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

Diff of /code/trunk/ChangeLog

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

revision 118 by ph10, Fri Mar 9 16:17:46 2007 UTC revision 1394 by ph10, Sat Nov 9 09:17:20 2013 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 7.1 05-Mar-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    32. Added the "forbid" facility to pcretest so that putting tests into the
156        wrong test files can sometimes be quickly detected.
157    
158    33. There is now a limit (default 250) on the depth of nesting of parentheses.
159        This limit is imposed to control the amount of system stack used at compile
160        time. It can be changed at build time by --with-parens-nest-limit=xxx or
161        the equivalent in CMake.
162    
163    34. Character classes such as [A-\d] or [a-[:digit:]] now cause compile-time
164        errors. Perl warns for these when in warning mode, but PCRE has no facility
165        for giving warnings.
166    
167    35. Change 34 for 8.13 allowed quantifiers on assertions, because Perl does.
168        However, this was not working for (?!) because it is optimized to (*FAIL),
169        for which PCRE does not allow quantifiers. The optimization is now disabled
170        when a quantifier follows (?!). I can't see any use for this, but it makes
171        things uniform.
172    
173    36. Perl no longer allows group names to start with digits, so I have made this
174        change also in PCRE. It simplifies the code a bit.
175    
176    
177    Version 8.33 28-May-2013
178    ------------------------
179    
180    1.  Added 'U' to some constants that are compared to unsigned integers, to
181        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
182        variables that are added to signed variables, to ensure the result is
183        signed and can be negated.
184    
185    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
186        CMake config files.
187    
188    3.  Revise the creation of config.h.generic so that all boolean macros are
189        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
190        overriding via -D on the command line possible.
191    
192    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
193        to unsigned int is reported to make a quite noticeable speed difference in
194        a specific Windows environment. Testing on Linux did also appear to show
195        some benefit (and it is clearly not harmful). Also fixed the definition of
196        Xop which should be unsigned.
197    
198    5.  Related to (4), changing the definition of the intermediate variable cc
199        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
200        improvements.
201    
202    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
203        unnecessary spaces.
204    
205    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
206        and later.
207    
208    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
209    
210    9.  Optimizing fast_forward_start_bits in JIT.
211    
212    10. Adding support for callouts in JIT, and fixing some issues revealed
213        during this work. Namely:
214    
215        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
216    
217        (b) Minimum length was not checked before the matching is started.
218    
219    11. The value of capture_last that is passed to callouts was incorrect in some
220        cases when there was a capture on one path that was subsequently abandoned
221        after a backtrack. Also, the capture_last value is now reset after a
222        recursion, since all captures are also reset in this case.
223    
224    12. The interpreter no longer returns the "too many substrings" error in the
225        case when an overflowing capture is in a branch that is subsequently
226        abandoned after a backtrack.
227    
228    13. In the pathological case when an offset vector of size 2 is used, pcretest
229        now prints out the matched string after a yield of 0 or 1.
230    
231    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
232        Only supported by the JIT compiler at the moment.
233    
234    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
235    
236    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
237        offset of the starting point of the matching process, provided the offsets
238        vector is large enough.
239    
240    17. The \A escape now records a lookbehind value of 1, though its execution
241        does not actually inspect the previous character. This is to ensure that,
242        in partial multi-segment matching, at least one character from the old
243        segment is retained when a new segment is processed. Otherwise, if there
244        are no lookbehinds in the pattern, \A might match incorrectly at the start
245        of a new segment.
246    
247    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
248    
249    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
250        gives some modest performance improvement in 8-bit mode.
251    
252    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
253        be expressed in certain programming languages using Universal Character
254        Names.
255    
256    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
257        which points out that "non characters" are not "characters that may not
258        appear in Unicode strings" but rather "characters that are reserved for
259        internal use and have only local meaning".
260    
261    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
262        there was a conditional group that depended on an assertion, if the
263        assertion was false, the callout that immediately followed the alternation
264        in the condition was skipped when pcre_exec() was used for matching.
265    
266    23. Allow an explicit callout to be inserted before an assertion that is the
267        condition for a conditional group, for compatibility with automatic
268        callouts, which always insert a callout at this point.
269    
270    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
271        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
272    
273    25. (*PRUNE) is now supported by the JIT compiler.
274    
275    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
276    
277    27. Fix the case where there are two or more SKIPs with arguments that may be
278        ignored.
279    
280    28. (*SKIP) is now supported by the JIT compiler.
281    
282    29. (*THEN) is now supported by the JIT compiler.
283    
284    30. Update RunTest with additional test selector options.
285    
286    31. The way PCRE handles backtracking verbs has been changed in two ways.
287    
288        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
289        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
290        backtracking. In some cases this makes it more Perl-compatible, but Perl's
291        rather obscure rules do not always do the same thing.
292    
293        (2) Previously, backtracking verbs were confined within assertions. This is
294        no longer the case for positive assertions, except for (*ACCEPT). Again,
295        this sometimes improves Perl compatibility, and sometimes does not.
296    
297    32. A number of tests that were in test 2 because Perl did things differently
298        have been moved to test 1, because either Perl or PCRE has changed, and
299        these tests are now compatible.
300    
301    32. Backtracking control verbs are now handled in the same way in JIT and
302        interpreter.
303    
304    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
305        contained a forward subroutine reference caused a compile error.
306    
307    34. Auto-detect and optimize limited repetitions in JIT.
308    
309    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
310        blocking (*UTF) etc.
311    
312    36. In the interpreter, maximizing pattern repetitions for characters and
313        character types now use tail recursion, which reduces stack usage.
314    
315    37. The value of the max lookbehind was not correctly preserved if a compiled
316        and saved regex was reloaded on a host of different endianness.
317    
318    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
319        of the compiled pattern block, expand the flags field from 16 to 32 bits
320        because it was almost full.
321    
322    39. Try madvise first before posix_madvise.
323    
324    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
325        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
326        original change locked out all matches of empty strings. This has been
327        changed so that one match of an empty string per line is recognized.
328        Subsequent searches on the same line (for colouring or for --only-matching,
329        for example) do not recognize empty strings.
330    
331    41. Applied a user patch to fix a number of spelling mistakes in comments.
332    
333    42. Data lines longer than 65536 caused pcretest to crash.
334    
335    43. Clarified the data type for length and startoffset arguments for pcre_exec
336        and pcre_dfa_exec in the function-specific man pages, where they were
337        explicitly stated to be in bytes, never having been updated. I also added
338        some clarification to the pcreapi man page.
339    
340    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
341        a segmentation fault.
342    
343    
344    Version 8.32 30-November-2012
345    -----------------------------
346    
347    1.  Improved JIT compiler optimizations for first character search and single
348        character iterators.
349    
350    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
351        Patch by Daniel Richard G.
352    
353    3.  Single character iterator optimizations in the JIT compiler.
354    
355    4.  Improved JIT compiler optimizations for character ranges.
356    
357    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
358        Reported by Giuseppe D'Angelo.
359    
360    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
361        of a line, was being set incorrectly in cases where .* appeared inside
362        atomic brackets at the start of a pattern, or where there was a subsequent
363        *PRUNE or *SKIP.
364    
365    7.  Improved instruction cache flush for POWER/PowerPC.
366        Patch by Daniel Richard G.
367    
368    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
369        grep:
370    
371        (a) There is now no limit to the number of patterns to be matched.
372    
373        (b) An error is given if a pattern is too long.
374    
375        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
376            are now supported.
377    
378        (d) --exclude-from and --include-from (multiple use) have been added.
379    
380        (e) Exclusions and inclusions now apply to all files and directories, not
381            just to those obtained from scanning a directory recursively.
382    
383        (f) Multiple uses of -f and --file-list are now supported.
384    
385        (g) In a Windows environment, the default for -d has been changed from
386            "read" (the GNU grep default) to "skip", because otherwise the presence
387            of a directory in the file list provokes an error.
388    
389        (h) The documentation has been revised and clarified in places.
390    
391    9.  Improve the matching speed of capturing brackets.
392    
393    10. Changed the meaning of \X so that it now matches a Unicode extended
394        grapheme cluster.
395    
396    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
397        out POSIX threads when JIT support is configured.
398    
399    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
400    
401    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
402        structure could go wrong in environments where size_t is not the same size
403        as int.
404    
405    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
406        set.
407    
408    15. The EBCDIC support had decayed; later updates to the code had included
409        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
410        general tidy up of EBCDIC-related issues, and the documentation was also
411        not quite right. There is now a test that can be run on ASCII systems to
412        check some of the EBCDIC-related things (but is it not a full test).
413    
414    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
415        in a small tidy to the code.
416    
417    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
418    
419    18. If the --only-matching (-o) option in pcregrep is specified multiple
420        times, each one causes appropriate output. For example, -o1 -o2 outputs the
421        substrings matched by the 1st and 2nd capturing parentheses. A separating
422        string can be specified by --om-separator (default empty).
423    
424    19. Improving the first n character searches.
425    
426    20. Turn case lists for horizontal and vertical white space into macros so that
427        they are defined only once.
428    
429    21. This set of changes together give more compatible Unicode case-folding
430        behaviour for characters that have more than one other case when UCP
431        support is available.
432    
433        (a) The Unicode property table now has offsets into a new table of sets of
434            three or more characters that are case-equivalent. The MultiStage2.py
435            script that generates these tables (the pcre_ucd.c file) now scans
436            CaseFolding.txt instead of UnicodeData.txt for character case
437            information.
438    
439        (b) The code for adding characters or ranges of characters to a character
440            class has been abstracted into a generalized function that also handles
441            case-independence. In UTF-mode with UCP support, this uses the new data
442            to handle characters with more than one other case.
443    
444        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
445            whose other case is greater than 256 are now correctly matched
446            caselessly. Previously, the high codepoint matched the low one, but not
447            vice versa.
448    
449        (d) The processing of \h, \H, \v, and \ in character classes now makes use
450            of the new class addition function, using character lists defined as
451            macros alongside the case definitions of 20 above.
