tries to detect cases of infinite recursion at compile time, but it
1135        cannot analyze patterns in sufficient detail to catch mutual recursions
1136        such as ((?1))((?2)). There is now a runtime test that gives an error if a
1137        subgroup is called recursively as a subpattern for a second time at the
1138        same position in the subject string. In previous releases this might have
1139        been caught by the recursion limit, or it might have run out of stack.
1140    
1141    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
1142        happen only once. PCRE was, however incorrectly giving a compile time error
1143        "recursive call could loop indefinitely" because it cannot analyze the
1144        pattern in sufficient detail. The compile time test no longer happens when
1145        PCRE is compiling a conditional subpattern, but actual runaway loops are
1146        now caught at runtime (see 40 above).
1147    
1148    42. It seems that Perl allows any characters other than a closing parenthesis
1149        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
1150        has been changed to be the same.
1151    
1152    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
1153        as not to get warnings when autogen.sh is called. Also changed
1154        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
1155    
1156    44. To help people who use pcregrep to scan files containing exceedingly long
1157        lines, the following changes have been made:
1158    
1159        (a) The default value of the buffer size parameter has been increased from
1160            8K to 20K. (The actual buffer used is three times this size.)
1161    
1162        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
1163            PCRE is built.
1164    
1165        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
1166            to be set at run time.
1167    
1168        (d) Numerical values in pcregrep options can be followed by K or M, for
1169            example --buffer-size=50K.
1170    
1171        (e) If a line being scanned overflows pcregrep's buffer, an error is now
1172            given and the return code is set to 2.
1173    
1174    45. Add a pointer to the latest mark to the callout data block.
1175    
1176    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
1177        partial match of an empty string instead of no match. This was specific to
1178        the use of ".".
1179    
1180    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
1181        complete match instead of a partial match. This bug was dependent on both
1182        the PCRE_UTF8 and PCRE_DOTALL options being set.
1183    
1184    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1185        starting byte set, because \b was not being ignored.
1186    
1187    
1188    Version 8.12 15-Jan-2011
1189    ------------------------
1190    
1191    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1192        checks for such things as part of the documentation building process.
1193    
1194    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1195        --match-limit and --recursion-limit options (added for 8.11). In
1196        particular, this made one of the standard tests fail. (The integer value
1197        went into the wrong half of a long int.)
1198    
1199    3.  If the --colour option was given to pcregrep with -v (invert match), it
1200        did strange things, either producing crazy output, or crashing. It should,
1201        of course, ignore a request for colour when reporting lines that do not
1202        match.
1203    
1204    4.  Another pcregrep bug caused similar problems if --colour was specified with
1205        -M (multiline) and the pattern match finished with a line ending.
1206    
1207    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1208        matched in multiline mode, the following line was shown as part of the
1209        match. This seems wrong, so I have changed it.
1210    
1211    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1212        the check for further matches in the same line (so they could be coloured)
1213        to overrun the end of the current line. If another match was found, it was
1214        incorrectly shown (and then shown again when found in the next line).
1215    
1216    7.  If pcregrep was compiled under Windows, there was a reference to the
1217        function pcregrep_exit() before it was defined. I am assuming this was
1218        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1219        reported by a user. I've moved the definition above the reference.
1220    
1221    
1222    Version 8.11 10-Dec-2010
1223    ------------------------
1224    
1225    1.  (*THEN) was not working properly if there were untried alternatives prior
1226        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1227        backtracked to try for "b" instead of moving to the next alternative branch
1228        at the same level (in this case, to look for "c"). The Perl documentation
1229        is clear that when (*THEN) is backtracked onto, it goes to the "next
1230        alternative in the innermost enclosing group".
1231    
1232    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1233        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1234        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1235        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1236        (*THEN).
1237    
1238    3.  If \s appeared in a character class, it removed the VT character from
1239        the class, even if it had been included by some previous item, for example
1240        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1241        of \s, but is part of the POSIX "space" class.)
