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

Diff of /code/trunk/ChangeLog

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

revision 221 by ph10, Fri Aug 17 09:25:08 2007 UTC revision 1387 by ph10, Sat Nov 2 18:29:05 2013 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 7.3 17-Aug-07  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    
156    Version 8.33 28-May-2013
157    ------------------------
158    
159    1.  Added 'U' to some constants that are compared to unsigned integers, to
160        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
161        variables that are added to signed variables, to ensure the result is
162        signed and can be negated.
163    
164    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
165        CMake config files.
166    
167    3.  Revise the creation of config.h.generic so that all boolean macros are
168        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
169        overriding via -D on the command line possible.
170    
171    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
172        to unsigned int is reported to make a quite noticeable speed difference in
173        a specific Windows environment. Testing on Linux did also appear to show
174        some benefit (and it is clearly not harmful). Also fixed the definition of
175        Xop which should be unsigned.
176    
177    5.  Related to (4), changing the definition of the intermediate variable cc
178        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
179        improvements.
180    
181    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
182        unnecessary spaces.
183    
184    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
185        and later.
186    
187    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
188    
189    9.  Optimizing fast_forward_start_bits in JIT.
190    
191    10. Adding support for callouts in JIT, and fixing some issues revealed
192        during this work. Namely:
193    
194        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
195    
196        (b) Minimum length was not checked before the matching is started.
197    
198    11. The value of capture_last that is passed to callouts was incorrect in some
199        cases when there was a capture on one path that was subsequently abandoned
200        after a backtrack. Also, the capture_last value is now reset after a
201        recursion, since all captures are also reset in this case.
202    
203    12. The interpreter no longer returns the "too many substrings" error in the
204        case when an overflowing capture is in a branch that is subsequently
205        abandoned after a backtrack.
206    
207    13. In the pathological case when an offset vector of size 2 is used, pcretest
208        now prints out the matched string after a yield of 0 or 1.
209    
210    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
211        Only supported by the JIT compiler at the moment.
212    
213    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
214    
215    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
216        offset of the starting point of the matching process, provided the offsets
217        vector is large enough.
218    
219    17. The \A escape now records a lookbehind value of 1, though its execution
220        does not actually inspect the previous character. This is to ensure that,
221        in partial multi-segment matching, at least one character from the old
222        segment is retained when a new segment is processed. Otherwise, if there
223        are no lookbehinds in the pattern, \A might match incorrectly at the start
224        of a new segment.
225    
226    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
227    
228    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
229        gives some modest performance improvement in 8-bit mode.
230    
231    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
232        be expressed in certain programming languages using Universal Character
233        Names.
234    
235    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
236        which points out that "non characters" are not "characters that may not
237        appear in Unicode strings" but rather "characters that are reserved for
238        internal use and have only local meaning".
239    
240    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
241        there was a conditional group that depended on an assertion, if the
242        assertion was false, the callout that immediately followed the alternation
243        in the condition was skipped when pcre_exec() was used for matching.
244    
245    23. Allow an explicit callout to be inserted before an assertion that is the
246        condition for a conditional group, for compatibility with automatic
247        callouts, which always insert a callout at this point.
248    
249    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
250        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
251    
252    25. (*PRUNE) is now supported by the JIT compiler.
253    
254    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
255    
256    27. Fix the case where there are two or more SKIPs with arguments that may be
257        ignored.
258    
259    28. (*SKIP) is now supported by the JIT compiler.
260    
261    29. (*THEN) is now supported by the JIT compiler.
262    
263    30. Update RunTest with additional test selector options.
264    
265    31. The way PCRE handles backtracking verbs has been changed in two ways.
266    
267        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
268        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
269        backtracking. In some cases this makes it more Perl-compatible, but Perl's
270        rather obscure rules do not always do the same thing.
271    
272        (2) Previously, backtracking verbs were confined within assertions. This is
273        no longer the case for positive assertions, except for (*ACCEPT). Again,
274        this sometimes improves Perl compatibility, and sometimes does not.
275    
276    32. A number of tests that were in test 2 because Perl did things differently
277        have been moved to test 1, because either Perl or PCRE has changed, and
278        these tests are now compatible.
279    
280    32. Backtracking control verbs are now handled in the same way in JIT and
281        interpreter.
282    
283    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
284        contained a forward subroutine reference caused a compile error.
285    
286    34. Auto-detect and optimize limited repetitions in JIT.
287    
288    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
289        blocking (*UTF) etc.
290    
291    36. In the interpreter, maximizing pattern repetitions for characters and
292        character types now use tail recursion, which reduces stack usage.
293    
294    37. The value of the max lookbehind was not correctly preserved if a compiled
295        and saved regex was reloaded on a host of different endianness.
296    
297    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
298        of the compiled pattern block, expand the flags field from 16 to 32 bits
299        because it was almost full.
300    
301    39. Try madvise first before posix_madvise.
302    
303    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
304        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
305        original change locked out all matches of empty strings. This has been
306        changed so that one match of an empty string per line is recognized.
307        Subsequent searches on the same line (for colouring or for --only-matching,
308        for example) do not recognize empty strings.
309    
310    41. Applied a user patch to fix a number of spelling mistakes in comments.
311    
312    42. Data lines longer than 65536 caused pcretest to crash.
313    
314    43. Clarified the data type for length and startoffset arguments for pcre_exec
315        and pcre_dfa_exec in the function-specific man pages, where they were
316        explicitly stated to be in bytes, never having been updated. I also added
317        some clarification to the pcreapi man page.
318    
319    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
320        a segmentation fault.
321    
322    
323    Version 8.32 30-November-2012
324    -----------------------------
325    
326    1.  Improved JIT compiler optimizations for first character search and single
327        character iterators.
328    
329    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
330        Patch by Daniel Richard G.
331    
332    3.  Single character iterator optimizations in the JIT compiler.
333    
334    4.  Improved JIT compiler optimizations for character ranges.
335    
336    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
337        Reported by Giuseppe D'Angelo.
338    
339    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
340        of a line, was being set incorrectly in cases where .* appeared inside
341        atomic brackets at the start of a pattern, or where there was a subsequent
342        *PRUNE or *SKIP.
343    
344    7.  Improved instruction cache flush for POWER/PowerPC.
345        Patch by Daniel Richard G.
346    
347    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
348        grep:
349    
350        (a) There is now no limit to the number of patterns to be matched.
351    
352        (b) An error is given if a pattern is too long.
353    
354        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
355            are now supported.
356    
357        (d) --exclude-from and --include-from (multiple use) have been added.
358    
359        (e) Exclusions and inclusions now apply to all files and directories, not
360            just to those obtained from scanning a directory recursively.
361    
362        (f) Multiple uses of -f and --file-list are now supported.
363    
364        (g) In a Windows environment, the default for -d has been changed from
365            "read" (the GNU grep default) to "skip", because otherwise the presence
366            of a directory in the file list provokes an error.
367    
368        (h) The documentation has been revised and clarified in places.
369    
370    9.  Improve the matching speed of capturing brackets.
371    
372    10. Changed the meaning of \X so that it now matches a Unicode extended
373        grapheme cluster.
374    
375    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
376        out POSIX threads when JIT support is configured.
377    
378    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
379    
380    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
381        structure could go wrong in environments where size_t is not the same size
382        as int.
383    
384    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
385        set.
386    
387    15. The EBCDIC support had decayed; later updates to the code had included
388        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
389        general tidy up of EBCDIC-related issues, and the documentation was also
390        not quite right. There is now a test that can be run on ASCII systems to
391        check some of the EBCDIC-related things (but is it not a full test).
392    
393    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
394        in a small tidy to the code.
395    
396    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
397    
398    18. If the --only-matching (-o) option in pcregrep is specified multiple
399        times, each one causes appropriate output. For example, -o1 -o2 outputs the
400        substrings matched by the 1st and 2nd capturing parentheses. A separating
401        string can be specified by --om-separator (default empty).
402    
403    19. Improving the first n character searches.
404    
405    20. Turn case lists for horizontal and vertical white space into macros so that
406        they are defined only once.
407    
408    21. This set of changes together give more compatible Unicode case-folding
409        behaviour for characters that have more than one other case when UCP
410        support is available.
411    
412        (a) The Unicode property table now has offsets into a new table of sets of
413            three or more characters that are case-equivalent. The MultiStage2.py
414            script that generates these tables (the pcre_ucd.c file) now scans
415            CaseFolding.txt instead of UnicodeData.txt for character case
416            information.
417    
418        (b) The code for adding characters or ranges of characters to a character
419            class has been abstracted into a generalized function that also handles
420            case-independence. In UTF-mode with UCP support, this uses the new data
421            to handle characters with more than one other case.
422    
423        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
424            whose other case is greater than 256 are now correctly matched
425            caselessly. Previously, the high codepoint matched the low one, but not
426            vice versa.
427    
428        (d) The processing of \h, \H, \v, and \ in character classes now makes use
429            of the new class addition function, using character lists defined as
430            macros alongside the case definitions of 20 above.
431    
432        (e) Caseless back references now work with characters that have more than
433            one other case.
434    
435        (f) General caseless matching of characters with more than one other case
436            is supported.
437    
438    22. Unicode character properties were updated from Unicode 6.2.0
439    
440    23. Improved CMake support under Windows. Patch by Daniel Richard G.
441    
442    24. Add support for 32-bit character strings, and UTF-32
443    
444    25. Major JIT compiler update (code refactoring and bugfixing).
445        Experimental Sparc 32 support is added.
446    
447    26. Applied a modified version of Daniel Richard G's patch to create
448        pcre.h.generic and config.h.generic by "make" instead of in the
449        PrepareRelease script.
450    
451    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
452        pcre_compile.c when checking for a zero character.
453    
454    28. Introducing a native interface for JIT. Through this interface, the compiled
455        machine code can be directly executed. The purpose of this interface is to
456        provide fast pattern matching, so several sanity checks are not performed.