452    
453        (e) Caseless back references now work with characters that have more than
454            one other case.
455    
456        (f) General caseless matching of characters with more than one other case
457            is supported.
458    
459    22. Unicode character properties were updated from Unicode 6.2.0
460    
461    23. Improved CMake support under Windows. Patch by Daniel Richard G.
462    
463    24. Add support for 32-bit character strings, and UTF-32
464    
465    25. Major JIT compiler update (code refactoring and bugfixing).
466        Experimental Sparc 32 support is added.
467    
468    26. Applied a modified version of Daniel Richard G's patch to create
469        pcre.h.generic and config.h.generic by "make" instead of in the
470        PrepareRelease script.
471    
472    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
473        pcre_compile.c when checking for a zero character.
474    
475    28. Introducing a native interface for JIT. Through this interface, the compiled
476        machine code can be directly executed. The purpose of this interface is to
477        provide fast pattern matching, so several sanity checks are not performed.
478        However, feature tests are still performed. The new interface provides
479        1.4x speedup compared to the old one.
480    
481    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
482        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
483        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
484    
485    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
486        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
487        the "old" RFC 2279). Instead, it ended up passing a negative length to
488        pcre_exec().
489    
490    31. Add support for GCC's visibility feature to hide internal functions.
491    
492    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
493        "unknown -C option" after outputting 0 or 1.
494    
495    33. There is now support for generating a code coverage report for the test
496        suite in environments where gcc is the compiler and lcov is installed. This
497        is mainly for the benefit of the developers.
498    
499    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
500        unaddressable using valgrind annotations, allowing valgrind to detect
501        invalid memory accesses. This is mainly for the benefit of the developers.
502    
503    25. (*UTF) can now be used to start a pattern in any of the three libraries.
504    
505    26. Give configure error if --enable-cpp but no C++ compiler found.
506    
507    
508    Version 8.31 06-July-2012
509    -------------------------
510    
511    1.  Fixing a wrong JIT test case and some compiler warnings.
512    
513    2.  Removed a bashism from the RunTest script.
514    
515    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
516        to unsigned type, result still unsigned" that was given by an MS compiler
517        on encountering the code "-sizeof(xxx)".
518    
519    4.  Partial matching support is added to the JIT compiler.
520    
521    5.  Fixed several bugs concerned with partial matching of items that consist
522        of more than one character:
523    
524        (a) /^(..)\1/ did not partially match "aba" because checking references was
525            done on an "all or nothing" basis. This also applied to repeated
526            references.
527    
528        (b) \R did not give a hard partial match if \r was found at the end of the
529            subject.
530    
531        (c) \X did not give a hard partial match after matching one or more
532            characters at the end of the subject.
533    
534        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
535            a partial match for the string "\r".
536    
537        (e) When newline was set to CRLF, the metacharacter "." did not recognize
538            a partial match for a CR character at the end of the subject string.
539    
540    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
541        running pcretest, the text "(JIT)" added to the output whenever JIT is
542        actually used to run the match.
543    
544    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
545        or /S+[+] with a digit between 1 and 7.
546    
547    8.  OP_NOT now supports any UTF character not just single-byte ones.
548    
549    9.  (*MARK) control verb is now supported by the JIT compiler.
550    
551    10. The command "./RunTest list" lists the available tests without actually
552        running any of them. (Because I keep forgetting what they all are.)
553    
554    11. Add PCRE_INFO_MAXLOOKBEHIND.
555    
556    12. Applied a (slightly modified) user-supplied patch that improves performance
557        when the heap is used for recursion (compiled with --disable-stack-for-
558        recursion). Instead of malloc and free for each heap frame each time a
559        logical recursion happens, frames are retained on a chain and re-used where
560        possible. This sometimes gives as much as 30% improvement.
561    
562    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
563        call.
564    
565    14. As documented, (*COMMIT) is now confined to within a positive assertion.
566    
567    15. It is now possible to link pcretest with libedit as an alternative to
568        libreadline.
569    
570    16. (*COMMIT) control verb is now supported by the JIT compiler.
571    
572    17. The Unicode data tables have been updated to Unicode 6.1.0.
573    
574    18. Added --file-list option to pcregrep.
575    
576    19. Added binary file support to pcregrep, including the -a, --binary-files,
577        -I, and --text options.
578    
579    20. The madvise function is renamed for posix_madvise for QNX compatibility
580        reasons. Fixed by Giuseppe D'Angelo.
581    
582    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
583    
584    22. Changed the option for creating symbolic links for 16-bit man pages from
585        -s to -sf so that re-installing does not cause issues.
586    
587    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
588    
589    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
590        matches in certain environments (the workspace was not being correctly
591        retained). Also added to pcre_dfa_exec() a simple plausibility check on
592        some of the workspace data at the beginning of a restart.
593    
594    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
595        was not doing so when it should - probably a typo introduced by SVN 528
596        (change 8.10/14).
597    
598    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
599        \w+ when the character tables indicated that \x{c4} was a word character.
600        There were several related cases, all because the tests for doing a table
601        lookup were testing for characters less than 127 instead of 255.
602    
603    27. If a pattern contains capturing parentheses that are not used in a match,
604        their slots in the ovector are set to -1. For those that are higher than
605        any matched groups, this happens at the end of processing. In the case when
606        there were back references that the ovector was too small to contain
607        (causing temporary malloc'd memory to be used during matching), and the
608        highest capturing number was not used, memory off the end of the ovector
609        was incorrectly being set to -1. (It was using the size of the temporary
610        memory instead of the true size.)
611    
612    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
613        ovector size, it uses memory at the end of the block that it has got.
614    
615    29. Check for an overlong MARK name and give an error at compile time. The
616        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
617    
618    30. JIT compiler update.
619    
620    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
621        Rill for the patch.
622    
623    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
624    
625    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
626    
627    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
628        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
629        was enabled.
630    
631    35. Improve JIT code generation for greedy plus quantifier.
632    
633    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
634        1 to "aa" instead of to an empty string. The bug affected repeated groups
635        that could potentially match an empty string.
636    
637    37. Optimizing single character iterators in JIT.
638    
639    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
640        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
641        codepoints that are too big for the mode are faulted, and in a UTF mode,
642        disallowed codepoints are also faulted.
643    
644    39. If PCRE was compiled with UTF support, in three places in the DFA
645        matcher there was code that should only have been obeyed in UTF mode, but
646        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
647        processing when bytes with values greater than 127 were present. In 16-bit
648        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
649        both cases the values are those that cannot be the first data item in a UTF
650        character. The three items that might have provoked this were recursions,
651        possessively repeated groups, and atomic groups.
652    
653    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
654        and pcregrep, because some OS require shared objects to be explicitly
655        passed to ld, causing the link step to fail if they are not.
656    
657    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
658        mode, patterns that started with \h* or \R* might be incorrectly matched.
659    
660    
661    Version 8.30 04-February-2012
662    -----------------------------
663    
664    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
665        name is defined in ctype.h.
666    
667    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
668        only in quite long subpatterns.
669    
670    3.  Removed the function pcre_info(), which has been obsolete and deprecated
671        since it was replaced by pcre_fullinfo() in February 2000.
672    
673    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
674        match a (*MARK), and the match failed at the start of the subject, a
675        reference to memory before the start of the subject could occur. This bug
676        was introduced by fix 17 of release 8.21.
677    
678    5.  A reference to an unset group with zero minimum repetition was giving
679        totally wrong answers (in non-JavaScript-compatibility mode). For example,
680        /(another)?(\1?)test/ matched against "hello world test". This bug was
681        introduced in release 8.13.
682    
683    6.  Add support for 16-bit character strings (a large amount of work involving
684        many changes and refactorings).
685    
686    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
687        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
688        from a file.
689    
690    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
691        rounding is not applied in this particular case).
692    
693    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
694        if they appear, or are escaped, in patterns.
695    
696    10. Get rid of a number of -Wunused-but-set-variable warnings.
697    
698    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
699        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
700        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
701        also returns the mark "x". This bug applied to capturing parentheses,
702        non-capturing parentheses, and atomic parentheses. It also applied to some
703        assertions.
704    
705    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
706        information out of configure.ac instead of relying on pcre.h.generic, which
707        is not stored in the repository.
708    
709    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
710        -lreadline.
711    
712    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
713    
714    15. Applied Graycode's patch to put the top-level frame on the stack rather
715        than the heap when not using the stack for recursion. This gives a
716        performance improvement in many cases when recursion is not deep.
717    
718    16. Experimental code added to "pcretest -C" to output the stack frame size.
719    
720    
721    Version 8.21 12-Dec-2011
722    ------------------------
723    
724    1.  Updating the JIT compiler.
725    
726    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
727        are added as well.
728    
729    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
730        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
731        calling _pcre_jit_exec. Some extra comments are added.
732    
733    4.  (*MARK) settings inside atomic groups that do not contain any capturing
734        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
735        was introduced by change 18 for 8.20.
736    
737    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
738        ECMA-262 standard.
739    
740    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
741        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
742        This bug was probably introduced by change 9 of 8.13.
743    
744    7.  While fixing 6 above, I noticed that a number of other items were being
745        incorrectly rejected as "not fixed length". This arose partly because newer
746        opcodes had not been added to the fixed-length checking code. I have (a)
747        corrected the bug and added tests for these items, and (b) arranged for an
748        error to occur if an unknown opcode is encountered while checking for fixed
749        length instead of just assuming "not fixed length". The items that were
750        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
751        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
752        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
753    
754    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
755        being incorrectly compiled and would have given unpredicatble results.
756    
757    9.  A possessively repeated subpattern with minimum repeat count greater than
758        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
759        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
760        the first (A) could occur when it should not.
761    
762    10. Add a cast and remove a redundant test from the code.
763    
764    11. JIT should use pcre_malloc/pcre_free for allocation.
765    
766    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
767        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
768        is anything other than /usr, -L is still shown).