1242    
1243    4.  A partial match never returns an empty string (because you can always
1244        match an empty string at the end of the subject); however the checking for
1245        an empty string was starting at the "start of match" point. This has been
1246        changed to the "earliest inspected character" point, because the returned
1247        data for a partial match starts at this character. This means that, for
1248        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1249        (previously it gave "no match").
1250    
1251    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1252        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1253        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1254        has an implication that the given string is incomplete (because a partial
1255        match is preferred over a full match). For this reason, these items now
1256        give a partial match in this situation. [Aside: previously, the one case
1257        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1258        match rather than a full match, which was wrong by the old rules, but is
1259        now correct.]
1260    
1261    6.  There was a bug in the handling of #-introduced comments, recognized when
1262        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1263        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1264        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1265        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1266        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1267        places in pcre_compile().
1268    
1269    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1270        comments when looking ahead for named forward references to subpatterns,
1271        the only newline sequence it recognized was NL. It now handles newlines
1272        according to the set newline convention.
1273    
1274    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1275        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1276        cater for a lack of strerror(). These oversights have been fixed.
1277    
1278    9.  Added --match-limit and --recursion-limit to pcregrep.
1279    
1280    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1281    
1282    11. When the -o option was used, pcregrep was setting a return code of 1, even
1283        when matches were found, and --line-buffered was not being honoured.
1284    
1285    12. Added an optional parentheses number to the -o and --only-matching options
1286        of pcregrep.
1287    
1288    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1289        can match an empty string. The code to do it in pcretest and pcredemo
1290        needed fixing:
1291    
1292        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1293            only one byte after an empty string match just before CRLF (this case
1294            just got forgotten; "any" and "anycrlf" were OK).
1295    
1296        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1297            mode when an empty string match preceded an ASCII character followed by
1298            a non-ASCII character. (The code for advancing by one character rather
1299            than one byte was nonsense.)
1300    
1301        (c) The pcredemo.c sample program did not have any code at all to handle
1302            the cases when CRLF is a valid newline sequence.
1303    
1304    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1305        as a starting offset was within the subject string. There is now a new
1306        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1307        negative or greater than the length of the string. In order to test this,
1308        pcretest is extended to allow the setting of negative starting offsets.
1309    
1310    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1311        starting offset points to the beginning of a UTF-8 character was
1312        unnecessarily clumsy. I tidied it up.
1313    
1314    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1315        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1316    
1317    17. Nobody had reported that the --include_dir option, which was added in
1318        release 7.7 should have been called --include-dir (hyphen, not underscore)
1319        for compatibility with GNU grep. I have changed it to --include-dir, but
1320        left --include_dir as an undocumented synonym, and the same for
1321        --exclude-dir, though that is not available in GNU grep, at least as of
1322        release 2.5.4.
1323    
1324    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1325        characters from a string of bytes) have been redefined so as not to use
1326        loops, in order to improve performance in some environments. At the same
1327        time, I abstracted some of the common code into auxiliary macros to save
1328        repetition (this should not affect the compiled code).
1329    
1330    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1331        compile-time error is now given if \c is not followed by an ASCII
1332        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1333        different, and any byte value is allowed.)
1334    
1335    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1336        START_OPTIMIZE option, which is now allowed at compile time - but just
1337        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1338        to pcregrep and other applications that have no direct access to PCRE
1339        options. The new /Y option in pcretest sets this option when calling
1340        pcre_compile().
1341    
1342    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1343        back references. Groups containing recursive back references were forced to
1344        be atomic by that change, but in the case of named groups, the amount of
1345        memory required was incorrectly computed, leading to "Failed: internal
1346        error: code overflow". This has been fixed.
1347    
1348    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1349        pcretest.c, to avoid build problems in some Borland environments.
1350    
1351    
1352    Version 8.10 25-Jun-2010
1353    ------------------------
1354    
1355    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1356        THEN.
1357    
1358    2.  (*ACCEPT) was not working when inside an atomic group.
1359    
1360    3.  Inside a character class, \B is treated as a literal by default, but
1361        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1362        causes the error). The code is unchanged, but I tidied the documentation.