457        However, feature tests are still performed. The new interface provides
458        1.4x speedup compared to the old one.
459    
460    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
461        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
462        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
463    
464    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
465        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
466        the "old" RFC 2279). Instead, it ended up passing a negative length to
467        pcre_exec().
468    
469    31. Add support for GCC's visibility feature to hide internal functions.
470    
471    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
472        "unknown -C option" after outputting 0 or 1.
473    
474    33. There is now support for generating a code coverage report for the test
475        suite in environments where gcc is the compiler and lcov is installed. This
476        is mainly for the benefit of the developers.
477    
478    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
479        unaddressable using valgrind annotations, allowing valgrind to detect
480        invalid memory accesses. This is mainly for the benefit of the developers.
481    
482    25. (*UTF) can now be used to start a pattern in any of the three libraries.
483    
484    26. Give configure error if --enable-cpp but no C++ compiler found.
485    
486    
487    Version 8.31 06-July-2012
488    -------------------------
489    
490    1.  Fixing a wrong JIT test case and some compiler warnings.
491    
492    2.  Removed a bashism from the RunTest script.
493    
494    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
495        to unsigned type, result still unsigned" that was given by an MS compiler
496        on encountering the code "-sizeof(xxx)".
497    
498    4.  Partial matching support is added to the JIT compiler.
499    
500    5.  Fixed several bugs concerned with partial matching of items that consist
501        of more than one character:
502    
503        (a) /^(..)\1/ did not partially match "aba" because checking references was
504            done on an "all or nothing" basis. This also applied to repeated
505            references.
506    
507        (b) \R did not give a hard partial match if \r was found at the end of the
508            subject.
509    
510        (c) \X did not give a hard partial match after matching one or more
511            characters at the end of the subject.
512    
513        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
514            a partial match for the string "\r".
515    
516        (e) When newline was set to CRLF, the metacharacter "." did not recognize
517            a partial match for a CR character at the end of the subject string.
518    
519    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
520        running pcretest, the text "(JIT)" added to the output whenever JIT is
521        actually used to run the match.
522    
523    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
524        or /S+[+] with a digit between 1 and 7.
525    
526    8.  OP_NOT now supports any UTF character not just single-byte ones.
527    
528    9.  (*MARK) control verb is now supported by the JIT compiler.
529    
530    10. The command "./RunTest list" lists the available tests without actually
531        running any of them. (Because I keep forgetting what they all are.)
532    
533    11. Add PCRE_INFO_MAXLOOKBEHIND.
534    
535    12. Applied a (slightly modified) user-supplied patch that improves performance
536        when the heap is used for recursion (compiled with --disable-stack-for-
537        recursion). Instead of malloc and free for each heap frame each time a
538        logical recursion happens, frames are retained on a chain and re-used where
539        possible. This sometimes gives as much as 30% improvement.
540    
541    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
542        call.
543    
544    14. As documented, (*COMMIT) is now confined to within a positive assertion.
545    
546    15. It is now possible to link pcretest with libedit as an alternative to
547        libreadline.
548    
549    16. (*COMMIT) control verb is now supported by the JIT compiler.
550    
551    17. The Unicode data tables have been updated to Unicode 6.1.0.
552    
553    18. Added --file-list option to pcregrep.
554    
555    19. Added binary file support to pcregrep, including the -a, --binary-files,
556        -I, and --text options.
557    
558    20. The madvise function is renamed for posix_madvise for QNX compatibility
559        reasons. Fixed by Giuseppe D'Angelo.
560    
561    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
562    
563    22. Changed the option for creating symbolic links for 16-bit man pages from
564        -s to -sf so that re-installing does not cause issues.
565    
566    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
567    
568    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
569        matches in certain environments (the workspace was not being correctly
570        retained). Also added to pcre_dfa_exec() a simple plausibility check on
571        some of the workspace data at the beginning of a restart.
572    
573    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
574        was not doing so when it should - probably a typo introduced by SVN 528
575        (change 8.10/14).
576    
577    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
578        \w+ when the character tables indicated that \x{c4} was a word character.
579        There were several related cases, all because the tests for doing a table
580        lookup were testing for characters less than 127 instead of 255.
581    
582    27. If a pattern contains capturing parentheses that are not used in a match,
583        their slots in the ovector are set to -1. For those that are higher than
584        any matched groups, this happens at the end of processing. In the case when
585        there were back references that the ovector was too small to contain
586        (causing temporary malloc'd memory to be used during matching), and the
587        highest capturing number was not used, memory off the end of the ovector
588        was incorrectly being set to -1. (It was using the size of the temporary
589        memory instead of the true size.)
590    
591    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
592        ovector size, it uses memory at the end of the block that it has got.
593    
594    29. Check for an overlong MARK name and give an error at compile time. The
595        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
596    
597    30. JIT compiler update.
598    
599    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
600        Rill for the patch.
601    
602    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
603    
604    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
605    
606    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
607        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
608        was enabled.
609    
610    35. Improve JIT code generation for greedy plus quantifier.
611    
612    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
613        1 to "aa" instead of to an empty string. The bug affected repeated groups
614        that could potentially match an empty string.
615    
616    37. Optimizing single character iterators in JIT.
617    
618    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
619        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
620        codepoints that are too big for the mode are faulted, and in a UTF mode,
621        disallowed codepoints are also faulted.
622    
623    39. If PCRE was compiled with UTF support, in three places in the DFA
624        matcher there was code that should only have been obeyed in UTF mode, but
625        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
626        processing when bytes with values greater than 127 were present. In 16-bit
627        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
628        both cases the values are those that cannot be the first data item in a UTF
629        character. The three items that might have provoked this were recursions,
630        possessively repeated groups, and atomic groups.
631    
632    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
633        and pcregrep, because some OS require shared objects to be explicitly
634        passed to ld, causing the link step to fail if they are not.
635    
636    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
637        mode, patterns that started with \h* or \R* might be incorrectly matched.
638    
639    
640    Version 8.30 04-February-2012
641    -----------------------------
642    
643    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
644        name is defined in ctype.h.
645    
646    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
647        only in quite long subpatterns.
648    
649    3.  Removed the function pcre_info(), which has been obsolete and deprecated
650        since it was replaced by pcre_fullinfo() in February 2000.
651    
652    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
653        match a (*MARK), and the match failed at the start of the subject, a
654        reference to memory before the start of the subject could occur. This bug
655        was introduced by fix 17 of release 8.21.
656    
657    5.  A reference to an unset group with zero minimum repetition was giving
658        totally wrong answers (in non-JavaScript-compatibility mode). For example,
659        /(another)?(\1?)test/ matched against "hello world test". This bug was
660        introduced in release 8.13.
661    
662    6.  Add support for 16-bit character strings (a large amount of work involving
663        many changes and refactorings).
664    
665    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
666        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
667        from a file.
668    
669    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
670        rounding is not applied in this particular case).
671    
672    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
673        if they appear, or are escaped, in patterns.
674    
675    10. Get rid of a number of -Wunused-but-set-variable warnings.
676    
677    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
678        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
679        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
680        also returns the mark "x". This bug applied to capturing parentheses,
681        non-capturing parentheses, and atomic parentheses. It also applied to some
682        assertions.
683    
684    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
685        information out of configure.ac instead of relying on pcre.h.generic, which
686        is not stored in the repository.
687    
688    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
689        -lreadline.
690    
691    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
692    
693    15. Applied Graycode's patch to put the top-level frame on the stack rather
694        than the heap when not using the stack for recursion. This gives a
695        performance improvement in many cases when recursion is not deep.
696    
697    16. Experimental code added to "pcretest -C" to output the stack frame size.
698    
699    
700    Version 8.21 12-Dec-2011
701    ------------------------
702    
703    1.  Updating the JIT compiler.
704    
705    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
706        are added as well.
707    
708    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
709        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
710        calling _pcre_jit_exec. Some extra comments are added.
711    
712    4.  (*MARK) settings inside atomic groups that do not contain any capturing
713        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
714        was introduced by change 18 for 8.20.
715    
716    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
717        ECMA-262 standard.
718    
719    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
720        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
721        This bug was probably introduced by change 9 of 8.13.
722    
723    7.  While fixing 6 above, I noticed that a number of other items were being
724        incorrectly rejected as "not fixed length". This arose partly because newer
725        opcodes had not been added to the fixed-length checking code. I have (a)
726        corrected the bug and added tests for these items, and (b) arranged for an
727        error to occur if an unknown opcode is encountered while checking for fixed
728        length instead of just assuming "not fixed length". The items that were
729        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
730        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
731        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
732    
733    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
734        being incorrectly compiled and would have given unpredicatble results.
735    
736    9.  A possessively repeated subpattern with minimum repeat count greater than
737        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
738        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
739        the first (A) could occur when it should not.
740    
741    10. Add a cast and remove a redundant test from the code.
742    
743    11. JIT should use pcre_malloc/pcre_free for allocation.
744    
745    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
746        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
747        is anything other than /usr, -L is still shown).
748    
749    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
750    
751    14. Perl does not support \N without a following name in a [] class; PCRE now
752        also gives an error.
753    
754    15. If a forward reference was repeated with an upper limit of around 2000,
755        it caused the error "internal error: overran compiling workspace". The
756        maximum number of forward references (including repeats) was limited by the
757        internal workspace, and dependent on the LINK_SIZE. The code has been
758        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
759        the default depends on LINK_SIZE. There is a new upper limit (for safety)
760        of around 200,000 forward references. While doing this, I also speeded up
761        the filling in of repeated forward references.
762    
763    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
764        incorrectly expecting the subject to contain another "a" after the start.
765    
766    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
767        in the match, the SKIP should be ignored. This was not happening; instead
768        the SKIP was being treated as NOMATCH. For patterns such as
769        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
770        tested.