769    
770    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
771    
772    14. Perl does not support \N without a following name in a [] class; PCRE now
773        also gives an error.
774    
775    15. If a forward reference was repeated with an upper limit of around 2000,
776        it caused the error "internal error: overran compiling workspace". The
777        maximum number of forward references (including repeats) was limited by the
778        internal workspace, and dependent on the LINK_SIZE. The code has been
779        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
780        the default depends on LINK_SIZE. There is a new upper limit (for safety)
781        of around 200,000 forward references. While doing this, I also speeded up
782        the filling in of repeated forward references.
783    
784    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
785        incorrectly expecting the subject to contain another "a" after the start.
786    
787    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
788        in the match, the SKIP should be ignored. This was not happening; instead
789        the SKIP was being treated as NOMATCH. For patterns such as
790        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
791        tested.
792    
793    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
794        now much more compatible with Perl, in particular in cases where the result
795        is a non-match for a non-anchored pattern. For example, if
796        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
797        "m", where previously it did not return a name. A side effect of this
798        change is that for partial matches, the last encountered mark name is
799        returned, as for non matches. A number of tests that were previously not
800        Perl-compatible have been moved into the Perl-compatible test files. The
801        refactoring has had the pleasing side effect of removing one argument from
802        the match() function, thus reducing its stack requirements.
803    
804    19. If the /S+ option was used in pcretest to study a pattern using JIT,
805        subsequent uses of /S (without +) incorrectly behaved like /S+.
806    
807    21. Retrieve executable code size support for the JIT compiler and fixing
808        some warnings.
809    
810    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
811        not work when the shorter character appeared right at the end of the
812        subject string.
813    
814    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
815        systems.
816    
817    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
818        output it when the /M option is used in pcretest.
819    
820    25. The CheckMan script was not being included in the distribution. Also, added
821        an explicit "perl" to run Perl scripts from the PrepareRelease script
822        because this is reportedly needed in Windows.
823    
824    26. If study data was being save in a file and studying had not found a set of
825        "starts with" bytes for the pattern, the data written to the file (though
826        never used) was taken from uninitialized memory and so caused valgrind to
827        complain.
828    
829    27. Updated RunTest.bat as provided by Sheri Pierce.
830    
831    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
832    
833    29. Computation of memory usage for the table of capturing group names was
834        giving an unnecessarily large value.
835    
836    
837    Version 8.20 21-Oct-2011
838    ------------------------
839    
840    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
841        a POSIX class. After further experiments with Perl, which convinced me that
842        Perl has bugs and confusions, a closing square bracket is no longer allowed
843        in a POSIX name. This bug also affected patterns with classes that started
844        with full stops.
845    
846    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
847        captured substring, but while checking the failing first alternative,
848        substring 1 is temporarily captured. If the output vector supplied to
849        pcre_exec() was not big enough for this capture, the yield of the function
850        was still zero ("insufficient space for captured substrings"). This cannot
851        be totally fixed without adding another stack variable, which seems a lot
852        of expense for a edge case. However, I have improved the situation in cases
853        such as /(a)(b)x|abc/ matched against "abc", where the return code
854        indicates that fewer than the maximum number of slots in the ovector have
855        been set.
856    
857    3.  Related to (2) above: when there are more back references in a pattern than
858        slots in the output vector, pcre_exec() uses temporary memory during
859        matching, and copies in the captures as far as possible afterwards. It was
860        using the entire output vector, but this conflicts with the specification
861        that only 2/3 is used for passing back captured substrings. Now it uses
862        only the first 2/3, for compatibility. This is, of course, another edge
863        case.
864    
865    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
866        main code base, and can be used by building with --enable-jit. When this is
867        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
868        runtime --no-jit option is given.
869    
870    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
871        ovector, the return from the function was zero, implying that there were
872        other matches that did not fit. The correct "exactly full" value is now
873        returned.
874    
875    6.  If a subpattern that was called recursively or as a subroutine contained
876        (*PRUNE) or any other control that caused it to give a non-standard return,
877        invalid errors such as "Error -26 (nested recursion at the same subject
878        position)" or even infinite loops could occur.
879    
880    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
881        computing the minimum length on reaching *ACCEPT, and so ended up with the
882        wrong value of 1 rather than 0. Further investigation indicates that
883        computing a minimum subject length in the presence of *ACCEPT is difficult
884        (think back references, subroutine calls), and so I have changed the code
885        so that no minimum is registered for a pattern that contains *ACCEPT.
886    
887    8.  If (*THEN) was present in the first (true) branch of a conditional group,
888        it was not handled as intended. [But see 16 below.]
889    
890    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
891        Sheri Pierce.
892    
893    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
894        the first byte in a match must be "a".
895    
896    11. Change 17 for 8.13 increased the recursion depth for patterns like
897        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
898        pattern contains any instances of (*THEN). If it does not, the old
899        optimizations are restored. It would be nice to do this on a per-group
900        basis, but at the moment that is not feasible.
901    
902    12. In some environments, the output of pcretest -C is CRLF terminated. This
903        broke RunTest's code that checks for the link size. A single white space
904        character after the value is now allowed for.
905    
906    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
907        For "fr", it uses the Windows-specific input and output files.
908    
909    14. If (*THEN) appeared in a group that was called recursively or as a
910        subroutine, it did not work as intended. [But see next item.]
911    
912    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
913        pattern fragments (but not containing any | characters). If A and B are
914        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
915        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
916        D. In other words, Perl considers parentheses that do not contain any |
917        characters to be part of a surrounding alternative, whereas PCRE was
918        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
919        differently. PCRE now behaves in the same way as Perl, except in the case
920        of subroutine/recursion calls such as (?1) which have in any case always
921        been different (but PCRE had them first :-).
922    
923    16. Related to 15 above: Perl does not treat the | in a conditional group as
924        creating alternatives. Such a group is treated in the same way as an
925        ordinary group without any | characters when processing (*THEN). PCRE has
926        been changed to match Perl's behaviour.
927    
928    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
929        RunGrepTest script failed.
930    
931    18. Change 22 for version 13 caused atomic groups to use more stack. This is
932        inevitable for groups that contain captures, but it can lead to a lot of
933        stack use in large patterns. The old behaviour has been restored for atomic
934        groups that do not contain any capturing parentheses.
935    
936    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
937        suppress the check for a minimum subject length at run time. (If it was
938        given to pcre_exec() or pcre_dfa_exec() it did work.)
939    
940    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
941        fail to work when decoding hex characters in data strings in EBCDIC
942        environments.
943    
944    21. It appears that in at least one Mac OS environment, the isxdigit() function
945        is implemented as a macro that evaluates to its argument more than once,
946        contravening the C 90 Standard (I haven't checked a later standard). There
947        was an instance in pcretest which caused it to go wrong when processing
948        \x{...} escapes in subject strings. The has been rewritten to avoid using
949        things like p++ in the argument of isxdigit().
950    
951    
952    Version 8.13 16-Aug-2011
953    ------------------------
954    
955    1.  The Unicode data tables have been updated to Unicode 6.0.0.
956    
957    2.  Two minor typos in pcre_internal.h have been fixed.
958    
959    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
960        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
961        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
962    
963    4.  There were a number of related bugs in the code for matching backrefences
964        caselessly in UTF-8 mode when codes for the characters concerned were
965        different numbers of bytes. For example, U+023A and U+2C65 are an upper
966        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
967        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
968        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
969        2-byte code at the end of the subject (it thought there wasn't enough data
970        left).
971    
972    5.  Comprehensive information about what went wrong is now returned by
973        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
974        as the output vector has at least 2 elements. The offset of the start of
975        the failing character and a reason code are placed in the vector.
976    
977    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
978        now returned is for the first byte of the failing character, instead of the
979        last byte inspected. This is an incompatible change, but I hope it is small
980        enough not to be a problem. It makes the returned offset consistent with
981        pcre_exec() and pcre_dfa_exec().
982    
983    7.  pcretest now gives a text phrase as well as the error number when
984        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
985        failure, the offset and reason code are output.
986    
987    8.  When \R was used with a maximizing quantifier it failed to skip backwards
988        over a \r\n pair if the subsequent match failed. Instead, it just skipped
989        back over a single character (\n). This seems wrong (because it treated the
990        two characters as a single entity when going forwards), conflicts with the
991        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
992        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
993        has been changed.
994    
995    9.  Some internal refactoring has changed the processing so that the handling
996        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
997        time (the PCRE_DOTALL option was changed this way some time ago: version
998        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
999        which was always a bit of a fudge. It also means that there is one less
1000        argument for the match() function, which reduces its stack requirements
1001        slightly. This change also fixes an incompatibility with Perl: the pattern
1002        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
1003    
1004    10. More internal refactoring has drastically reduced the number of recursive
1005        calls to match() for possessively repeated groups such as (abc)++ when
1006        using pcre_exec().
1007    
1008    11. While implementing 10, a number of bugs in the handling of groups were
1009        discovered and fixed:
1010    
1011        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
1012        (a|)*(?1) gave a compile-time internal error.
1013        ((a|)+)+  did not notice that the outer group could match an empty string.
1014        (^a|^)+   was not marked as anchored.
1015        (.*a|.*)+ was not marked as matching at start or after a newline.
1016    
1017    12. Yet more internal refactoring has removed another argument from the match()
1018        function. Special calls to this function are now indicated by setting a
1019        value in a variable in the "match data" data block.
1020    
1021    13. Be more explicit in pcre_study() instead of relying on "default" for
1022        opcodes that mean there is no starting character; this means that when new
1023        ones are added and accidentally left out of pcre_study(), testing should
1024        pick them up.
1025    
1026    14. The -s option of pcretest has been documented for ages as being an old
1027        synonym of -m (show memory usage). I have changed it to mean "force study
1028        for every regex", that is, assume /S for every regex. This is similar to -i
1029        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
1030        using it. It makes it easier to run collections of tests with and without
1031        study enabled, and thereby test pcre_study() more easily. All the standard
1032        tests are now run with and without -s (but some patterns can be marked as
1033        "never study" - see 20 below).