1363    
1364    4.  Inside a character class, PCRE always treated \R and \X as literals,
1365        whereas Perl faults them if its -w option is set. I have changed PCRE so
1366        that it faults them when PCRE_EXTRA is set.
1367    
1368    5.  Added support for \N, which always matches any character other than
1369        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1370    
1371    6.  When compiling pcregrep with newer versions of gcc which may have
1372        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1373        declared with attribute warn_unused_result" were given. Just casting the
1374        result to (void) does not stop the warnings; a more elaborate fudge is
1375        needed. I've used a macro to implement this.
1376    
1377    7.  Minor change to pcretest.c to avoid a compiler warning.
1378    
1379    8.  Added four artifical Unicode properties to help with an option to make
1380        \s etc use properties (see next item). The new properties are: Xan
1381        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1382    
1383    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1384        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1385        this option. Modified pcretest to add /W to test this facility. Added
1386        REG_UCP to make it available via the POSIX interface.
1387    
1388    10. Added --line-buffered to pcregrep.
1389    
1390    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1391        studied, and the match started with a letter with a code point greater than
1392        127 whose first byte was different to the first byte of the other case of
1393        the letter, the other case of this starting letter was not recognized
1394        (#976).
1395    
1396    12. If a pattern that was studied started with a repeated Unicode property
1397        test, for example, \p{Nd}+, there was the theoretical possibility of
1398        setting up an incorrect bitmap of starting bytes, but fortunately it could
1399        not have actually happened in practice until change 8 above was made (it
1400        added property types that matched character-matching opcodes).
1401    
1402    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1403        possible starting bytes for non-anchored patterns.
1404    
1405    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1406        \R, and also a number of cases that involve Unicode properties, both
1407        explicit and implicit when PCRE_UCP is set.
1408    
1409    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1410        input, it could crash or give wrong results if characters with values
1411        greater than 0xc0 were present in the subject string. (Detail: it assumed
1412        UTF-8 input when processing these items.)
1413    
1414    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1415        size_t is 64-bit (#991).
1416    
1417    17. Added a check for running out of memory when PCRE is compiled with
1418        --disable-stack-for-recursion (#990).
1419    
1420    18. If the last data line in a file for pcretest does not have a newline on
1421        the end, a newline was missing in the output.
1422    
1423    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1424        less than 128) in its various bitmaps. However, there is a facility for
1425        generating tables according to the current locale when PCRE is compiled. It
1426        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1427        characters, are recognized by isspace() and therefore were getting set in
1428        these tables, and indeed these tables seem to approximate to ISO 8859. This
1429        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1430        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1431        which of course cannot start UTF-8 characters. I have changed the code so
1432        that only real ASCII characters (less than 128) and the correct starting
1433        bytes for UTF-8 encodings are set for characters greater than 127 when in
1434        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1435        altogether.)
1436    
1437    20. Added the /T option to pcretest so as to be able to run tests with non-
1438        standard character tables, thus making it possible to include the tests
1439        used for 19 above in the standard set of tests.
1440    
1441    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1442        reference to a subpattern the other side of a comment that contains an
1443        opening parenthesis caused either an internal compiling error, or a
1444        reference to the wrong subpattern.
1445    
1446    
1447    Version 8.02 19-Mar-2010
1448    ------------------------
1449    
1450    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1451    
1452    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1453        configured.
1454    
1455    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1456        original author of that file, following a query about its status.
1457    
1458    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1459        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1460    
1461    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1462        quantifier applied to a forward-referencing subroutine call, could compile
1463        incorrect code or give the error "internal error: previously-checked
1464        referenced subpattern not found".
1465    
1466    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1467        variables to point to external functions. For these systems, therefore,
1468        pcre_malloc etc. are now initialized to local functions that call the
1469        relevant global functions.
1470    
1471    7.  There were two entries missing in the vectors called coptable and poptable
1472        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1473        I've fixed the data, and added a kludgy way of testing at compile time that
1474        the lengths are correct (equal to the number of opcodes).
1475    
1476    8.  Following on from 7, I added a similar kludge to check the length of the
1477        eint vector in pcreposix.c.