771    
772    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
773        now much more compatible with Perl, in particular in cases where the result
774        is a non-match for a non-anchored pattern. For example, if
775        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
776        "m", where previously it did not return a name. A side effect of this
777        change is that for partial matches, the last encountered mark name is
778        returned, as for non matches. A number of tests that were previously not
779        Perl-compatible have been moved into the Perl-compatible test files. The
780        refactoring has had the pleasing side effect of removing one argument from
781        the match() function, thus reducing its stack requirements.
782    
783    19. If the /S+ option was used in pcretest to study a pattern using JIT,
784        subsequent uses of /S (without +) incorrectly behaved like /S+.
785    
786    21. Retrieve executable code size support for the JIT compiler and fixing
787        some warnings.
788    
789    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
790        not work when the shorter character appeared right at the end of the
791        subject string.
792    
793    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
794        systems.
795    
796    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
797        output it when the /M option is used in pcretest.
798    
799    25. The CheckMan script was not being included in the distribution. Also, added
800        an explicit "perl" to run Perl scripts from the PrepareRelease script
801        because this is reportedly needed in Windows.
802    
803    26. If study data was being save in a file and studying had not found a set of
804        "starts with" bytes for the pattern, the data written to the file (though
805        never used) was taken from uninitialized memory and so caused valgrind to
806        complain.
807    
808    27. Updated RunTest.bat as provided by Sheri Pierce.
809    
810    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
811    
812    29. Computation of memory usage for the table of capturing group names was
813        giving an unnecessarily large value.
814    
815    
816    Version 8.20 21-Oct-2011
817    ------------------------
818    
819    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
820        a POSIX class. After further experiments with Perl, which convinced me that
821        Perl has bugs and confusions, a closing square bracket is no longer allowed
822        in a POSIX name. This bug also affected patterns with classes that started
823        with full stops.
824    
825    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
826        captured substring, but while checking the failing first alternative,
827        substring 1 is temporarily captured. If the output vector supplied to
828        pcre_exec() was not big enough for this capture, the yield of the function
829        was still zero ("insufficient space for captured substrings"). This cannot
830        be totally fixed without adding another stack variable, which seems a lot
831        of expense for a edge case. However, I have improved the situation in cases
832        such as /(a)(b)x|abc/ matched against "abc", where the return code
833        indicates that fewer than the maximum number of slots in the ovector have
834        been set.
835    
836    3.  Related to (2) above: when there are more back references in a pattern than
837        slots in the output vector, pcre_exec() uses temporary memory during
838        matching, and copies in the captures as far as possible afterwards. It was
839        using the entire output vector, but this conflicts with the specification
840        that only 2/3 is used for passing back captured substrings. Now it uses
841        only the first 2/3, for compatibility. This is, of course, another edge
842        case.
843    
844    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
845        main code base, and can be used by building with --enable-jit. When this is
846        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
847        runtime --no-jit option is given.
848    
849    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
850        ovector, the return from the function was zero, implying that there were
851        other matches that did not fit. The correct "exactly full" value is now
852        returned.
853    
854    6.  If a subpattern that was called recursively or as a subroutine contained
855        (*PRUNE) or any other control that caused it to give a non-standard return,
856        invalid errors such as "Error -26 (nested recursion at the same subject
857        position)" or even infinite loops could occur.
858    
859    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
860        computing the minimum length on reaching *ACCEPT, and so ended up with the
861        wrong value of 1 rather than 0. Further investigation indicates that
862        computing a minimum subject length in the presence of *ACCEPT is difficult
863        (think back references, subroutine calls), and so I have changed the code
864        so that no minimum is registered for a pattern that contains *ACCEPT.
865    
866    8.  If (*THEN) was present in the first (true) branch of a conditional group,
867        it was not handled as intended. [But see 16 below.]
868    
869    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
870        Sheri Pierce.
871    
872    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
873        the first byte in a match must be "a".
874    
875    11. Change 17 for 8.13 increased the recursion depth for patterns like
876        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
877        pattern contains any instances of (*THEN). If it does not, the old
878        optimizations are restored. It would be nice to do this on a per-group
879        basis, but at the moment that is not feasible.
880    
881    12. In some environments, the output of pcretest -C is CRLF terminated. This
882        broke RunTest's code that checks for the link size. A single white space
883        character after the value is now allowed for.
884    
885    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
886        For "fr", it uses the Windows-specific input and output files.
887    
888    14. If (*THEN) appeared in a group that was called recursively or as a
889        subroutine, it did not work as intended. [But see next item.]
890    
891    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
892        pattern fragments (but not containing any | characters). If A and B are
893        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
894        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
895        D. In other words, Perl considers parentheses that do not contain any |
896        characters to be part of a surrounding alternative, whereas PCRE was
897        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
898        differently. PCRE now behaves in the same way as Perl, except in the case
899        of subroutine/recursion calls such as (?1) which have in any case always
900        been different (but PCRE had them first :-).
901    
902    16. Related to 15 above: Perl does not treat the | in a conditional group as
903        creating alternatives. Such a group is treated in the same way as an
904        ordinary group without any | characters when processing (*THEN). PCRE has
905        been changed to match Perl's behaviour.
906    
907    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
908        RunGrepTest script failed.
909    
910    18. Change 22 for version 13 caused atomic groups to use more stack. This is
911        inevitable for groups that contain captures, but it can lead to a lot of
912        stack use in large patterns. The old behaviour has been restored for atomic
913        groups that do not contain any capturing parentheses.
914    
915    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
916        suppress the check for a minimum subject length at run time. (If it was
917        given to pcre_exec() or pcre_dfa_exec() it did work.)
918    
919    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
920        fail to work when decoding hex characters in data strings in EBCDIC
921        environments.
922    
923    21. It appears that in at least one Mac OS environment, the isxdigit() function
924        is implemented as a macro that evaluates to its argument more than once,
925        contravening the C 90 Standard (I haven't checked a later standard). There
926        was an instance in pcretest which caused it to go wrong when processing
927        \x{...} escapes in subject strings. The has been rewritten to avoid using
928        things like p++ in the argument of isxdigit().
929    
930    
931    Version 8.13 16-Aug-2011
932    ------------------------
933    
934    1.  The Unicode data tables have been updated to Unicode 6.0.0.
935    
936    2.  Two minor typos in pcre_internal.h have been fixed.
937    
938    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
939        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
940        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
941    
942    4.  There were a number of related bugs in the code for matching backrefences
943        caselessly in UTF-8 mode when codes for the characters concerned were
944        different numbers of bytes. For example, U+023A and U+2C65 are an upper
945        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
946        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
947        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
948        2-byte code at the end of the subject (it thought there wasn't enough data
949        left).
950    
951    5.  Comprehensive information about what went wrong is now returned by
952        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
953        as the output vector has at least 2 elements. The offset of the start of
954        the failing character and a reason code are placed in the vector.
955    
956    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
957        now returned is for the first byte of the failing character, instead of the
958        last byte inspected. This is an incompatible change, but I hope it is small
959        enough not to be a problem. It makes the returned offset consistent with
960        pcre_exec() and pcre_dfa_exec().
961    
962    7.  pcretest now gives a text phrase as well as the error number when
963        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
964        failure, the offset and reason code are output.
965    
966    8.  When \R was used with a maximizing quantifier it failed to skip backwards
967        over a \r\n pair if the subsequent match failed. Instead, it just skipped
968        back over a single character (\n). This seems wrong (because it treated the
969        two characters as a single entity when going forwards), conflicts with the
970        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
971        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
972        has been changed.
973    
974    9.  Some internal refactoring has changed the processing so that the handling
975        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
976        time (the PCRE_DOTALL option was changed this way some time ago: version
977        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
978        which was always a bit of a fudge. It also means that there is one less
979        argument for the match() function, which reduces its stack requirements
980        slightly. This change also fixes an incompatibility with Perl: the pattern
981        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
982    
983    10. More internal refactoring has drastically reduced the number of recursive
984        calls to match() for possessively repeated groups such as (abc)++ when
985        using pcre_exec().
986    
987    11. While implementing 10, a number of bugs in the handling of groups were
988        discovered and fixed:
989    
990        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
991        (a|)*(?1) gave a compile-time internal error.
992        ((a|)+)+  did not notice that the outer group could match an empty string.
993        (^a|^)+   was not marked as anchored.
994        (.*a|.*)+ was not marked as matching at start or after a newline.
995    
996    12. Yet more internal refactoring has removed another argument from the match()
997        function. Special calls to this function are now indicated by setting a
998        value in a variable in the "match data" data block.
999    
1000    13. Be more explicit in pcre_study() instead of relying on "default" for
1001        opcodes that mean there is no starting character; this means that when new
1002        ones are added and accidentally left out of pcre_study(), testing should
1003        pick them up.
1004    
1005    14. The -s option of pcretest has been documented for ages as being an old
1006        synonym of -m (show memory usage). I have changed it to mean "force study
1007        for every regex", that is, assume /S for every regex. This is similar to -i
1008        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
1009        using it. It makes it easier to run collections of tests with and without
1010        study enabled, and thereby test pcre_study() more easily. All the standard
1011        tests are now run with and without -s (but some patterns can be marked as
1012        "never study" - see 20 below).
1013    
1014    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
1015        restoration of the capturing data to the outer values was not happening
1016        correctly.
1017    
1018    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
1019        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
1020        pattern had matched an empty string, and so incorrectly returned a no
1021        match.
1022    
1023    17. There was optimizing code for the last branch of non-capturing parentheses,
1024        and also for the obeyed branch of a conditional subexpression, which used
1025        tail recursion to cut down on stack usage. Unfortunately, now that there is
1026        the possibility of (*THEN) occurring in these branches, tail recursion is
1027        no longer possible because the return has to be checked for (*THEN). These
1028        two optimizations have therefore been removed. [But see 8.20/11 above.]