1034    
1035    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
1036        restoration of the capturing data to the outer values was not happening
1037        correctly.
1038    
1039    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
1040        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
1041        pattern had matched an empty string, and so incorrectly returned a no
1042        match.
1043    
1044    17. There was optimizing code for the last branch of non-capturing parentheses,
1045        and also for the obeyed branch of a conditional subexpression, which used
1046        tail recursion to cut down on stack usage. Unfortunately, now that there is
1047        the possibility of (*THEN) occurring in these branches, tail recursion is
1048        no longer possible because the return has to be checked for (*THEN). These
1049        two optimizations have therefore been removed. [But see 8.20/11 above.]
1050    
1051    18. If a pattern containing \R was studied, it was assumed that \R always
1052        matched two bytes, thus causing the minimum subject length to be
1053        incorrectly computed because \R can also match just one byte.
1054    
1055    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
1056        was incorrectly computed.
1057    
1058    20. If /S is present twice on a test pattern in pcretest input, it now
1059        *disables* studying, thereby overriding the use of -s on the command line
1060        (see 14 above). This is necessary for one or two tests to keep the output
1061        identical in both cases.
1062    
1063    21. When (*ACCEPT) was used in an assertion that matched an empty string and
1064        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
1065    
1066    22. When an atomic group that contained a capturing parenthesis was
1067        successfully matched, but the branch in which it appeared failed, the
1068        capturing was not being forgotten if a higher numbered group was later
1069        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
1070        group 1 to "a", when in fact it should be unset. This applied to multi-
1071        branched capturing and non-capturing groups, repeated or not, and also to
1072        positive assertions (capturing in negative assertions does not happen
1073        in PCRE) and also to nested atomic groups.
1074    
1075    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
1076        subject after a captured substring, to make it easier to tell which of a
1077        number of identical substrings has been captured.
1078    
1079    24. The way atomic groups are processed by pcre_exec() has been changed so that
1080        if they are repeated, backtracking one repetition now resets captured
1081        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
1082        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
1083        "aaa". Previously, it would have been "a". As part of this code
1084        refactoring, the way recursive calls are handled has also been changed.
1085    
1086    25. If an assertion condition captured any substrings, they were not passed
1087        back unless some other capturing happened later. For example, if
1088        (?(?=(a))a) was matched against "a", no capturing was returned.
1089    
1090    26. When studying a pattern that contained subroutine calls or assertions,
1091        the code for finding the minimum length of a possible match was handling
1092        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
1093        group 1 called group 2 while simultaneously a separate group 2 called group
1094        1). A stack overflow occurred in this case. I have fixed this by limiting
1095        the recursion depth to 10.
1096    
1097    27. Updated RunTest.bat in the distribution to the version supplied by Tom
1098        Fortmann. This supports explicit test numbers on the command line, and has
1099        argument validation and error reporting.
1100    
1101    28. An instance of \X with an unlimited repeat could fail if at any point the
1102        first character it looked at was a mark character.
1103    
1104    29. Some minor code refactoring concerning Unicode properties and scripts
1105        should reduce the stack requirement of match() slightly.
1106    
1107    30. Added the '=' option to pcretest to check the setting of unused capturing
1108        slots at the end of the pattern, which are documented as being -1, but are
1109        not included in the return count.
1110    
1111    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
1112        compiled something random. Now it gives a compile-time error (as does
1113        Perl).
1114    
1115    32. A *MARK encountered during the processing of a positive assertion is now
1116        recorded and passed back (compatible with Perl).
1117    
1118    33. If --only-matching or --colour was set on a pcregrep call whose pattern
1119        had alternative anchored branches, the search for a second match in a line
1120        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
1121        matched the line "0102" twice. The same bug affected patterns that started
1122        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
1123        twice.
1124    
1125    34. Previously, PCRE did not allow quantification of assertions. However, Perl
1126        does, and because of capturing effects, quantifying parenthesized
1127        assertions may at times be useful. Quantifiers are now allowed for
1128        parenthesized assertions.
1129    
1130    35. A minor code tidy in pcre_compile() when checking options for \R usage.
1131    
1132    36. \g was being checked for fancy things in a character class, when it should
1133        just be a literal "g".
1134    
1135    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
1136        appearance of a nested POSIX class supersedes an apparent external class.
1137        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
1138        unescaped square brackets may also appear as part of class names. For
1139        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
1140        more like Perl. (But see 8.20/1 above.)
1141    
1142    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
1143        was because it thought it was \N{name}, which is not supported).
1144    
1145    39. Add minix to OS list not supporting the -S option in pcretest.
1146    
1147    40. PCRE tries to detect cases of infinite recursion at compile time, but it
1148        cannot analyze patterns in sufficient detail to catch mutual recursions
1149        such as ((?1))((?2)). There is now a runtime test that gives an error if a
1150        subgroup is called recursively as a subpattern for a second time at the
1151        same position in the subject string. In previous releases this might have
1152        been caught by the recursion limit, or it might have run out of stack.
1153    
1154    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
1155        happen only once. PCRE was, however incorrectly giving a compile time error
1156        "recursive call could loop indefinitely" because it cannot analyze the
1157        pattern in sufficient detail. The compile time test no longer happens when
1158        PCRE is compiling a conditional subpattern, but actual runaway loops are
1159        now caught at runtime (see 40 above).
1160    
1161    42. It seems that Perl allows any characters other than a closing parenthesis
1162        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
1163        has been changed to be the same.
1164    
1165    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
1166        as not to get warnings when autogen.sh is called. Also changed
1167        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
1168    
1169    44. To help people who use pcregrep to scan files containing exceedingly long
1170        lines, the following changes have been made:
1171    
1172        (a) The default value of the buffer size parameter has been increased from
1173            8K to 20K. (The actual buffer used is three times this size.)
1174    
1175        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
1176            PCRE is built.
1177    
1178        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
1179            to be set at run time.
1180    
1181        (d) Numerical values in pcregrep options can be followed by K or M, for
1182            example --buffer-size=50K.
1183    
1184        (e) If a line being scanned overflows pcregrep's buffer, an error is now
1185            given and the return code is set to 2.
1186    
1187    45. Add a pointer to the latest mark to the callout data block.
1188    
1189    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
1190        partial match of an empty string instead of no match. This was specific to
1191        the use of ".".
1192    
1193    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
1194        complete match instead of a partial match. This bug was dependent on both
1195        the PCRE_UTF8 and PCRE_DOTALL options being set.
1196    
1197    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1198        starting byte set, because \b was not being ignored.
1199    
1200    
1201    Version 8.12 15-Jan-2011
1202    ------------------------
1203    
1204    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1205        checks for such things as part of the documentation building process.
1206    
1207    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1208        --match-limit and --recursion-limit options (added for 8.11). In
1209        particular, this made one of the standard tests fail. (The integer value
1210        went into the wrong half of a long int.)
1211    
1212    3.  If the --colour option was given to pcregrep with -v (invert match), it
1213        did strange things, either producing crazy output, or crashing. It should,
1214        of course, ignore a request for colour when reporting lines that do not
1215        match.
1216    
1217    4.  Another pcregrep bug caused similar problems if --colour was specified with
1218        -M (multiline) and the pattern match finished with a line ending.
1219    
1220    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1221        matched in multiline mode, the following line was shown as part of the
1222        match. This seems wrong, so I have changed it.
1223    
1224    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1225        the check for further matches in the same line (so they could be coloured)
1226        to overrun the end of the current line. If another match was found, it was
1227        incorrectly shown (and then shown again when found in the next line).
1228    
1229    7.  If pcregrep was compiled under Windows, there was a reference to the
1230        function pcregrep_exit() before it was defined. I am assuming this was
1231        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1232        reported by a user. I've moved the definition above the reference.
1233    
1234    
1235    Version 8.11 10-Dec-2010
1236    ------------------------
1237    
1238    1.  (*THEN) was not working properly if there were untried alternatives prior
1239        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1240        backtracked to try for "b" instead of moving to the next alternative branch
1241        at the same level (in this case, to look for "c"). The Perl documentation
1242        is clear that when (*THEN) is backtracked onto, it goes to the "next
1243        alternative in the innermost enclosing group".
1244    
1245    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1246        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1247        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1248        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1249        (*THEN).
1250    
1251    3.  If \s appeared in a character class, it removed the VT character from
1252        the class, even if it had been included by some previous item, for example
1253        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1254        of \s, but is part of the POSIX "space" class.)
1255    
1256    4.  A partial match never returns an empty string (because you can always
1257        match an empty string at the end of the subject); however the checking for
1258        an empty string was starting at the "start of match" point. This has been
1259        changed to the "earliest inspected character" point, because the returned
1260        data for a partial match starts at this character. This means that, for
1261        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1262        (previously it gave "no match").
1263    
1264    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1265        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1266        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1267        has an implication that the given string is incomplete (because a partial
1268        match is preferred over a full match). For this reason, these items now
1269        give a partial match in this situation. [Aside: previously, the one case
1270        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1271        match rather than a full match, which was wrong by the old rules, but is
1272        now correct.]
1273    
1274    6.  There was a bug in the handling of #-introduced comments, recognized when
1275        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1276        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1277        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1278        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1279        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1280        places in pcre_compile().
1281    
1282    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1283        comments when looking ahead for named forward references to subpatterns,
1284        the only newline sequence it recognized was NL. It now handles newlines
1285        according to the set newline convention.
1286    
1287    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1288        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1289        cater for a lack of strerror(). These oversights have been fixed.
1290    
1291    9.  Added --match-limit and --recursion-limit to pcregrep.
1292    
1293    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1294    
1295    11. When the -o option was used, pcregrep was setting a return code of 1, even
1296        when matches were found, and --line-buffered was not being honoured.
1297    
1298    12. Added an optional parentheses number to the -o and --only-matching options
1299        of pcregrep.