1478    
1479    9.  Error texts for pcre_compile() are held as one long string to avoid too
1480        much relocation at load time. To find a text, the string is searched,
1481        counting zeros. There was no check for running off the end of the string,
1482        which could happen if a new error number was added without updating the
1483        string.
1484    
1485    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1486    
1487    11. \K was not working if it appeared in an atomic group or in a group that
1488        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1489        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1490        the assertion is positive, but not if it is negative.
1491    
1492    12. Change 11 fortuitously reduced the size of the stack frame used in the
1493        "match()" function of pcre_exec.c by one pointer. Forthcoming
1494        implementation of support for (*MARK) will need an extra pointer on the
1495        stack; I have reserved it now, so that the stack frame size does not
1496        decrease.
1497    
1498    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1499        item in branch that calls a recursion is a subroutine call - as in the
1500        second branch in the above example - was incorrectly given the compile-
1501        time error "recursive call could loop indefinitely" because pcre_compile()
1502        was not correctly checking the subroutine for matching a non-empty string.
1503    
1504    14. The checks for overrunning compiling workspace could trigger after an
1505        overrun had occurred. This is a "should never occur" error, but it can be
1506        triggered by pathological patterns such as hundreds of nested parentheses.
1507        The checks now trigger 100 bytes before the end of the workspace.
1508    
1509    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1510    
1511    
1512    Version 8.01 19-Jan-2010
1513    ------------------------
1514    
1515    1.  If a pattern contained a conditional subpattern with only one branch (in
1516        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1517        computed the wrong minimum data length (which is of course zero for such
1518        subpatterns). This could cause incorrect "no match" results.
1519    
1520    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1521        the pattern is reset in the first branch, pcre_compile() failed with
1522        "internal error: code overflow at offset...". This happened only when
1523        the reset was to the original external option setting. (An optimization
1524        abstracts leading options settings into an external setting, which was the
1525        cause of this.)
1526    
1527    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1528        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1529        assertion pattern did not match (meaning that the assertion was true), it
1530        was incorrectly treated as false if the SKIP had been reached during the
1531        matching. This also applied to assertions used as conditions.
1532    
1533    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1534        assertion subpattern, including such a pattern used as a condition,
1535        unpredictable results occurred, instead of the error return
1536        PCRE_ERROR_DFA_UITEM.
1537    
1538    5.  The C++ GlobalReplace function was not working like Perl for the special
1539        situation when an empty string is matched. It now does the fancy magic
1540        stuff that is necessary.
1541    
1542    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1543        removed. (These were left over from very, very early versions of PCRE.)
1544    
1545    7.  Some cosmetic changes to the code to make life easier when compiling it
1546        as part of something else:
1547    
1548        (a) Change DEBUG to PCRE_DEBUG.
1549    
1550        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1551            called "current" as "current_branch", to prevent a collision with the
1552            Linux macro when compiled as a kernel module.
1553    
1554        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1555            prevent a collision with the Linux macro when compiled as a kernel
1556            module.
1557    
1558    8.  In pcre_compile() there are some checks for integer overflows that used to
1559        cast potentially large values to (double). This has been changed to that
1560        when building, a check for int64_t is made, and if it is found, it is used
1561        instead, thus avoiding the use of floating point arithmetic. (There is no
1562        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1563        double.
1564    
1565    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1566        2005 (difference between two addresses compared to an unsigned value).
1567    
1568    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1569        custom one, because of the following reported problem in Windows:
1570    
1571          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1572              under Win32.
1573          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1574              therefore missing the function definition.
1575          - The compiler thus generates a "C" signature for the test function.
1576          - The linker fails to find the "C" function.
1577          - PCRE fails to configure if asked to do so against libbz2.
1578    
1579    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1580        messages were output:
1581    
1582          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1583          rerunning libtoolize, to keep the correct libtool macros in-tree.
1584          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1585    
1586        I have done both of these things.
1587    
1588    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1589        most of the time, it *can* run out if it is given a pattern that contains a
1590        runaway infinite recursion. I updated the discussion in the pcrestack man
1591        page.