1029    
1030    18. If a pattern containing \R was studied, it was assumed that \R always
1031        matched two bytes, thus causing the minimum subject length to be
1032        incorrectly computed because \R can also match just one byte.
1033    
1034    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
1035        was incorrectly computed.
1036    
1037    20. If /S is present twice on a test pattern in pcretest input, it now
1038        *disables* studying, thereby overriding the use of -s on the command line
1039        (see 14 above). This is necessary for one or two tests to keep the output
1040        identical in both cases.
1041    
1042    21. When (*ACCEPT) was used in an assertion that matched an empty string and
1043        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
1044    
1045    22. When an atomic group that contained a capturing parenthesis was
1046        successfully matched, but the branch in which it appeared failed, the
1047        capturing was not being forgotten if a higher numbered group was later
1048        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
1049        group 1 to "a", when in fact it should be unset. This applied to multi-
1050        branched capturing and non-capturing groups, repeated or not, and also to
1051        positive assertions (capturing in negative assertions does not happen
1052        in PCRE) and also to nested atomic groups.
1053    
1054    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
1055        subject after a captured substring, to make it easier to tell which of a
1056        number of identical substrings has been captured.
1057    
1058    24. The way atomic groups are processed by pcre_exec() has been changed so that
1059        if they are repeated, backtracking one repetition now resets captured
1060        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
1061        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
1062        "aaa". Previously, it would have been "a". As part of this code
1063        refactoring, the way recursive calls are handled has also been changed.
1064    
1065    25. If an assertion condition captured any substrings, they were not passed
1066        back unless some other capturing happened later. For example, if
1067        (?(?=(a))a) was matched against "a", no capturing was returned.
1068    
1069    26. When studying a pattern that contained subroutine calls or assertions,
1070        the code for finding the minimum length of a possible match was handling
1071        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
1072        group 1 called group 2 while simultaneously a separate group 2 called group
1073        1). A stack overflow occurred in this case. I have fixed this by limiting
1074        the recursion depth to 10.
1075    
1076    27. Updated RunTest.bat in the distribution to the version supplied by Tom
1077        Fortmann. This supports explicit test numbers on the command line, and has
1078        argument validation and error reporting.
1079    
1080    28. An instance of \X with an unlimited repeat could fail if at any point the
1081        first character it looked at was a mark character.
1082    
1083    29. Some minor code refactoring concerning Unicode properties and scripts
1084        should reduce the stack requirement of match() slightly.
1085    
1086    30. Added the '=' option to pcretest to check the setting of unused capturing
1087        slots at the end of the pattern, which are documented as being -1, but are
1088        not included in the return count.
1089    
1090    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
1091        compiled something random. Now it gives a compile-time error (as does
1092        Perl).
1093    
1094    32. A *MARK encountered during the processing of a positive assertion is now
1095        recorded and passed back (compatible with Perl).
1096    
1097    33. If --only-matching or --colour was set on a pcregrep call whose pattern
1098        had alternative anchored branches, the search for a second match in a line
1099        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
1100        matched the line "0102" twice. The same bug affected patterns that started
1101        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
1102        twice.
1103    
1104    34. Previously, PCRE did not allow quantification of assertions. However, Perl
1105        does, and because of capturing effects, quantifying parenthesized
1106        assertions may at times be useful. Quantifiers are now allowed for
1107        parenthesized assertions.
1108    
1109    35. A minor code tidy in pcre_compile() when checking options for \R usage.
1110    
1111    36. \g was being checked for fancy things in a character class, when it should
1112        just be a literal "g".
1113    
1114    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
1115        appearance of a nested POSIX class supersedes an apparent external class.
1116        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
1117        unescaped square brackets may also appear as part of class names. For
1118        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
1119        more like Perl. (But see 8.20/1 above.)
1120    
1121    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
1122        was because it thought it was \N{name}, which is not supported).
1123    
1124    39. Add minix to OS list not supporting the -S option in pcretest.
1125    
1126    40. PCRE tries to detect cases of infinite recursion at compile time, but it
1127        cannot analyze patterns in sufficient detail to catch mutual recursions
1128        such as ((?1))((?2)). There is now a runtime test that gives an error if a
1129        subgroup is called recursively as a subpattern for a second time at the
1130        same position in the subject string. In previous releases this might have
1131        been caught by the recursion limit, or it might have run out of stack.
1132    
1133    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
1134        happen only once. PCRE was, however incorrectly giving a compile time error
1135        "recursive call could loop indefinitely" because it cannot analyze the
1136        pattern in sufficient detail. The compile time test no longer happens when
1137        PCRE is compiling a conditional subpattern, but actual runaway loops are
1138        now caught at runtime (see 40 above).
1139    
1140    42. It seems that Perl allows any characters other than a closing parenthesis
1141        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
1142        has been changed to be the same.
1143    
1144    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
1145        as not to get warnings when autogen.sh is called. Also changed
1146        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
1147    
1148    44. To help people who use pcregrep to scan files containing exceedingly long
1149        lines, the following changes have been made:
1150    
1151        (a) The default value of the buffer size parameter has been increased from
1152            8K to 20K. (The actual buffer used is three times this size.)
1153    
1154        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
1155            PCRE is built.
1156    
1157        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
1158            to be set at run time.
1159    
1160        (d) Numerical values in pcregrep options can be followed by K or M, for
1161            example --buffer-size=50K.
1162    
1163        (e) If a line being scanned overflows pcregrep's buffer, an error is now
1164            given and the return code is set to 2.
1165    
1166    45. Add a pointer to the latest mark to the callout data block.
1167    
1168    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
1169        partial match of an empty string instead of no match. This was specific to
1170        the use of ".".
1171    
1172    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
1173        complete match instead of a partial match. This bug was dependent on both
1174        the PCRE_UTF8 and PCRE_DOTALL options being set.
1175    
1176    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1177        starting byte set, because \b was not being ignored.
1178    
1179    
1180    Version 8.12 15-Jan-2011
1181    ------------------------
1182    
1183    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1184        checks for such things as part of the documentation building process.
1185    
1186    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1187        --match-limit and --recursion-limit options (added for 8.11). In
1188        particular, this made one of the standard tests fail. (The integer value
1189        went into the wrong half of a long int.)
1190    
1191    3.  If the --colour option was given to pcregrep with -v (invert match), it
1192        did strange things, either producing crazy output, or crashing. It should,
1193        of course, ignore a request for colour when reporting lines that do not
1194        match.
1195    
1196    4.  Another pcregrep bug caused similar problems if --colour was specified with
1197        -M (multiline) and the pattern match finished with a line ending.
1198    
1199    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1200        matched in multiline mode, the following line was shown as part of the
1201        match. This seems wrong, so I have changed it.
1202    
1203    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1204        the check for further matches in the same line (so they could be coloured)
1205        to overrun the end of the current line. If another match was found, it was
1206        incorrectly shown (and then shown again when found in the next line).
1207    
1208    7.  If pcregrep was compiled under Windows, there was a reference to the
1209        function pcregrep_exit() before it was defined. I am assuming this was
1210        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1211        reported by a user. I've moved the definition above the reference.
1212    
1213    
1214    Version 8.11 10-Dec-2010
1215    ------------------------
1216    
1217    1.  (*THEN) was not working properly if there were untried alternatives prior
1218        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1219        backtracked to try for "b" instead of moving to the next alternative branch
1220        at the same level (in this case, to look for "c"). The Perl documentation
1221        is clear that when (*THEN) is backtracked onto, it goes to the "next
1222        alternative in the innermost enclosing group".
1223    
1224    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1225        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1226        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1227        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1228        (*THEN).
1229    
1230    3.  If \s appeared in a character class, it removed the VT character from
1231        the class, even if it had been included by some previous item, for example
1232        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1233        of \s, but is part of the POSIX "space" class.)
1234    
1235    4.  A partial match never returns an empty string (because you can always
1236        match an empty string at the end of the subject); however the checking for
1237        an empty string was starting at the "start of match" point. This has been
1238        changed to the "earliest inspected character" point, because the returned
1239        data for a partial match starts at this character. This means that, for
1240        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1241        (previously it gave "no match").
1242    
1243    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1244        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1245        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1246        has an implication that the given string is incomplete (because a partial
1247        match is preferred over a full match). For this reason, these items now
1248        give a partial match in this situation. [Aside: previously, the one case
1249        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1250        match rather than a full match, which was wrong by the old rules, but is
1251        now correct.]
1252    
1253    6.  There was a bug in the handling of #-introduced comments, recognized when
1254        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1255        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1256        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1257        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1258        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1259        places in pcre_compile().
1260    
1261    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1262        comments when looking ahead for named forward references to subpatterns,
1263        the only newline sequence it recognized was NL. It now handles newlines
1264        according to the set newline convention.
1265    
1266    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1267        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1268        cater for a lack of strerror(). These oversights have been fixed.
1269    
1270    9.  Added --match-limit and --recursion-limit to pcregrep.
1271    
1272    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1273    
1274    11. When the -o option was used, pcregrep was setting a return code of 1, even
1275        when matches were found, and --line-buffered was not being honoured.
1276    
1277    12. Added an optional parentheses number to the -o and --only-matching options
1278        of pcregrep.
1279    
1280    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1281        can match an empty string. The code to do it in pcretest and pcredemo
1282        needed fixing:
1283    
1284        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1285            only one byte after an empty string match just before CRLF (this case
1286            just got forgotten; "any" and "anycrlf" were OK).
1287    
1288        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1289            mode when an empty string match preceded an ASCII character followed by
1290            a non-ASCII character. (The code for advancing by one character rather
1291            than one byte was nonsense.)
1292    
1293        (c) The pcredemo.c sample program did not have any code at all to handle
1294            the cases when CRLF is a valid newline sequence.
1295    
1296    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1297        as a starting offset was within the subject string. There is now a new
1298        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1299        negative or greater than the length of the string. In order to test this,
1300        pcretest is extended to allow the setting of negative starting offsets.