1300    
1301    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1302        can match an empty string. The code to do it in pcretest and pcredemo
1303        needed fixing:
1304    
1305        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1306            only one byte after an empty string match just before CRLF (this case
1307            just got forgotten; "any" and "anycrlf" were OK).
1308    
1309        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1310            mode when an empty string match preceded an ASCII character followed by
1311            a non-ASCII character. (The code for advancing by one character rather
1312            than one byte was nonsense.)
1313    
1314        (c) The pcredemo.c sample program did not have any code at all to handle
1315            the cases when CRLF is a valid newline sequence.
1316    
1317    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1318        as a starting offset was within the subject string. There is now a new
1319        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1320        negative or greater than the length of the string. In order to test this,
1321        pcretest is extended to allow the setting of negative starting offsets.
1322    
1323    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1324        starting offset points to the beginning of a UTF-8 character was
1325        unnecessarily clumsy. I tidied it up.
1326    
1327    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1328        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1329    
1330    17. Nobody had reported that the --include_dir option, which was added in
1331        release 7.7 should have been called --include-dir (hyphen, not underscore)
1332        for compatibility with GNU grep. I have changed it to --include-dir, but
1333        left --include_dir as an undocumented synonym, and the same for
1334        --exclude-dir, though that is not available in GNU grep, at least as of
1335        release 2.5.4.
1336    
1337    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1338        characters from a string of bytes) have been redefined so as not to use
1339        loops, in order to improve performance in some environments. At the same
1340        time, I abstracted some of the common code into auxiliary macros to save
1341        repetition (this should not affect the compiled code).
1342    
1343    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1344        compile-time error is now given if \c is not followed by an ASCII
1345        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1346        different, and any byte value is allowed.)
1347    
1348    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1349        START_OPTIMIZE option, which is now allowed at compile time - but just
1350        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1351        to pcregrep and other applications that have no direct access to PCRE
1352        options. The new /Y option in pcretest sets this option when calling
1353        pcre_compile().
1354    
1355    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1356        back references. Groups containing recursive back references were forced to
1357        be atomic by that change, but in the case of named groups, the amount of
1358        memory required was incorrectly computed, leading to "Failed: internal
1359        error: code overflow". This has been fixed.
1360    
1361    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1362        pcretest.c, to avoid build problems in some Borland environments.
1363    
1364    
1365    Version 8.10 25-Jun-2010
1366    ------------------------
1367    
1368    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1369        THEN.
1370    
1371    2.  (*ACCEPT) was not working when inside an atomic group.
1372    
1373    3.  Inside a character class, \B is treated as a literal by default, but
1374        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1375        causes the error). The code is unchanged, but I tidied the documentation.
1376    
1377    4.  Inside a character class, PCRE always treated \R and \X as literals,
1378        whereas Perl faults them if its -w option is set. I have changed PCRE so
1379        that it faults them when PCRE_EXTRA is set.
1380    
1381    5.  Added support for \N, which always matches any character other than
1382        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1383    
1384    6.  When compiling pcregrep with newer versions of gcc which may have
1385        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1386        declared with attribute warn_unused_result" were given. Just casting the
1387        result to (void) does not stop the warnings; a more elaborate fudge is
1388        needed. I've used a macro to implement this.
1389    
1390    7.  Minor change to pcretest.c to avoid a compiler warning.
1391    
1392    8.  Added four artifical Unicode properties to help with an option to make
1393        \s etc use properties (see next item). The new properties are: Xan
1394        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1395    
1396    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1397        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1398        this option. Modified pcretest to add /W to test this facility. Added
1399        REG_UCP to make it available via the POSIX interface.
1400    
1401    10. Added --line-buffered to pcregrep.
1402    
1403    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1404        studied, and the match started with a letter with a code point greater than
1405        127 whose first byte was different to the first byte of the other case of
1406        the letter, the other case of this starting letter was not recognized
1407        (#976).
1408    
1409    12. If a pattern that was studied started with a repeated Unicode property
1410        test, for example, \p{Nd}+, there was the theoretical possibility of
1411        setting up an incorrect bitmap of starting bytes, but fortunately it could
1412        not have actually happened in practice until change 8 above was made (it
1413        added property types that matched character-matching opcodes).
1414    
1415    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1416        possible starting bytes for non-anchored patterns.
1417    
1418    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1419        \R, and also a number of cases that involve Unicode properties, both
1420        explicit and implicit when PCRE_UCP is set.
1421    
1422    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1423        input, it could crash or give wrong results if characters with values
1424        greater than 0xc0 were present in the subject string. (Detail: it assumed
1425        UTF-8 input when processing these items.)
1426    
1427    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1428        size_t is 64-bit (#991).
1429    
1430    17. Added a check for running out of memory when PCRE is compiled with
1431        --disable-stack-for-recursion (#990).
1432    
1433    18. If the last data line in a file for pcretest does not have a newline on
1434        the end, a newline was missing in the output.
1435    
1436    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1437        less than 128) in its various bitmaps. However, there is a facility for
1438        generating tables according to the current locale when PCRE is compiled. It
1439        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1440        characters, are recognized by isspace() and therefore were getting set in
1441        these tables, and indeed these tables seem to approximate to ISO 8859. This
1442        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1443        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1444        which of course cannot start UTF-8 characters. I have changed the code so
1445        that only real ASCII characters (less than 128) and the correct starting
1446        bytes for UTF-8 encodings are set for characters greater than 127 when in
1447        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1448        altogether.)
1449    
1450    20. Added the /T option to pcretest so as to be able to run tests with non-
1451        standard character tables, thus making it possible to include the tests
1452        used for 19 above in the standard set of tests.
1453    
1454    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1455        reference to a subpattern the other side of a comment that contains an
1456        opening parenthesis caused either an internal compiling error, or a
1457        reference to the wrong subpattern.
1458    
1459    
1460    Version 8.02 19-Mar-2010
1461    ------------------------
1462    
1463    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1464    
1465    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1466        configured.
1467    
1468    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1469        original author of that file, following a query about its status.
1470    
1471    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1472        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1473    
1474    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1475        quantifier applied to a forward-referencing subroutine call, could compile
1476        incorrect code or give the error "internal error: previously-checked
1477        referenced subpattern not found".
1478    
1479    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1480        variables to point to external functions. For these systems, therefore,
1481        pcre_malloc etc. are now initialized to local functions that call the
1482        relevant global functions.
1483    
1484    7.  There were two entries missing in the vectors called coptable and poptable
1485        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1486        I've fixed the data, and added a kludgy way of testing at compile time that
1487        the lengths are correct (equal to the number of opcodes).
1488    
1489    8.  Following on from 7, I added a similar kludge to check the length of the
1490        eint vector in pcreposix.c.
1491    
1492    9.  Error texts for pcre_compile() are held as one long string to avoid too
1493        much relocation at load time. To find a text, the string is searched,
1494        counting zeros. There was no check for running off the end of the string,
1495        which could happen if a new error number was added without updating the
1496        string.
1497    
1498    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1499    
1500    11. \K was not working if it appeared in an atomic group or in a group that
1501        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1502        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1503        the assertion is positive, but not if it is negative.
1504    
1505    12. Change 11 fortuitously reduced the size of the stack frame used in the
1506        "match()" function of pcre_exec.c by one pointer. Forthcoming
1507        implementation of support for (*MARK) will need an extra pointer on the
1508        stack; I have reserved it now, so that the stack frame size does not
1509        decrease.
1510    
1511    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1512        item in branch that calls a recursion is a subroutine call - as in the
1513        second branch in the above example - was incorrectly given the compile-
1514        time error "recursive call could loop indefinitely" because pcre_compile()
1515        was not correctly checking the subroutine for matching a non-empty string.
1516    
1517    14. The checks for overrunning compiling workspace could trigger after an
1518        overrun had occurred. This is a "should never occur" error, but it can be
1519        triggered by pathological patterns such as hundreds of nested parentheses.
1520        The checks now trigger 100 bytes before the end of the workspace.
1521    
1522    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1523    
1524    
1525    Version 8.01 19-Jan-2010
1526    ------------------------
1527    
1528    1.  If a pattern contained a conditional subpattern with only one branch (in
1529        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1530        computed the wrong minimum data length (which is of course zero for such
1531        subpatterns). This could cause incorrect "no match" results.
1532    
1533    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1534        the pattern is reset in the first branch, pcre_compile() failed with
1535        "internal error: code overflow at offset...". This happened only when
1536        the reset was to the original external option setting. (An optimization
1537        abstracts leading options settings into an external setting, which was the
1538        cause of this.)
1539    
1540    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1541        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1542        assertion pattern did not match (meaning that the assertion was true), it
1543        was incorrectly treated as false if the SKIP had been reached during the
1544        matching. This also applied to assertions used as conditions.
1545    
1546    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1547        assertion subpattern, including such a pattern used as a condition,
1548        unpredictable results occurred, instead of the error return
1549        PCRE_ERROR_DFA_UITEM.
1550    
1551    5.  The C++ GlobalReplace function was not working like Perl for the special
1552        situation when an empty string is matched. It now does the fancy magic
1553        stuff that is necessary.
1554    
1555    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1556        removed. (These were left over from very, very early versions of PCRE.)
1557    
1558    7.  Some cosmetic changes to the code to make life easier when compiling it
1559        as part of something else:
1560    
1561        (a) Change DEBUG to PCRE_DEBUG.
1562    
1563        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1564            called "current" as "current_branch", to prevent a collision with the
1565            Linux macro when compiled as a kernel module.
1566    
1567        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1568            prevent a collision with the Linux macro when compiled as a kernel
1569            module.
1570    
1571    8.  In pcre_compile() there are some checks for integer overflows that used to
1572        cast potentially large values to (double). This has been changed to that
1573        when building, a check for int64_t is made, and if it is found, it is used
1574        instead, thus avoiding the use of floating point arithmetic. (There is no
1575        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1576        double.
1577    
1578    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1579        2005 (difference between two addresses compared to an unsigned value).