1592    
1593    13. Now that we have gone to the x.xx style of version numbers, the minor
1594        version may start with zero. Using 08 or 09 is a bad idea because users
1595        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1596        interpreted as invalid octal numbers. I've updated the previous comment in
1597        configure.ac, and also added a check that gives an error if 08 or 09 are
1598        used.
1599    
1600    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1601        causing partial matching to fail when the end of the subject matched \W
1602        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1603    
1604    15. There were some discrepancies between the declarations in pcre_internal.h
1605        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1606        their definitions. The declarations used "const uschar *" and the
1607        definitions used USPTR. Even though USPTR is normally defined as "const
1608        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1609        reported that: "This difference in casting confuses some C++ compilers, for
1610        example, SunCC recognizes above declarations as different functions and
1611        generates broken code for hbpcre." I have changed the declarations to use
1612        USPTR.
1613    
1614    16. GNU libtool is named differently on some systems. The autogen.sh script now
1615        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1616        (FreeBSD).
1617    
1618    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1619        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1620        comment: "Figure out how to create a longlong from a string: strtoll and
1621        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1622        instance, but it only takes 2 args instead of 3!"
1623    
1624    18. A subtle bug concerned with back references has been fixed by a change of
1625        specification, with a corresponding code fix. A pattern such as
1626        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1627        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1628        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1629        same bug. Such groups have to be quantified to be useful, or contained
1630        inside another quantified group. (If there's no repetition, the reference
1631        can never match.) The problem arises because, having left the group and
1632        moved on to the rest of the pattern, a later failure that backtracks into
1633        the group uses the captured value from the final iteration of the group
1634        rather than the correct earlier one. I have fixed this in PCRE by forcing
1635        any group that contains a reference to itself to be an atomic group; that
1636        is, there cannot be any backtracking into it once it has completed. This is
1637        similar to recursive and subroutine calls.
1638    
1639    
1640    Version 8.00 19-Oct-09
1641    ----------------------
1642    
1643    1.  The table for translating pcre_compile() error codes into POSIX error codes
1644        was out-of-date, and there was no check on the pcre_compile() error code
1645        being within the table. This could lead to an OK return being given in
1646        error.
1647    
1648    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1649        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1650        in a Windows environment.
1651    
1652    3.  The pcregrep --count option prints the count for each file even when it is
1653        zero, as does GNU grep. However, pcregrep was also printing all files when
1654        --files-with-matches was added. Now, when both options are given, it prints
1655        counts only for those files that have at least one match. (GNU grep just
1656        prints the file name in this circumstance, but including the count seems
1657        more useful - otherwise, why use --count?) Also ensured that the
1658        combination -clh just lists non-zero counts, with no names.
1659    
1660    4.  The long form of the pcregrep -F option was incorrectly implemented as
1661        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1662        but it seems right to fix it, and I didn't think it was worth preserving
1663        the old behaviour.
1664    
1665    5.  The command line items --regex=pattern and --regexp=pattern were not
1666        recognized by pcregrep, which required --regex pattern or --regexp pattern
1667        (with a space rather than an '='). The man page documented the '=' forms,
1668        which are compatible with GNU grep; these now work.
1669    
1670    6.  No libpcreposix.pc file was created for pkg-config; there was just
1671        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1672    
1673    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1674        when UCP support is not needed, by modifying the Python script that
1675        generates it from Unicode data files. This should not matter if the module
1676        is correctly used as a library, but I received one complaint about 50K of
1677        unwanted data. My guess is that the person linked everything into his
1678        program rather than using a library. Anyway, it does no harm.
1679    
1680    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1681        was a minimum greater than 1 for a wide character in a possessive
1682        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1683        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1684        character. Chaos in the form of incorrect output or a compiling loop could
1685        result.
1686    
1687    9.  The restrictions on what a pattern can contain when partial matching is
1688        requested for pcre_exec() have been removed. All patterns can now be
1689        partially matched by this function. In addition, if there are at least two
1690        slots in the offset vector, the offset of the earliest inspected character
1691        for the match and the offset of the end of the subject are set in them when
1692        PCRE_ERROR_PARTIAL is returned.
1693    
1694    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1695        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1696        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1697        and may be more useful for multi-segment matching.