1301    
1302    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1303        starting offset points to the beginning of a UTF-8 character was
1304        unnecessarily clumsy. I tidied it up.
1305    
1306    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1307        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1308    
1309    17. Nobody had reported that the --include_dir option, which was added in
1310        release 7.7 should have been called --include-dir (hyphen, not underscore)
1311        for compatibility with GNU grep. I have changed it to --include-dir, but
1312        left --include_dir as an undocumented synonym, and the same for
1313        --exclude-dir, though that is not available in GNU grep, at least as of
1314        release 2.5.4.
1315    
1316    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1317        characters from a string of bytes) have been redefined so as not to use
1318        loops, in order to improve performance in some environments. At the same
1319        time, I abstracted some of the common code into auxiliary macros to save
1320        repetition (this should not affect the compiled code).
1321    
1322    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1323        compile-time error is now given if \c is not followed by an ASCII
1324        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1325        different, and any byte value is allowed.)
1326    
1327    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1328        START_OPTIMIZE option, which is now allowed at compile time - but just
1329        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1330        to pcregrep and other applications that have no direct access to PCRE
1331        options. The new /Y option in pcretest sets this option when calling
1332        pcre_compile().
1333    
1334    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1335        back references. Groups containing recursive back references were forced to
1336        be atomic by that change, but in the case of named groups, the amount of
1337        memory required was incorrectly computed, leading to "Failed: internal
1338        error: code overflow". This has been fixed.
1339    
1340    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1341        pcretest.c, to avoid build problems in some Borland environments.
1342    
1343    
1344    Version 8.10 25-Jun-2010
1345    ------------------------
1346    
1347    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1348        THEN.
1349    
1350    2.  (*ACCEPT) was not working when inside an atomic group.
1351    
1352    3.  Inside a character class, \B is treated as a literal by default, but
1353        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1354        causes the error). The code is unchanged, but I tidied the documentation.
1355    
1356    4.  Inside a character class, PCRE always treated \R and \X as literals,
1357        whereas Perl faults them if its -w option is set. I have changed PCRE so
1358        that it faults them when PCRE_EXTRA is set.
1359    
1360    5.  Added support for \N, which always matches any character other than
1361        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1362    
1363    6.  When compiling pcregrep with newer versions of gcc which may have
1364        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1365        declared with attribute warn_unused_result" were given. Just casting the
1366        result to (void) does not stop the warnings; a more elaborate fudge is
1367        needed. I've used a macro to implement this.
1368    
1369    7.  Minor change to pcretest.c to avoid a compiler warning.
1370    
1371    8.  Added four artifical Unicode properties to help with an option to make
1372        \s etc use properties (see next item). The new properties are: Xan
1373        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1374    
1375    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1376        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1377        this option. Modified pcretest to add /W to test this facility. Added
1378        REG_UCP to make it available via the POSIX interface.
1379    
1380    10. Added --line-buffered to pcregrep.
1381    
1382    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1383        studied, and the match started with a letter with a code point greater than
1384        127 whose first byte was different to the first byte of the other case of
1385        the letter, the other case of this starting letter was not recognized
1386        (#976).
1387    
1388    12. If a pattern that was studied started with a repeated Unicode property
1389        test, for example, \p{Nd}+, there was the theoretical possibility of
1390        setting up an incorrect bitmap of starting bytes, but fortunately it could
1391        not have actually happened in practice until change 8 above was made (it
1392        added property types that matched character-matching opcodes).
1393    
1394    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1395        possible starting bytes for non-anchored patterns.
1396    
1397    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1398        \R, and also a number of cases that involve Unicode properties, both
1399        explicit and implicit when PCRE_UCP is set.
1400    
1401    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1402        input, it could crash or give wrong results if characters with values
1403        greater than 0xc0 were present in the subject string. (Detail: it assumed
1404        UTF-8 input when processing these items.)
1405    
1406    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1407        size_t is 64-bit (#991).
1408    
1409    17. Added a check for running out of memory when PCRE is compiled with
1410        --disable-stack-for-recursion (#990).
1411    
1412    18. If the last data line in a file for pcretest does not have a newline on
1413        the end, a newline was missing in the output.
1414    
1415    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1416        less than 128) in its various bitmaps. However, there is a facility for
1417        generating tables according to the current locale when PCRE is compiled. It
1418        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1419        characters, are recognized by isspace() and therefore were getting set in
1420        these tables, and indeed these tables seem to approximate to ISO 8859. This
1421        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1422        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1423        which of course cannot start UTF-8 characters. I have changed the code so
1424        that only real ASCII characters (less than 128) and the correct starting
1425        bytes for UTF-8 encodings are set for characters greater than 127 when in
1426        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1427        altogether.)
1428    
1429    20. Added the /T option to pcretest so as to be able to run tests with non-
1430        standard character tables, thus making it possible to include the tests
1431        used for 19 above in the standard set of tests.
1432    
1433    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1434        reference to a subpattern the other side of a comment that contains an
1435        opening parenthesis caused either an internal compiling error, or a
1436        reference to the wrong subpattern.
1437    
1438    
1439    Version 8.02 19-Mar-2010
1440    ------------------------
1441    
1442    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1443    
1444    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1445        configured.
1446    
1447    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1448        original author of that file, following a query about its status.
1449    
1450    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1451        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1452    
1453    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1454        quantifier applied to a forward-referencing subroutine call, could compile
1455        incorrect code or give the error "internal error: previously-checked
1456        referenced subpattern not found".
1457    
1458    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1459        variables to point to external functions. For these systems, therefore,
1460        pcre_malloc etc. are now initialized to local functions that call the
1461        relevant global functions.
1462    
1463    7.  There were two entries missing in the vectors called coptable and poptable
1464        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1465        I've fixed the data, and added a kludgy way of testing at compile time that
1466        the lengths are correct (equal to the number of opcodes).
1467    
1468    8.  Following on from 7, I added a similar kludge to check the length of the
1469        eint vector in pcreposix.c.
1470    
1471    9.  Error texts for pcre_compile() are held as one long string to avoid too
1472        much relocation at load time. To find a text, the string is searched,
1473        counting zeros. There was no check for running off the end of the string,
1474        which could happen if a new error number was added without updating the
1475        string.
1476    
1477    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1478    
1479    11. \K was not working if it appeared in an atomic group or in a group that
1480        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1481        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1482        the assertion is positive, but not if it is negative.
1483    
1484    12. Change 11 fortuitously reduced the size of the stack frame used in the
1485        "match()" function of pcre_exec.c by one pointer. Forthcoming
1486        implementation of support for (*MARK) will need an extra pointer on the
1487        stack; I have reserved it now, so that the stack frame size does not
1488        decrease.
1489    
1490    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1491        item in branch that calls a recursion is a subroutine call - as in the
1492        second branch in the above example - was incorrectly given the compile-
1493        time error "recursive call could loop indefinitely" because pcre_compile()
1494        was not correctly checking the subroutine for matching a non-empty string.
1495    
1496    14. The checks for overrunning compiling workspace could trigger after an
1497        overrun had occurred. This is a "should never occur" error, but it can be
1498        triggered by pathological patterns such as hundreds of nested parentheses.
1499        The checks now trigger 100 bytes before the end of the workspace.
1500    
1501    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1502    
1503    
1504    Version 8.01 19-Jan-2010
1505    ------------------------
1506    
1507    1.  If a pattern contained a conditional subpattern with only one branch (in
1508        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1509        computed the wrong minimum data length (which is of course zero for such
1510        subpatterns). This could cause incorrect "no match" results.
1511    
1512    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1513        the pattern is reset in the first branch, pcre_compile() failed with
1514        "internal error: code overflow at offset...". This happened only when
1515        the reset was to the original external option setting. (An optimization
1516        abstracts leading options settings into an external setting, which was the
1517        cause of this.)
1518    
1519    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1520        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1521        assertion pattern did not match (meaning that the assertion was true), it
1522        was incorrectly treated as false if the SKIP had been reached during the
1523        matching. This also applied to assertions used as conditions.
1524    
1525    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1526        assertion subpattern, including such a pattern used as a condition,
1527        unpredictable results occurred, instead of the error return
1528        PCRE_ERROR_DFA_UITEM.
1529    
1530    5.  The C++ GlobalReplace function was not working like Perl for the special
1531        situation when an empty string is matched. It now does the fancy magic
1532        stuff that is necessary.
1533    
1534    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1535        removed. (These were left over from very, very early versions of PCRE.)
1536    
1537    7.  Some cosmetic changes to the code to make life easier when compiling it
1538        as part of something else:
1539    
1540        (a) Change DEBUG to PCRE_DEBUG.
1541    
1542        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1543            called "current" as "current_branch", to prevent a collision with the
1544            Linux macro when compiled as a kernel module.
1545    
1546        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1547            prevent a collision with the Linux macro when compiled as a kernel
1548            module.
1549    
1550    8.  In pcre_compile() there are some checks for integer overflows that used to
1551        cast potentially large values to (double). This has been changed to that
1552        when building, a check for int64_t is made, and if it is found, it is used
1553        instead, thus avoiding the use of floating point arithmetic. (There is no
1554        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1555        double.
1556    
1557    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1558        2005 (difference between two addresses compared to an unsigned value).
1559    
1560    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1561        custom one, because of the following reported problem in Windows:
1562    
1563          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1564              under Win32.
1565          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1566              therefore missing the function definition.
1567          - The compiler thus generates a "C" signature for the test function.
1568          - The linker fails to find the "C" function.
1569          - PCRE fails to configure if asked to do so against libbz2.
1570    
1571    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1572        messages were output:
1573    
1574          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1575          rerunning libtoolize, to keep the correct libtool macros in-tree.