1580    
1581    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1582        custom one, because of the following reported problem in Windows:
1583    
1584          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1585              under Win32.
1586          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1587              therefore missing the function definition.
1588          - The compiler thus generates a "C" signature for the test function.
1589          - The linker fails to find the "C" function.
1590          - PCRE fails to configure if asked to do so against libbz2.
1591    
1592    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1593        messages were output:
1594    
1595          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1596          rerunning libtoolize, to keep the correct libtool macros in-tree.
1597          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1598    
1599        I have done both of these things.
1600    
1601    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1602        most of the time, it *can* run out if it is given a pattern that contains a
1603        runaway infinite recursion. I updated the discussion in the pcrestack man
1604        page.
1605    
1606    13. Now that we have gone to the x.xx style of version numbers, the minor
1607        version may start with zero. Using 08 or 09 is a bad idea because users
1608        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1609        interpreted as invalid octal numbers. I've updated the previous comment in
1610        configure.ac, and also added a check that gives an error if 08 or 09 are
1611        used.
1612    
1613    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1614        causing partial matching to fail when the end of the subject matched \W
1615        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1616    
1617    15. There were some discrepancies between the declarations in pcre_internal.h
1618        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1619        their definitions. The declarations used "const uschar *" and the
1620        definitions used USPTR. Even though USPTR is normally defined as "const
1621        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1622        reported that: "This difference in casting confuses some C++ compilers, for
1623        example, SunCC recognizes above declarations as different functions and
1624        generates broken code for hbpcre." I have changed the declarations to use
1625        USPTR.
1626    
1627    16. GNU libtool is named differently on some systems. The autogen.sh script now
1628        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1629        (FreeBSD).
1630    
1631    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1632        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1633        comment: "Figure out how to create a longlong from a string: strtoll and
1634        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1635        instance, but it only takes 2 args instead of 3!"
1636    
1637    18. A subtle bug concerned with back references has been fixed by a change of
1638        specification, with a corresponding code fix. A pattern such as
1639        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1640        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1641        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1642        same bug. Such groups have to be quantified to be useful, or contained
1643        inside another quantified group. (If there's no repetition, the reference
1644        can never match.) The problem arises because, having left the group and
1645        moved on to the rest of the pattern, a later failure that backtracks into
1646        the group uses the captured value from the final iteration of the group
1647        rather than the correct earlier one. I have fixed this in PCRE by forcing
1648        any group that contains a reference to itself to be an atomic group; that
1649        is, there cannot be any backtracking into it once it has completed. This is
1650        similar to recursive and subroutine calls.
1651    
1652    
1653    Version 8.00 19-Oct-09
1654    ----------------------
1655    
1656    1.  The table for translating pcre_compile() error codes into POSIX error codes
1657        was out-of-date, and there was no check on the pcre_compile() error code
1658        being within the table. This could lead to an OK return being given in
1659        error.
1660    
1661    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1662        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1663        in a Windows environment.
1664    
1665    3.  The pcregrep --count option prints the count for each file even when it is
1666        zero, as does GNU grep. However, pcregrep was also printing all files when
1667        --files-with-matches was added. Now, when both options are given, it prints
1668        counts only for those files that have at least one match. (GNU grep just
1669        prints the file name in this circumstance, but including the count seems
1670        more useful - otherwise, why use --count?) Also ensured that the
1671        combination -clh just lists non-zero counts, with no names.
1672    
1673    4.  The long form of the pcregrep -F option was incorrectly implemented as
1674        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1675        but it seems right to fix it, and I didn't think it was worth preserving
1676        the old behaviour.
1677    
1678    5.  The command line items --regex=pattern and --regexp=pattern were not
1679        recognized by pcregrep, which required --regex pattern or --regexp pattern
1680        (with a space rather than an '='). The man page documented the '=' forms,
1681        which are compatible with GNU grep; these now work.
1682    
1683    6.  No libpcreposix.pc file was created for pkg-config; there was just
1684        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1685    
1686    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1687        when UCP support is not needed, by modifying the Python script that
1688        generates it from Unicode data files. This should not matter if the module
1689        is correctly used as a library, but I received one complaint about 50K of
1690        unwanted data. My guess is that the person linked everything into his
1691        program rather than using a library. Anyway, it does no harm.
1692    
1693    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1694        was a minimum greater than 1 for a wide character in a possessive
1695        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1696        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1697        character. Chaos in the form of incorrect output or a compiling loop could
1698        result.
1699    
1700    9.  The restrictions on what a pattern can contain when partial matching is
1701        requested for pcre_exec() have been removed. All patterns can now be
1702        partially matched by this function. In addition, if there are at least two
1703        slots in the offset vector, the offset of the earliest inspected character
1704        for the match and the offset of the end of the subject are set in them when
1705        PCRE_ERROR_PARTIAL is returned.
1706    
1707    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1708        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1709        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1710        and may be more useful for multi-segment matching.
1711    
1712    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1713        used to be given if ever the end of the subject was reached; now it is
1714        given only if matching could not proceed because another character was
1715        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1716        string "Z", which now yields "no match" instead of "partial match". In the
1717        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1718        final character ended with (*FAIL).
1719    
1720    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1721        if the pattern had a "must contain" character that was already found in the
1722        earlier partial match, unless partial matching was again requested. For
1723        example, with the pattern /dog.(body)?/, the "must contain" character is
1724        "g". If the first part-match was for the string "dog", restarting with
1725        "sbody" failed. This bug has been fixed.
1726    
1727    13. The string returned by pcre_dfa_exec() after a partial match has been
1728        changed so that it starts at the first inspected character rather than the
1729        first character of the match. This makes a difference only if the pattern
1730        starts with a lookbehind assertion or \b or \B (\K is not supported by
1731        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1732        matching functions compatible, and I think it's the right thing to do.
1733    
1734    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1735        so that the demonstration program is easily available in environments where
1736        PCRE has not been installed from source.
1737    
1738    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1739        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1740        library.
1741    
1742    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1743        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1744        is not the first non-POSIX option to be added. Clearly some people find
1745        these options useful.
1746    
1747    17. If a caller to the POSIX matching function regexec() passes a non-zero
1748        value for nmatch with a NULL value for pmatch, the value of
1749        nmatch is forced to zero.
1750    
1751    18. RunGrepTest did not have a test for the availability of the -u option of
1752        the diff command, as RunTest does. It now checks in the same way as
1753        RunTest, and also checks for the -b option.
1754    
1755    19. If an odd number of negated classes containing just a single character
1756        interposed, within parentheses, between a forward reference to a named
1757        subpattern and the definition of the subpattern, compilation crashed with
1758        an internal error, complaining that it could not find the referenced
1759        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1760        [The bug was that it was starting one character too far in when skipping
1761        over the character class, thus treating the ] as data rather than
1762        terminating the class. This meant it could skip too much.]
1763    
1764    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1765        /g option in pcretest when the pattern contains \K, which makes it possible
1766        to have an empty string match not at the start, even when the pattern is
1767        anchored. Updated pcretest and pcredemo to use this option.
1768    
1769    21. If the maximum number of capturing subpatterns in a recursion was greater
1770        than the maximum at the outer level, the higher number was returned, but
1771        with unset values at the outer level. The correct (outer level) value is
1772        now given.
1773    
1774    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1775        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1776        make it do so. The string so far is captured, making this feature
1777        compatible with Perl.
1778    
1779    23. The tests have been re-organized, adding tests 11 and 12, to make it
1780        possible to check the Perl 5.10 features against Perl 5.10.
1781    
1782    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1783        pattern matches a fixed length string. PCRE did not allow this; now it
1784        does. Neither allows recursion.
1785    
1786    25. I finally figured out how to implement a request to provide the minimum
1787        length of subject string that was needed in order to match a given pattern.
1788        (It was back references and recursion that I had previously got hung up
1789        on.) This code has now been added to pcre_study(); it finds a lower bound
1790        to the length of subject needed. It is not necessarily the greatest lower
1791        bound, but using it to avoid searching strings that are too short does give
1792        some useful speed-ups. The value is available to calling programs via
1793        pcre_fullinfo().
1794    
1795    26. While implementing 25, I discovered to my embarrassment that pcretest had
1796        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1797        study optimizations had never been tested with that matching function.
1798        Oops. What is worse, even when it was passed study data, there was a bug in
1799        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1800        were also very few tests of studied patterns with pcre_dfa_exec().
1801    
1802    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1803        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1804        on the other side of the coin, they are no longer allowed to have different
1805        names, because these cannot be distinguished in PCRE, and this has caused
1806        confusion. (This is a difference from Perl.)
1807    
1808    28. When duplicate subpattern names are present (necessarily with different
1809        numbers, as required by 27 above), and a test is made by name in a
1810        conditional pattern, either for a subpattern having been matched, or for
1811        recursion in such a pattern, all the associated numbered subpatterns are
1812        tested, and the overall condition is true if the condition is true for any
1813        one of them. This is the way Perl works, and is also more like the way
1814        testing by number works.
1815    
1816    
1817    Version 7.9 11-Apr-09
1818    ---------------------
1819    
1820    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1821        (pcretest), all targets were linked against these libraries. This included
1822        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1823        libraries. This caused unwanted dependencies to be created. This problem
1824        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1825        pcretest is linked with readline.
1826    
1827    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1828        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1829        moved outside it again, because FALSE and TRUE are already defined in AIX,
1830        but BOOL is not.
1831    
1832    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1833        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1834    
1835    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1836        hyphens) following file names and line numbers when outputting matching
1837        lines. This is not true; no spaces are inserted. I have also clarified the
1838        wording for the --colour (or --color) option.
1839    
1840    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1841        was not coloured; this is different to GNU grep, so I have changed it to be
1842        the same.
1843    
1844    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1845        each matching line was coloured. Now it goes on to look for further matches
1846        of any of the test patterns, which is the same behaviour as GNU grep.