1698    
1699    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1700        used to be given if ever the end of the subject was reached; now it is
1701        given only if matching could not proceed because another character was
1702        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1703        string "Z", which now yields "no match" instead of "partial match". In the
1704        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1705        final character ended with (*FAIL).
1706    
1707    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1708        if the pattern had a "must contain" character that was already found in the
1709        earlier partial match, unless partial matching was again requested. For
1710        example, with the pattern /dog.(body)?/, the "must contain" character is
1711        "g". If the first part-match was for the string "dog", restarting with
1712        "sbody" failed. This bug has been fixed.
1713    
1714    13. The string returned by pcre_dfa_exec() after a partial match has been
1715        changed so that it starts at the first inspected character rather than the
1716        first character of the match. This makes a difference only if the pattern
1717        starts with a lookbehind assertion or \b or \B (\K is not supported by
1718        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1719        matching functions compatible, and I think it's the right thing to do.
1720    
1721    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1722        so that the demonstration program is easily available in environments where
1723        PCRE has not been installed from source.
1724    
1725    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1726        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1727        library.
1728    
1729    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1730        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1731        is not the first non-POSIX option to be added. Clearly some people find
1732        these options useful.
1733    
1734    17. If a caller to the POSIX matching function regexec() passes a non-zero
1735        value for nmatch with a NULL value for pmatch, the value of
1736        nmatch is forced to zero.
1737    
1738    18. RunGrepTest did not have a test for the availability of the -u option of
1739        the diff command, as RunTest does. It now checks in the same way as
1740        RunTest, and also checks for the -b option.
1741    
1742    19. If an odd number of negated classes containing just a single character
1743        interposed, within parentheses, between a forward reference to a named
1744        subpattern and the definition of the subpattern, compilation crashed with
1745        an internal error, complaining that it could not find the referenced
1746        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1747        [The bug was that it was starting one character too far in when skipping
1748        over the character class, thus treating the ] as data rather than
1749        terminating the class. This meant it could skip too much.]
1750    
1751    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1752        /g option in pcretest when the pattern contains \K, which makes it possible
1753        to have an empty string match not at the start, even when the pattern is
1754        anchored. Updated pcretest and pcredemo to use this option.
1755    
1756    21. If the maximum number of capturing subpatterns in a recursion was greater
1757        than the maximum at the outer level, the higher number was returned, but
1758        with unset values at the outer level. The correct (outer level) value is
1759        now given.
1760    
1761    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1762        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1763        make it do so. The string so far is captured, making this feature
1764        compatible with Perl.
1765    
1766    23. The tests have been re-organized, adding tests 11 and 12, to make it
1767        possible to check the Perl 5.10 features against Perl 5.10.
1768    
1769    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1770        pattern matches a fixed length string. PCRE did not allow this; now it
1771        does. Neither allows recursion.
1772    
1773    25. I finally figured out how to implement a request to provide the minimum
1774        length of subject string that was needed in order to match a given pattern.
1775        (It was back references and recursion that I had previously got hung up
1776        on.) This code has now been added to pcre_study(); it finds a lower bound
1777        to the length of subject needed. It is not necessarily the greatest lower
1778        bound, but using it to avoid searching strings that are too short does give
1779        some useful speed-ups. The value is available to calling programs via
1780        pcre_fullinfo().
1781    
1782    26. While implementing 25, I discovered to my embarrassment that pcretest had
1783        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1784        study optimizations had never been tested with that matching function.
1785        Oops. What is worse, even when it was passed study data, there was a bug in
1786        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1787        were also very few tests of studied patterns with pcre_dfa_exec().
1788    
1789    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1790        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1791        on the other side of the coin, they are no longer allowed to have different
1792        names, because these cannot be distinguished in PCRE, and this has caused
1793        confusion. (This is a difference from Perl.)