1576          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1577    
1578        I have done both of these things.
1579    
1580    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1581        most of the time, it *can* run out if it is given a pattern that contains a
1582        runaway infinite recursion. I updated the discussion in the pcrestack man
1583        page.
1584    
1585    13. Now that we have gone to the x.xx style of version numbers, the minor
1586        version may start with zero. Using 08 or 09 is a bad idea because users
1587        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1588        interpreted as invalid octal numbers. I've updated the previous comment in
1589        configure.ac, and also added a check that gives an error if 08 or 09 are
1590        used.
1591    
1592    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1593        causing partial matching to fail when the end of the subject matched \W
1594        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1595    
1596    15. There were some discrepancies between the declarations in pcre_internal.h
1597        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1598        their definitions. The declarations used "const uschar *" and the
1599        definitions used USPTR. Even though USPTR is normally defined as "const
1600        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1601        reported that: "This difference in casting confuses some C++ compilers, for
1602        example, SunCC recognizes above declarations as different functions and
1603        generates broken code for hbpcre." I have changed the declarations to use
1604        USPTR.
1605    
1606    16. GNU libtool is named differently on some systems. The autogen.sh script now
1607        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1608        (FreeBSD).
1609    
1610    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1611        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1612        comment: "Figure out how to create a longlong from a string: strtoll and
1613        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1614        instance, but it only takes 2 args instead of 3!"
1615    
1616    18. A subtle bug concerned with back references has been fixed by a change of
1617        specification, with a corresponding code fix. A pattern such as
1618        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1619        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1620        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1621        same bug. Such groups have to be quantified to be useful, or contained
1622        inside another quantified group. (If there's no repetition, the reference
1623        can never match.) The problem arises because, having left the group and
1624        moved on to the rest of the pattern, a later failure that backtracks into
1625        the group uses the captured value from the final iteration of the group
1626        rather than the correct earlier one. I have fixed this in PCRE by forcing
1627        any group that contains a reference to itself to be an atomic group; that
1628        is, there cannot be any backtracking into it once it has completed. This is
1629        similar to recursive and subroutine calls.
1630    
1631    
1632    Version 8.00 19-Oct-09
1633    ----------------------
1634    
1635    1.  The table for translating pcre_compile() error codes into POSIX error codes
1636        was out-of-date, and there was no check on the pcre_compile() error code
1637        being within the table. This could lead to an OK return being given in
1638        error.
1639    
1640    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1641        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1642        in a Windows environment.
1643    
1644    3.  The pcregrep --count option prints the count for each file even when it is
1645        zero, as does GNU grep. However, pcregrep was also printing all files when
1646        --files-with-matches was added. Now, when both options are given, it prints
1647        counts only for those files that have at least one match. (GNU grep just
1648        prints the file name in this circumstance, but including the count seems
1649        more useful - otherwise, why use --count?) Also ensured that the
1650        combination -clh just lists non-zero counts, with no names.
1651    
1652    4.  The long form of the pcregrep -F option was incorrectly implemented as
1653        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1654        but it seems right to fix it, and I didn't think it was worth preserving
1655        the old behaviour.
1656    
1657    5.  The command line items --regex=pattern and --regexp=pattern were not
1658        recognized by pcregrep, which required --regex pattern or --regexp pattern
1659        (with a space rather than an '='). The man page documented the '=' forms,
1660        which are compatible with GNU grep; these now work.
1661    
1662    6.  No libpcreposix.pc file was created for pkg-config; there was just
1663        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1664    
1665    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1666        when UCP support is not needed, by modifying the Python script that
1667        generates it from Unicode data files. This should not matter if the module
1668        is correctly used as a library, but I received one complaint about 50K of
1669        unwanted data. My guess is that the person linked everything into his
1670        program rather than using a library. Anyway, it does no harm.
1671    
1672    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1673        was a minimum greater than 1 for a wide character in a possessive
1674        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1675        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1676        character. Chaos in the form of incorrect output or a compiling loop could
1677        result.
1678    
1679    9.  The restrictions on what a pattern can contain when partial matching is
1680        requested for pcre_exec() have been removed. All patterns can now be
1681        partially matched by this function. In addition, if there are at least two
1682        slots in the offset vector, the offset of the earliest inspected character
1683        for the match and the offset of the end of the subject are set in them when
1684        PCRE_ERROR_PARTIAL is returned.
1685    
1686    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1687        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1688        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1689        and may be more useful for multi-segment matching.
1690    
1691    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1692        used to be given if ever the end of the subject was reached; now it is
1693        given only if matching could not proceed because another character was
1694        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1695        string "Z", which now yields "no match" instead of "partial match". In the
1696        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1697        final character ended with (*FAIL).
1698    
1699    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1700        if the pattern had a "must contain" character that was already found in the
1701        earlier partial match, unless partial matching was again requested. For
1702        example, with the pattern /dog.(body)?/, the "must contain" character is
1703        "g". If the first part-match was for the string "dog", restarting with
1704        "sbody" failed. This bug has been fixed.
1705    
1706    13. The string returned by pcre_dfa_exec() after a partial match has been
1707        changed so that it starts at the first inspected character rather than the
1708        first character of the match. This makes a difference only if the pattern
1709        starts with a lookbehind assertion or \b or \B (\K is not supported by
1710        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1711        matching functions compatible, and I think it's the right thing to do.
1712    
1713    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1714        so that the demonstration program is easily available in environments where
1715        PCRE has not been installed from source.
1716    
1717    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1718        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1719        library.
1720    
1721    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1722        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1723        is not the first non-POSIX option to be added. Clearly some people find
1724        these options useful.
1725    
1726    17. If a caller to the POSIX matching function regexec() passes a non-zero
1727        value for nmatch with a NULL value for pmatch, the value of
1728        nmatch is forced to zero.
1729    
1730    18. RunGrepTest did not have a test for the availability of the -u option of
1731        the diff command, as RunTest does. It now checks in the same way as
1732        RunTest, and also checks for the -b option.
1733    
1734    19. If an odd number of negated classes containing just a single character
1735        interposed, within parentheses, between a forward reference to a named
1736        subpattern and the definition of the subpattern, compilation crashed with
1737        an internal error, complaining that it could not find the referenced
1738        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1739        [The bug was that it was starting one character too far in when skipping
1740        over the character class, thus treating the ] as data rather than
1741        terminating the class. This meant it could skip too much.]
1742    
1743    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1744        /g option in pcretest when the pattern contains \K, which makes it possible
1745        to have an empty string match not at the start, even when the pattern is
1746        anchored. Updated pcretest and pcredemo to use this option.
1747    
1748    21. If the maximum number of capturing subpatterns in a recursion was greater
1749        than the maximum at the outer level, the higher number was returned, but
1750        with unset values at the outer level. The correct (outer level) value is
1751        now given.
1752    
1753    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1754        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1755        make it do so. The string so far is captured, making this feature
1756        compatible with Perl.
1757    
1758    23. The tests have been re-organized, adding tests 11 and 12, to make it
1759        possible to check the Perl 5.10 features against Perl 5.10.
1760    
1761    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1762        pattern matches a fixed length string. PCRE did not allow this; now it
1763        does. Neither allows recursion.
1764    
1765    25. I finally figured out how to implement a request to provide the minimum
1766        length of subject string that was needed in order to match a given pattern.
1767        (It was back references and recursion that I had previously got hung up
1768        on.) This code has now been added to pcre_study(); it finds a lower bound
1769        to the length of subject needed. It is not necessarily the greatest lower
1770        bound, but using it to avoid searching strings that are too short does give
1771        some useful speed-ups. The value is available to calling programs via
1772        pcre_fullinfo().
1773    
1774    26. While implementing 25, I discovered to my embarrassment that pcretest had
1775        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1776        study optimizations had never been tested with that matching function.
1777        Oops. What is worse, even when it was passed study data, there was a bug in
1778        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1779        were also very few tests of studied patterns with pcre_dfa_exec().
1780    
1781    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1782        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1783        on the other side of the coin, they are no longer allowed to have different
1784        names, because these cannot be distinguished in PCRE, and this has caused
1785        confusion. (This is a difference from Perl.)
1786    
1787    28. When duplicate subpattern names are present (necessarily with different
1788        numbers, as required by 27 above), and a test is made by name in a
1789        conditional pattern, either for a subpattern having been matched, or for
1790        recursion in such a pattern, all the associated numbered subpatterns are
1791        tested, and the overall condition is true if the condition is true for any
1792        one of them. This is the way Perl works, and is also more like the way
1793        testing by number works.
1794    
1795    
1796    Version 7.9 11-Apr-09
1797    ---------------------
1798    
1799    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1800        (pcretest), all targets were linked against these libraries. This included
1801        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1802        libraries. This caused unwanted dependencies to be created. This problem
1803        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1804        pcretest is linked with readline.
1805    
1806    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1807        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1808        moved outside it again, because FALSE and TRUE are already defined in AIX,
1809        but BOOL is not.
1810    
1811    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1812        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1813    
1814    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1815        hyphens) following file names and line numbers when outputting matching
1816        lines. This is not true; no spaces are inserted. I have also clarified the
1817        wording for the --colour (or --color) option.
1818    
1819    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1820        was not coloured; this is different to GNU grep, so I have changed it to be
1821        the same.
1822    
1823    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1824        each matching line was coloured. Now it goes on to look for further matches
1825        of any of the test patterns, which is the same behaviour as GNU grep.
1826    
1827    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1828        doesn't make sense to accept an empty string match in pcregrep, so I have
1829        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1830        seems to be how GNU grep behaves. [But see later change 40 for release
1831        8.33.]
1832    
1833    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1834        start or after a newline", because the conditional assertion was not being
1835        correctly handled. The rule now is that both the assertion and what follows
1836        in the first alternative must satisfy the test.
1837    
1838    9.  If auto-callout was enabled in a pattern with a conditional group whose
1839        condition was an assertion, PCRE could crash during matching, both with
1840        pcre_exec() and pcre_dfa_exec().