1847    
1848    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1849        doesn't make sense to accept an empty string match in pcregrep, so I have
1850        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1851        seems to be how GNU grep behaves. [But see later change 40 for release
1852        8.33.]
1853    
1854    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1855        start or after a newline", because the conditional assertion was not being
1856        correctly handled. The rule now is that both the assertion and what follows
1857        in the first alternative must satisfy the test.
1858    
1859    9.  If auto-callout was enabled in a pattern with a conditional group whose
1860        condition was an assertion, PCRE could crash during matching, both with
1861        pcre_exec() and pcre_dfa_exec().
1862    
1863    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1864        used for matching.
1865    
1866    11. Unicode property support in character classes was not working for
1867        characters (bytes) greater than 127 when not in UTF-8 mode.
1868    
1869    12. Added the -M command line option to pcretest.
1870    
1871    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1872    
1873    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1874    
1875    16. Added comments and documentation about mis-use of no_arg in the C++
1876        wrapper.
1877    
1878    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1879        from Martin Jerabek that uses macro names for all relevant character and
1880        string constants.
1881    
1882    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1883        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1884        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1885        these, but not everybody uses configure.
1886    
1887    19. A conditional group that had only one branch was not being correctly
1888        recognized as an item that could match an empty string. This meant that an
1889        enclosing group might also not be so recognized, causing infinite looping
1890        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1891        with the subject "ab", where knowledge that the repeated group can match
1892        nothing is needed in order to break the loop.
1893    
1894    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1895        exec(), but without supplying a callout function, matching went wrong.
1896    
1897    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1898        leak if the size of the offset vector was greater than 30. When the vector
1899        is smaller, the saved offsets during recursion go onto a local stack
1900        vector, but for larger vectors malloc() is used. It was failing to free
1901        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1902        error, in fact).
1903    
1904    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1905        heapframe that is used only when UTF-8 support is enabled. This caused no
1906        problem, but was untidy.
1907    
1908    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1909        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1910        included within another project.
1911    
1912    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1913        slightly modified by me:
1914    
1915          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1916              not building pcregrep.
1917    
1918          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1919              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1920    
1921    25. Forward references, both numeric and by name, in patterns that made use of
1922        duplicate group numbers, could behave incorrectly or give incorrect errors,
1923        because when scanning forward to find the reference group, PCRE was not
1924        taking into account the duplicate group numbers. A pattern such as
1925        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1926    
1927    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1928        the feature of a custom pointer more persuasive (as requested by a user).
1929    
1930    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1931        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1932        pre-defined.
1933    
1934    28. Added support for (*UTF8) at the start of a pattern.
1935    
1936    29. Arrange for flags added by the "release type" setting in CMake to be shown
1937        in the configuration summary.
1938    
1939    
1940    Version 7.8 05-Sep-08
1941    ---------------------
1942    
1943    1.  Replaced UCP searching code with optimized version as implemented for Ad
1944        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1945        stage table and inline lookup instead of a function, giving speed ups of 2
1946        to 5 times on some simple patterns that I tested. Permission was given to
1947        distribute the MultiStage2.py script that generates the tables (it's not in
1948        the tarball, but is in the Subversion repository).
1949    
1950    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1951        scripts.
1952    
1953    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1954        a group with a zero qualifier. The result of the study could be incorrect,
1955        or the function might crash, depending on the pattern.
1956    
1957    4.  Caseless matching was not working for non-ASCII characters in back
1958        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1959        It now works when Unicode Property Support is available.
1960    
1961    5.  In pcretest, an escape such as \x{de} in the data was always generating
1962        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1963        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1964        truncation.
1965    
1966    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1967    
1968    7.  Added two (int) casts to pcregrep when printing the difference of two
1969        pointers, in case they are 64-bit values.
1970    
1971    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1972        test 2 if it fails.
1973    
1974    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1975        and a #define of that name to empty if it is not externally set. This is to
1976        allow users of MSVC to set it if necessary.
1977    
1978    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1979        the convenience functions in the pcre_get.c source file.
1980    
1981    11. An option change at the start of a pattern that had top-level alternatives
1982        could cause overwriting and/or a crash. This command provoked a crash in
1983        some environments:
1984    
1985          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1986    
1987        This potential security problem was recorded as CVE-2008-2371.
1988    
1989    12. For a pattern where the match had to start at the beginning or immediately
1990        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1991        pcre_dfa_exec() could read past the end of the passed subject if there was
1992        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1993        pcretest so that it places the subject at the end of its malloc-ed buffer.
1994    
1995    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1996        exec() might read past the end of the data buffer in UTF-8 mode.
1997    
1998    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1999        the data contained the byte 0x85 as part of a UTF-8 character within its
2000        first line. This applied both to normal and DFA matching.
2001    
2002    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
2003        /^[^d]*?$/8 failed to match "abc".
2004    
2005    16. Added a missing copyright notice to pcrecpp_internal.h.
2006    
2007    17. Make it more clear in the documentation that values returned from
2008        pcre_exec() in ovector are byte offsets, not character counts.
2009    
2010    18. Tidied a few places to stop certain compilers from issuing warnings.
2011    
2012    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
2013        supplied by Stefan Weber. I made a further small update for 7.8 because
2014        there is a change of source arrangements: the pcre_searchfuncs.c module is
2015        replaced by pcre_ucd.c.
2016    
2017    
2018    Version 7.7 07-May-08
2019    ---------------------
2020    
2021    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
2022        a string to a long long, pretend we don't even have a long long." This is
2023        done by checking for the strtoq, strtoll, and _strtoi64 functions.
2024    
2025    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
2026        pre-7.6 versions, which defined a global no_arg variable instead of putting
2027        it in the RE class. (See also #8 below.)
2028    
2029    3.  Remove a line of dead code, identified by coverity and reported by Nuno
2030        Lopes.
2031    
2032    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
2033    
2034        (1) The include/exclude patterns were being applied to the whole pathnames
2035            of files, instead of just to the final components.
2036    
2037        (2) If there was more than one level of directory, the subdirectories were
2038            skipped unless they satisfied the include/exclude conditions. This is
2039            inconsistent with GNU grep (and could even be seen as contrary to the
2040            pcregrep specification - which I improved to make it absolutely clear).
2041            The action now is always to scan all levels of directory, and just
2042            apply the include/exclude patterns to regular files.
2043    
2044    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
2045        --exclude_dir in the tests to avoid scanning .svn directories.
2046    
2047    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
2048        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
2049        doesn't support NULs in patterns.
2050    
2051    7.  Added some missing "const"s to declarations of static tables in
2052        pcre_compile.c and pcre_dfa_exec.c.
2053    
2054    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
2055        caused by fix #2  above. (Subsequently also a second patch to fix the
2056        first patch. And a third patch - this was a messy problem.)
2057    
2058    9.  Applied Craig's patch to remove the use of push_back().
2059    
2060    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
2061        matching function regexec().
2062    
2063    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
2064        which, however, unlike Perl's \g{...}, are subroutine calls, not back
2065        references. PCRE supports relative numbers with this syntax (I don't think
2066        Oniguruma does).
2067    
2068    12. Previously, a group with a zero repeat such as (...){0} was completely
2069        omitted from the compiled regex. However, this means that if the group
2070        was called as a subroutine from elsewhere in the pattern, things went wrong
2071        (an internal error was given). Such groups are now left in the compiled
2072        pattern, with a new opcode that causes them to be skipped at execution
2073        time.
2074    
2075    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
2076        to the way PCRE behaves:
2077    
2078        (a) A lone ] character is dis-allowed (Perl treats it as data).
2079    
2080        (b) A back reference to an unmatched subpattern matches an empty string
2081            (Perl fails the current match path).
2082    
2083        (c) A data ] in a character class must be notated as \] because if the
2084            first data character in a class is ], it defines an empty class. (In
2085            Perl it is not possible to have an empty class.) The empty class []
2086            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
2087            The negative empty class [^] matches any one character, independently
2088            of the DOTALL setting.
2089    
2090    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
2091        non-existent subpattern following a character class starting with ']' and
2092        containing () gave an internal compiling error instead of "reference to
2093        non-existent subpattern". Fortunately, when the pattern did exist, the
2094        compiled code was correct. (When scanning forwards to check for the
2095        existence of the subpattern, it was treating the data ']' as terminating
2096        the class, so got the count wrong. When actually compiling, the reference
2097        was subsequently set up correctly.)
2098    
2099    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
2100        it was being rejected as not supported by pcre_dfa_exec(), even though
2101        other assertions are supported. I have made pcre_dfa_exec() support
2102        (*FAIL).
2103    
2104    16. The implementation of 13c above involved the invention of a new opcode,
2105        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
2106        cannot be changed at match time, I realized I could make a small
2107        improvement to matching performance by compiling OP_ALLANY instead of
2108        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
2109        on the OP_ANY path.
2110    
2111    17. Compiling pcretest on Windows with readline support failed without the
2112        following two fixes: (1) Make the unistd.h include conditional on
2113        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
2114    
2115    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
2116        ncurses library to be included for pcretest when ReadLine support is
2117        requested, but also to allow for it to be overridden. This patch came from
2118        Daniel Bergström.
2119    
2120    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
2121        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
2122        any errors with the current Unicode tables. Thanks to Peter Kankowski for
2123        spotting this.
2124    
2125    
2126    Version 7.6 28-Jan-08
2127    ---------------------
2128    
2129    1.  A character class containing a very large number of characters with
2130        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
2131        overflow.
2132    
2133    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
2134        HAVE_LONG_LONG is not defined.
2135    
2136    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
2137        bring them up to date and include new features. This patch includes:
2138    
2139        - Fixed PH's badly added libz and libbz2 support.
2140        - Fixed a problem with static linking.
2141        - Added pcredemo. [But later removed - see 7 below.]
2142        - Fixed dftables problem and added an option.
2143        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
2144            HAVE_LONG_LONG.
2145        - Added readline support for pcretest.
2146        - Added an listing of the option settings after cmake has run.