1794    
1795    28. When duplicate subpattern names are present (necessarily with different
1796        numbers, as required by 27 above), and a test is made by name in a
1797        conditional pattern, either for a subpattern having been matched, or for
1798        recursion in such a pattern, all the associated numbered subpatterns are
1799        tested, and the overall condition is true if the condition is true for any
1800        one of them. This is the way Perl works, and is also more like the way
1801        testing by number works.
1802    
1803    
1804    Version 7.9 11-Apr-09
1805    ---------------------
1806    
1807    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1808        (pcretest), all targets were linked against these libraries. This included
1809        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1810        libraries. This caused unwanted dependencies to be created. This problem
1811        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1812        pcretest is linked with readline.
1813    
1814    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1815        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1816        moved outside it again, because FALSE and TRUE are already defined in AIX,
1817        but BOOL is not.
1818    
1819    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1820        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1821    
1822    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1823        hyphens) following file names and line numbers when outputting matching
1824        lines. This is not true; no spaces are inserted. I have also clarified the
1825        wording for the --colour (or --color) option.
1826    
1827    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1828        was not coloured; this is different to GNU grep, so I have changed it to be
1829        the same.
1830    
1831    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1832        each matching line was coloured. Now it goes on to look for further matches
1833        of any of the test patterns, which is the same behaviour as GNU grep.
1834    
1835    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1836        doesn't make sense to accept an empty string match in pcregrep, so I have
1837        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1838        seems to be how GNU grep behaves. [But see later change 40 for release
1839        8.33.]
1840    
1841    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1842        start or after a newline", because the conditional assertion was not being
1843        correctly handled. The rule now is that both the assertion and what follows
1844        in the first alternative must satisfy the test.
1845    
1846    9.  If auto-callout was enabled in a pattern with a conditional group whose
1847        condition was an assertion, PCRE could crash during matching, both with
1848        pcre_exec() and pcre_dfa_exec().
1849    
1850    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1851        used for matching.
1852    
1853    11. Unicode property support in character classes was not working for
1854        characters (bytes) greater than 127 when not in UTF-8 mode.
1855    
1856    12. Added the -M command line option to pcretest.
1857    
1858    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1859    
1860    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1861    
1862    16. Added comments and documentation about mis-use of no_arg in the C++
1863        wrapper.
1864    
1865    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1866        from Martin Jerabek that uses macro names for all relevant character and
1867        string constants.
1868    
1869    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1870        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1871        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1872        these, but not everybody uses configure.
1873    
1874    19. A conditional group that had only one branch was not being correctly
1875        recognized as an item that could match an empty string. This meant that an
1876        enclosing group might also not be so recognized, causing infinite looping
1877        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1878        with the subject "ab", where knowledge that the repeated group can match
1879        nothing is needed in order to break the loop.
1880    
1881    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1882        exec(), but without supplying a callout function, matching went wrong.
1883    
1884    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1885        leak if the size of the offset vector was greater than 30. When the vector
1886        is smaller, the saved offsets during recursion go onto a local stack
1887        vector, but for larger vectors malloc() is used. It was failing to free
1888        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1889        error, in fact).
1890    
1891    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1892        heapframe that is used only when UTF-8 support is enabled. This caused no
1893        problem, but was untidy.
1894    
1895    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1896        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1897        included within another project.
1898    
1899    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1900        slightly modified by me:
1901    
1902          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1903              not building pcregrep.
1904    
1905          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1906              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1907    
1908    25. Forward references, both numeric and by name, in patterns that made use of
1909        duplicate group numbers, could behave incorrectly or give incorrect errors,
1910        because when scanning forward to find the reference group, PCRE was not
1911        taking into account the duplicate group numbers. A pattern such as
1912        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1913    
1914    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1915        the feature of a custom pointer more persuasive (as requested by a user).
1916    
1917    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1918        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1919        pre-defined.
1920    
1921    28. Added support for (*UTF8) at the start of a pattern.
1922    
1923    29. Arrange for flags added by the "release type" setting in CMake to be shown
1924        in the configuration summary.