1841    
1842    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1843        used for matching.
1844    
1845    11. Unicode property support in character classes was not working for
1846        characters (bytes) greater than 127 when not in UTF-8 mode.
1847    
1848    12. Added the -M command line option to pcretest.
1849    
1850    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1851    
1852    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1853    
1854    16. Added comments and documentation about mis-use of no_arg in the C++
1855        wrapper.
1856    
1857    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1858        from Martin Jerabek that uses macro names for all relevant character and
1859        string constants.
1860    
1861    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1862        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1863        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1864        these, but not everybody uses configure.
1865    
1866    19. A conditional group that had only one branch was not being correctly
1867        recognized as an item that could match an empty string. This meant that an
1868        enclosing group might also not be so recognized, causing infinite looping
1869        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1870        with the subject "ab", where knowledge that the repeated group can match
1871        nothing is needed in order to break the loop.
1872    
1873    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1874        exec(), but without supplying a callout function, matching went wrong.
1875    
1876    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1877        leak if the size of the offset vector was greater than 30. When the vector
1878        is smaller, the saved offsets during recursion go onto a local stack
1879        vector, but for larger vectors malloc() is used. It was failing to free
1880        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1881        error, in fact).
1882    
1883    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1884        heapframe that is used only when UTF-8 support is enabled. This caused no
1885        problem, but was untidy.
1886    
1887    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1888        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1889        included within another project.
1890    
1891    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1892        slightly modified by me:
1893    
1894          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1895              not building pcregrep.
1896    
1897          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1898              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1899    
1900    25. Forward references, both numeric and by name, in patterns that made use of
1901        duplicate group numbers, could behave incorrectly or give incorrect errors,
1902        because when scanning forward to find the reference group, PCRE was not
1903        taking into account the duplicate group numbers. A pattern such as
1904        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1905    
1906    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1907        the feature of a custom pointer more persuasive (as requested by a user).
1908    
1909    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1910        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1911        pre-defined.
1912    
1913    28. Added support for (*UTF8) at the start of a pattern.
1914    
1915    29. Arrange for flags added by the "release type" setting in CMake to be shown
1916        in the configuration summary.
1917    
1918    
1919    Version 7.8 05-Sep-08
1920    ---------------------
1921    
1922    1.  Replaced UCP searching code with optimized version as implemented for Ad
1923        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1924        stage table and inline lookup instead of a function, giving speed ups of 2
1925        to 5 times on some simple patterns that I tested. Permission was given to
1926        distribute the MultiStage2.py script that generates the tables (it's not in
1927        the tarball, but is in the Subversion repository).
1928    
1929    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1930        scripts.
1931    
1932    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1933        a group with a zero qualifier. The result of the study could be incorrect,
1934        or the function might crash, depending on the pattern.
1935    
1936    4.  Caseless matching was not working for non-ASCII characters in back
1937        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1938        It now works when Unicode Property Support is available.
1939    
1940    5.  In pcretest, an escape such as \x{de} in the data was always generating
1941        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1942        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1943        truncation.
1944    
1945    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1946    
1947    7.  Added two (int) casts to pcregrep when printing the difference of two
1948        pointers, in case they are 64-bit values.
1949    
1950    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1951        test 2 if it fails.
1952    
1953    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1954        and a #define of that name to empty if it is not externally set. This is to
1955        allow users of MSVC to set it if necessary.
1956    
1957    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1958        the convenience functions in the pcre_get.c source file.
1959    
1960    11. An option change at the start of a pattern that had top-level alternatives
1961        could cause overwriting and/or a crash. This command provoked a crash in
1962        some environments:
1963    
1964          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1965    
1966        This potential security problem was recorded as CVE-2008-2371.
1967    
1968    12. For a pattern where the match had to start at the beginning or immediately
1969        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1970        pcre_dfa_exec() could read past the end of the passed subject if there was
1971        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1972        pcretest so that it places the subject at the end of its malloc-ed buffer.
1973    
1974    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1975        exec() might read past the end of the data buffer in UTF-8 mode.
1976    
1977    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1978        the data contained the byte 0x85 as part of a UTF-8 character within its
1979        first line. This applied both to normal and DFA matching.
1980    
1981    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1982        /^[^d]*?$/8 failed to match "abc".
1983    
1984    16. Added a missing copyright notice to pcrecpp_internal.h.
1985    
1986    17. Make it more clear in the documentation that values returned from
1987        pcre_exec() in ovector are byte offsets, not character counts.
1988    
1989    18. Tidied a few places to stop certain compilers from issuing warnings.
1990    
1991    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1992        supplied by Stefan Weber. I made a further small update for 7.8 because
1993        there is a change of source arrangements: the pcre_searchfuncs.c module is
1994        replaced by pcre_ucd.c.
1995    
1996    
1997    Version 7.7 07-May-08
1998    ---------------------
1999    
2000    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
2001        a string to a long long, pretend we don't even have a long long." This is
2002        done by checking for the strtoq, strtoll, and _strtoi64 functions.
2003    
2004    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
2005        pre-7.6 versions, which defined a global no_arg variable instead of putting
2006        it in the RE class. (See also #8 below.)
2007    
2008    3.  Remove a line of dead code, identified by coverity and reported by Nuno
2009        Lopes.
2010    
2011    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
2012    
2013        (1) The include/exclude patterns were being applied to the whole pathnames
2014            of files, instead of just to the final components.
2015    
2016        (2) If there was more than one level of directory, the subdirectories were
2017            skipped unless they satisfied the include/exclude conditions. This is
2018            inconsistent with GNU grep (and could even be seen as contrary to the
2019            pcregrep specification - which I improved to make it absolutely clear).
2020            The action now is always to scan all levels of directory, and just
2021            apply the include/exclude patterns to regular files.
2022    
2023    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
2024        --exclude_dir in the tests to avoid scanning .svn directories.
2025    
2026    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
2027        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
2028        doesn't support NULs in patterns.
2029    
2030    7.  Added some missing "const"s to declarations of static tables in
2031        pcre_compile.c and pcre_dfa_exec.c.
2032    
2033    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
2034        caused by fix #2  above. (Subsequently also a second patch to fix the
2035        first patch. And a third patch - this was a messy problem.)
2036    
2037    9.  Applied Craig's patch to remove the use of push_back().
2038    
2039    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
2040        matching function regexec().
2041    
2042    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
2043        which, however, unlike Perl's \g{...}, are subroutine calls, not back
2044        references. PCRE supports relative numbers with this syntax (I don't think
2045        Oniguruma does).
2046    
2047    12. Previously, a group with a zero repeat such as (...){0} was completely
2048        omitted from the compiled regex. However, this means that if the group
2049        was called as a subroutine from elsewhere in the pattern, things went wrong
2050        (an internal error was given). Such groups are now left in the compiled
2051        pattern, with a new opcode that causes them to be skipped at execution
2052        time.
2053    
2054    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
2055        to the way PCRE behaves:
2056    
2057        (a) A lone ] character is dis-allowed (Perl treats it as data).
2058    
2059        (b) A back reference to an unmatched subpattern matches an empty string
2060            (Perl fails the current match path).
2061    
2062        (c) A data ] in a character class must be notated as \] because if the
2063            first data character in a class is ], it defines an empty class. (In
2064            Perl it is not possible to have an empty class.) The empty class []
2065            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
2066            The negative empty class [^] matches any one character, independently
2067            of the DOTALL setting.
2068    
2069    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
2070        non-existent subpattern following a character class starting with ']' and
2071        containing () gave an internal compiling error instead of "reference to
2072        non-existent subpattern". Fortunately, when the pattern did exist, the
2073        compiled code was correct. (When scanning forwards to check for the
2074        existence of the subpattern, it was treating the data ']' as terminating
2075        the class, so got the count wrong. When actually compiling, the reference
2076        was subsequently set up correctly.)
2077    
2078    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
2079        it was being rejected as not supported by pcre_dfa_exec(), even though
2080        other assertions are supported. I have made pcre_dfa_exec() support
2081        (*FAIL).
2082    
2083    16. The implementation of 13c above involved the invention of a new opcode,
2084        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
2085        cannot be changed at match time, I realized I could make a small
2086        improvement to matching performance by compiling OP_ALLANY instead of
2087        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
2088        on the OP_ANY path.
2089    
2090    17. Compiling pcretest on Windows with readline support failed without the
2091        following two fixes: (1) Make the unistd.h include conditional on
2092        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
2093    
2094    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
2095        ncurses library to be included for pcretest when ReadLine support is
2096        requested, but also to allow for it to be overridden. This patch came from
2097        Daniel Bergström.
2098    
2099    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
2100        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
2101        any errors with the current Unicode tables. Thanks to Peter Kankowski for
2102        spotting this.
2103    
2104    
2105    Version 7.6 28-Jan-08
2106    ---------------------
2107    
2108    1.  A character class containing a very large number of characters with
2109        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
2110        overflow.
2111    
2112    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
2113        HAVE_LONG_LONG is not defined.
2114    
2115    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
2116        bring them up to date and include new features. This patch includes:
2117    
2118        - Fixed PH's badly added libz and libbz2 support.
2119        - Fixed a problem with static linking.
2120        - Added pcredemo. [But later removed - see 7 below.]
2121        - Fixed dftables problem and added an option.
2122        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
2123            HAVE_LONG_LONG.
2124        - Added readline support for pcretest.
2125        - Added an listing of the option settings after cmake has run.
2126    
2127    4.  A user submitted a patch to Makefile that makes it easy to create
2128        "pcre.dll" under mingw when using Configure/Make. I added stuff to
2129        Makefile.am that cause it to include this special target, without
2130        affecting anything else. Note that the same mingw target plus all
2131        the other distribution libraries and programs are now supported
2132        when configuring with CMake (see 6 below) instead of with
2133        Configure/Make.
2134    
2135    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
2136        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
2137        exported in the Windows port". It has not yet been confirmed that the patch
2138        solves the problem, but it does no harm.
2139    
2140    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
2141        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
2142        with CMake, and also correct the comment about stack recursion.
2143    
2144    7.  Remove the automatic building of pcredemo from the ./configure system and
2145        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
2146        of a program that users should build themselves after PCRE is installed, so
2147        building it automatically is not really right. What is more, it gave
2148        trouble in some build environments.
2149    
2150    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
2151    
2152    
2153    Version 7.5 10-Jan-08
2154    ---------------------
2155    
2156    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
2157        values in parens when parsing an RE using the C++ wrapper."
2158    
2159    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
2160        Characters greater than 255 were excluded from the class instead of being
2161        included.
2162    
2163    3.  The same bug as (2) above applied to negated POSIX classes such as
2164        [:^space:].
2165    
2166    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
2167        defined or documented. It seems to have been a typo for PCRE_STATIC, so
2168        I have changed it.
2169    
2170    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
2171        first named subpattern) and a construct such as (?&a) would reference the
2172        first named subpattern whose name started with "a" (in other words, the
2173        length check was missing). Both these problems are fixed. "Subpattern name
2174        expected" is now given for (?&) (a zero-length name), and this patch also
2175        makes it give the same error for \k'' (previously it complained that that
2176        was a reference to a non-existent subpattern).
2177    
2178    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
2179        this is right because (?- can be followed by option settings as well as by
2180        digits. I have, however, made the messages clearer.
2181    
2182    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
2183        than the number used in the conditional) now cause a compile-time error.
2184        This is actually not compatible with Perl, which accepts such patterns, but
2185        treats the conditional as always being FALSE (as PCRE used to), but it
2186        seems to me that giving a diagnostic is better.
2187    
2188    8.  Change "alphameric" to the more common word "alphanumeric" in comments
2189        and messages.
2190    
2191    9.  Fix two occurrences of "backslash" in comments that should have been
2192        "backspace".
2193    
2194    10. Remove two redundant lines of code that can never be obeyed (their function
2195        was moved elsewhere).
2196    
2197    11. The program that makes PCRE's Unicode character property table had a bug
2198        which caused it to generate incorrect table entries for sequences of
2199        characters that have the same character type, but are in different scripts.
2200        It amalgamated them into a single range, with the script of the first of
2201        them. In other words, some characters were in the wrong script. There were
2202        thirteen such cases, affecting characters in the following ranges:
2203    
2204          U+002b0 - U+002c1
2205          U+0060c - U+0060d
2206          U+0061e - U+00612
2207          U+0064b - U+0065e
2208          U+0074d - U+0076d
2209          U+01800 - U+01805
2210          U+01d00 - U+01d77
2211          U+01d9b - U+01dbf
2212          U+0200b - U+0200f
2213          U+030fc - U+030fe
2214          U+03260 - U+0327f
2215          U+0fb46 - U+0fbb1
2216          U+10450 - U+1049d
2217    
2218    12. The -o option (show only the matching part of a line) for pcregrep was not
2219        compatible with GNU grep in that, if there was more than one match in a
2220        line, it showed only the first of them. It now behaves in the same way as
2221        GNU grep.
2222    
2223    13. If the -o and -v options were combined for pcregrep, it printed a blank
2224        line for every non-matching line. GNU grep prints nothing, and pcregrep now
2225        does the same. The return code can be used to tell if there were any
2226        non-matching lines.
2227    
2228    14. Added --file-offsets and --line-offsets to pcregrep.
2229    
2230    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
2231        infinitely looping recursion. The bug was that positive lookaheads were not
2232        being skipped when checking for a possible empty match (negative lookaheads
2233        and both kinds of lookbehind were skipped).
2234    
2235    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
2236        inclusion of <windows.h> to before rather than after the definition of
2237        INVALID_FILE_ATTRIBUTES (patch from David Byron).
2238    
2239    17. Specifying a possessive quantifier with a specific limit for a Unicode
2240        character property caused pcre_compile() to compile bad code, which led at
2241        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
2242        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
2243        caused the error; without that there was no problem.
2244    
2245    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
2246    
2247    19. Added --enable-pcretest-libreadline.
2248    
2249    20. In pcrecpp.cc, the variable 'count' was incremented twice in
2250        RE::GlobalReplace(). As a result, the number of replacements returned was
2251        double what it should be. I removed one of the increments, but Craig sent a
2252        later patch that removed the other one (the right fix) and added unit tests
2253        that check the return values (which was not done before).
2254    
2255    21. Several CMake things:
2256    
2257        (1) Arranged that, when cmake is used on Unix, the libraries end up with
2258            the names libpcre and libpcreposix, not just pcre and pcreposix.
2259    
2260        (2) The above change means that pcretest and pcregrep are now correctly
2261            linked with the newly-built libraries, not previously installed ones.
2262    
2263        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
2264    
2265    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
2266        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
2267        UTF-8 newline character). The key issue is that the pattern starts .*;
2268        this means that the match must be either at the beginning, or after a
2269        newline. The bug was in the code for advancing after a failed match and
2270        checking that the new position followed a newline. It was not taking
2271        account of UTF-8 characters correctly.
2272    
2273    23. PCRE was behaving differently from Perl in the way it recognized POSIX
2274        character classes. PCRE was not treating the sequence [:...:] as a
2275        character class unless the ... were all letters. Perl, however, seems to
2276        allow any characters between [: and :], though of course it rejects as
2277        unknown any "names" that contain non-letters, because all the known class
2278        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
2279        for example, whereas PCRE did not - it did not recognize a POSIX character
2280        class. This seemed a bit dangerous, so the code has been changed to be
2281        closer to Perl. The behaviour is not identical to Perl, because PCRE will
2282        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
2283        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
2284        Perl does, and where it didn't before.
2285    
2286    24. Rewrite so as to remove the single use of %n from pcregrep because in some
2287        Windows environments %n is disabled by default.
2288    
2289    
2290    Version 7.4 21-Sep-07
2291    ---------------------
2292    
2293    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
2294        means that a class such as [\s] counted as "explicit reference to CR or
2295        LF". That isn't really right - the whole point of the change was to try to
2296        help when there was an actual mention of one of the two characters. So now
2297        the change happens only if \r or \n (or a literal CR or LF) character is
2298        encountered.
2299    
2300    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
2301        of both had grown to the point where there were only 3 bits left.
2302        Fortunately, there was spare space in the data structure, and so I have
2303        moved the internal flags into a new 16-bit field to free up more option
2304        bits.
2305    
2306    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
2307        but did not set the internal JCHANGED flag - either of these is enough to
2308        control the way the "get" function works - but the PCRE_INFO_JCHANGED
2309        facility is supposed to tell if (?J) was ever used, so now (?J) at the
2310        start sets both bits.
2311    
2312    4.  Added options (at build time, compile time, exec time) to change \R from
2313        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
2314    
2315    5.  doc/pcresyntax.html was missing from the distribution.
2316    
2317    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
2318        compatibility, even though it is no longer used.
2319    
2320    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
2321        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
2322        windows.h file is present (where different names are used). [This was
2323        reversed later after testing - see 16 below.]
2324    
2325    8.  Changed all #include <config.h> to #include "config.h". There were also
2326        some further <pcre.h> cases that I changed to "pcre.h".
2327    
2328    9.  When pcregrep was used with the --colour option, it missed the line ending
2329        sequence off the lines that it output.
2330    
2331    10. It was pointed out to me that arrays of string pointers cause lots of
2332        relocations when a shared library is dynamically loaded. A technique of
2333        using a single long string with a table of offsets can drastically reduce
2334        these. I have refactored PCRE in four places to do this. The result is
2335        dramatic:
2336    
2337          Originally:                          290
2338          After changing UCP table:            187
2339          After changing error message table:   43
2340          After changing table of "verbs"       36
2341          After changing table of Posix names   22
2342    
2343        Thanks to the folks working on Gregex for glib for this insight.
2344    
2345    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
2346        unicode-properties was also set.
2347    
2348    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
2349    
2350    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
2351        checked only for CRLF.
2352    
2353    14. Added casts to pcretest.c to avoid compiler warnings.
2354    
2355    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
2356    
2357    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
2358        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2359        entirely. This removes changes made in 7 above.
2360    
2361    17. The CMake files have been updated, and there is now more information about
2362        building with CMake in the NON-UNIX-USE document.
2363    
2364    
2365    Version 7.3 28-Aug-07
2366  ---------------------  ---------------------
2367    
2368   1. In the rejigging of the build system that eventually resulted in 7.1, the   1. In the rejigging of the build system that eventually resulted in 7.1, the
# Line 143  Version 7.3 17-Aug-07 Line 2504  Version 7.3 17-Aug-07
2504  24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte  24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2505      characters were involved (for example /\S{2}/8g with "A\x{a3}BC").      characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2506    
2507    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2508    
2509    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2510        character were causing crashes (broken optimization).
2511    
2512    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2513        \p or \P) caused a compile-time loop.
2514    
2515    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2516        break. For example, the unstudied pattern [\r\n]A does not match the string
2517        "\r\nA" because change 7.0/46 below moves the current point on by two
2518        characters after failing to match at the start. However, the pattern \nA
2519        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2520        the same is true. There doesn't seem any very clean way out of this, but
2521        what I have chosen to do makes the common cases work: PCRE now takes note
2522        of whether there can be an explicit match for \r or \n anywhere in the
2523        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2524        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2525        pattern has explicit CR or LF references.
2526    
2527    29. Added (*CR) etc for changing newline setting at start of pattern.
2528    
2529    
2530  Version 7.2 19-Jun-07  Version 7.2 19-Jun-07
2531  ---------------------  ---------------------

Legend:
Removed from v.221  
changed lines
  Added in v.1387

  ViewVC Help
Powered by ViewVC 1.1.5