1925    
1926    
1927    Version 7.8 05-Sep-08
1928  ---------------------  ---------------------
1929    
1930  1.  Replaced UCP searching code with optimized version as implemented for Ad  1.  Replaced UCP searching code with optimized version as implemented for Ad
# Line 31  Version 8.0 02 Jul-08 Line 1954  Version 8.0 02 Jul-08
1954    
1955  7.  Added two (int) casts to pcregrep when printing the difference of two  7.  Added two (int) casts to pcregrep when printing the difference of two
1956      pointers, in case they are 64-bit values.      pointers, in case they are 64-bit values.
1957    
1958  8.  Added comments about Mac OS X stack usage to the pcrestack man page and to  8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1959      test 2 if it fails.      test 2 if it fails.
1960    
1961  9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,  9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1962      and a #define of that name to empty if it is not externally set. This is to      and a #define of that name to empty if it is not externally set. This is to
1963      allow users of MSVC to set it if necessary.      allow users of MSVC to set it if necessary.
1964    
1965  10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from  10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1966      the convenience functions in the pcre_get.c source file.      the convenience functions in the pcre_get.c source file.
1967    
1968  11. An option change at the start of a pattern that had top-level alternatives  11. An option change at the start of a pattern that had top-level alternatives
1969      could cause overwriting and/or a crash. This command provoked a crash in      could cause overwriting and/or a crash. This command provoked a crash in
1970      some environments:      some environments:
1971    
1972        printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest        printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1973    
1974      This potential security problem was recorded as CVE-2008-2371.      This potential security problem was recorded as CVE-2008-2371.
1975    
1976  12. For a pattern where the match had to start at the beginning or immediately  12. For a pattern where the match had to start at the beginning or immediately
1977      after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and      after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1978      pcre_dfa_exec() could read past the end of the passed subject if there was      pcre_dfa_exec() could read past the end of the passed subject if there was
1979      no match. To help with detecting such bugs (e.g. with valgrind), I modified      no match. To help with detecting such bugs (e.g. with valgrind), I modified
1980      pcretest so that it places the subject at the end of its malloc-ed buffer.      pcretest so that it places the subject at the end of its malloc-ed buffer.
1981    
1982    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1983        exec() might read past the end of the data buffer in UTF-8 mode.
1984    
1985    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1986        the data contained the byte 0x85 as part of a UTF-8 character within its
1987        first line. This applied both to normal and DFA matching.
1988    
1989    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1990        /^[^d]*?$/8 failed to match "abc".
1991    
1992    16. Added a missing copyright notice to pcrecpp_internal.h.
1993    
1994    17. Make it more clear in the documentation that values returned from
1995        pcre_exec() in ovector are byte offsets, not character counts.
1996    
1997    18. Tidied a few places to stop certain compilers from issuing warnings.
1998    
1999    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
2000        supplied by Stefan Weber. I made a further small update for 7.8 because
2001        there is a change of source arrangements: the pcre_searchfuncs.c module is
2002        replaced by pcre_ucd.c.
2003    
2004    
2005  Version 7.7 07-May-08  Version 7.7 07-May-08
2006  ---------------------  ---------------------
# Line 134  Version 7.7 07-May-08 Line 2079  Version 7.7 07-May-08
2079      containing () gave an internal compiling error instead of "reference to      containing () gave an internal compiling error instead of "reference to
2080      non-existent subpattern". Fortunately, when the pattern did exist, the      non-existent subpattern". Fortunately, when the pattern did exist, the
2081      compiled code was correct. (When scanning forwards to check for the      compiled code was correct. (When scanning forwards to check for the
2082      existencd of the subpattern, it was treating the data ']' as terminating      existence of the subpattern, it was treating the data ']' as terminating
2083      the class, so got the count wrong. When actually compiling, the reference      the class, so got the count wrong. When actually compiling, the reference
2084      was subsequently set up correctly.)      was subsequently set up correctly.)
2085    

Legend:
Removed from v.361  
changed lines
  Added in v.1389

  ViewVC Help
Powered by ViewVC 1.1.5