2147    
2148    4.  A user submitted a patch to Makefile that makes it easy to create
2149        "pcre.dll" under mingw when using Configure/Make. I added stuff to
2150        Makefile.am that cause it to include this special target, without
2151        affecting anything else. Note that the same mingw target plus all
2152        the other distribution libraries and programs are now supported
2153        when configuring with CMake (see 6 below) instead of with
2154        Configure/Make.
2155    
2156    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
2157        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
2158        exported in the Windows port". It has not yet been confirmed that the patch
2159        solves the problem, but it does no harm.
2160    
2161    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
2162        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
2163        with CMake, and also correct the comment about stack recursion.
2164    
2165    7.  Remove the automatic building of pcredemo from the ./configure system and
2166        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
2167        of a program that users should build themselves after PCRE is installed, so
2168        building it automatically is not really right. What is more, it gave
2169        trouble in some build environments.
2170    
2171    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
2172    
2173    
2174    Version 7.5 10-Jan-08
2175    ---------------------
2176    
2177    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
2178        values in parens when parsing an RE using the C++ wrapper."
2179    
2180    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
2181        Characters greater than 255 were excluded from the class instead of being
2182        included.
2183    
2184    3.  The same bug as (2) above applied to negated POSIX classes such as
2185        [:^space:].
2186    
2187    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
2188        defined or documented. It seems to have been a typo for PCRE_STATIC, so
2189        I have changed it.
2190    
2191    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
2192        first named subpattern) and a construct such as (?&a) would reference the
2193        first named subpattern whose name started with "a" (in other words, the
2194        length check was missing). Both these problems are fixed. "Subpattern name
2195        expected" is now given for (?&) (a zero-length name), and this patch also
2196        makes it give the same error for \k'' (previously it complained that that
2197        was a reference to a non-existent subpattern).
2198    
2199    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
2200        this is right because (?- can be followed by option settings as well as by
2201        digits. I have, however, made the messages clearer.
2202    
2203    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
2204        than the number used in the conditional) now cause a compile-time error.
2205        This is actually not compatible with Perl, which accepts such patterns, but
2206        treats the conditional as always being FALSE (as PCRE used to), but it
2207        seems to me that giving a diagnostic is better.
2208    
2209    8.  Change "alphameric" to the more common word "alphanumeric" in comments
2210        and messages.
2211    
2212    9.  Fix two occurrences of "backslash" in comments that should have been
2213        "backspace".
2214    
2215    10. Remove two redundant lines of code that can never be obeyed (their function
2216        was moved elsewhere).
2217    
2218    11. The program that makes PCRE's Unicode character property table had a bug
2219        which caused it to generate incorrect table entries for sequences of
2220        characters that have the same character type, but are in different scripts.
2221        It amalgamated them into a single range, with the script of the first of
2222        them. In other words, some characters were in the wrong script. There were
2223        thirteen such cases, affecting characters in the following ranges:
2224    
2225          U+002b0 - U+002c1
2226          U+0060c - U+0060d
2227          U+0061e - U+00612
2228          U+0064b - U+0065e
2229          U+0074d - U+0076d
2230          U+01800 - U+01805
2231          U+01d00 - U+01d77
2232          U+01d9b - U+01dbf
2233          U+0200b - U+0200f
2234          U+030fc - U+030fe
2235          U+03260 - U+0327f
2236          U+0fb46 - U+0fbb1
2237          U+10450 - U+1049d
2238    
2239    12. The -o option (show only the matching part of a line) for pcregrep was not
2240        compatible with GNU grep in that, if there was more than one match in a
2241        line, it showed only the first of them. It now behaves in the same way as
2242        GNU grep.
2243    
2244    13. If the -o and -v options were combined for pcregrep, it printed a blank
2245        line for every non-matching line. GNU grep prints nothing, and pcregrep now
2246        does the same. The return code can be used to tell if there were any
2247        non-matching lines.
2248    
2249    14. Added --file-offsets and --line-offsets to pcregrep.
2250    
2251    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
2252        infinitely looping recursion. The bug was that positive lookaheads were not
2253        being skipped when checking for a possible empty match (negative lookaheads
2254        and both kinds of lookbehind were skipped).
2255    
2256    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
2257        inclusion of <windows.h> to before rather than after the definition of
2258        INVALID_FILE_ATTRIBUTES (patch from David Byron).
2259    
2260    17. Specifying a possessive quantifier with a specific limit for a Unicode
2261        character property caused pcre_compile() to compile bad code, which led at
2262        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
2263        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
2264        caused the error; without that there was no problem.
2265    
2266    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
2267    
2268    19. Added --enable-pcretest-libreadline.
2269    
2270    20. In pcrecpp.cc, the variable 'count' was incremented twice in
2271        RE::GlobalReplace(). As a result, the number of replacements returned was
2272        double what it should be. I removed one of the increments, but Craig sent a
2273        later patch that removed the other one (the right fix) and added unit tests
2274        that check the return values (which was not done before).
2275    
2276    21. Several CMake things:
2277    
2278        (1) Arranged that, when cmake is used on Unix, the libraries end up with
2279            the names libpcre and libpcreposix, not just pcre and pcreposix.
2280    
2281        (2) The above change means that pcretest and pcregrep are now correctly
2282            linked with the newly-built libraries, not previously installed ones.
2283    
2284        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
2285    
2286    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
2287        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
2288        UTF-8 newline character). The key issue is that the pattern starts .*;
2289        this means that the match must be either at the beginning, or after a
2290        newline. The bug was in the code for advancing after a failed match and
2291        checking that the new position followed a newline. It was not taking
2292        account of UTF-8 characters correctly.
2293    
2294    23. PCRE was behaving differently from Perl in the way it recognized POSIX
2295        character classes. PCRE was not treating the sequence [:...:] as a
2296        character class unless the ... were all letters. Perl, however, seems to
2297        allow any characters between [: and :], though of course it rejects as
2298        unknown any "names" that contain non-letters, because all the known class
2299        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
2300        for example, whereas PCRE did not - it did not recognize a POSIX character
2301        class. This seemed a bit dangerous, so the code has been changed to be
2302        closer to Perl. The behaviour is not identical to Perl, because PCRE will
2303        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
2304        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
2305        Perl does, and where it didn't before.
2306    
2307    24. Rewrite so as to remove the single use of %n from pcregrep because in some
2308        Windows environments %n is disabled by default.
2309    
2310    
2311    Version 7.4 21-Sep-07
2312    ---------------------
2313    
2314    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
2315        means that a class such as [\s] counted as "explicit reference to CR or
2316        LF". That isn't really right - the whole point of the change was to try to
2317        help when there was an actual mention of one of the two characters. So now
2318        the change happens only if \r or \n (or a literal CR or LF) character is
2319        encountered.
2320    
2321    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
2322        of both had grown to the point where there were only 3 bits left.
2323        Fortunately, there was spare space in the data structure, and so I have
2324        moved the internal flags into a new 16-bit field to free up more option
2325        bits.
2326    
2327    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
2328        but did not set the internal JCHANGED flag - either of these is enough to
2329        control the way the "get" function works - but the PCRE_INFO_JCHANGED
2330        facility is supposed to tell if (?J) was ever used, so now (?J) at the
2331        start sets both bits.
2332    
2333    4.  Added options (at build time, compile time, exec time) to change \R from
2334        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
2335    
2336    5.  doc/pcresyntax.html was missing from the distribution.
2337    
2338    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
2339        compatibility, even though it is no longer used.
2340    
2341    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
2342        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
2343        windows.h file is present (where different names are used). [This was
2344        reversed later after testing - see 16 below.]
2345    
2346    8.  Changed all #include <config.h> to #include "config.h". There were also
2347        some further <pcre.h> cases that I changed to "pcre.h".
2348    
2349    9.  When pcregrep was used with the --colour option, it missed the line ending
2350        sequence off the lines that it output.
2351    
2352    10. It was pointed out to me that arrays of string pointers cause lots of
2353        relocations when a shared library is dynamically loaded. A technique of
2354        using a single long string with a table of offsets can drastically reduce
2355        these. I have refactored PCRE in four places to do this. The result is
2356        dramatic:
2357    
2358          Originally:                          290
2359          After changing UCP table:            187
2360          After changing error message table:   43
2361          After changing table of "verbs"       36
2362          After changing table of Posix names   22
2363    
2364        Thanks to the folks working on Gregex for glib for this insight.
2365    
2366    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
2367        unicode-properties was also set.
2368    
2369    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
2370    
2371    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
2372        checked only for CRLF.
2373    
2374    14. Added casts to pcretest.c to avoid compiler warnings.
2375    
2376    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
2377    
2378    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
2379        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2380        entirely. This removes changes made in 7 above.
2381    
2382    17. The CMake files have been updated, and there is now more information about
2383        building with CMake in the NON-UNIX-USE document.
2384    
2385    
2386    Version 7.3 28-Aug-07
2387    ---------------------
2388    
2389     1. In the rejigging of the build system that eventually resulted in 7.1, the
2390        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
2391        brackets there is not right, since it causes compilers to look for an
2392        installed pcre.h, not the version that is in the source that is being
2393        compiled (which of course may be different). I have changed it back to:
2394    
2395          #include "pcre.h"
2396    
2397        I have a vague recollection that the change was concerned with compiling in
2398        different directories, but in the new build system, that is taken care of
2399        by the VPATH setting the Makefile.
2400    
2401     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
2402        when the subject happened to end in the byte 0x85 (e.g. if the last
2403        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
2404        characters but of course it shouldn't be taken as a newline when it is part
2405        of another character. The bug was that, for an unlimited repeat of . in
2406        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
2407        characters when looking for a newline.
2408    
2409     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
2410    
2411     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
2412        in debug output.
2413    
2414     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
2415        long printing in the pcrecpp unittest when running under MinGW.
2416    
2417     6. ESC_K was left out of the EBCDIC table.
2418    
2419     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
2420        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
2421        limit also applies to "virtual nesting" when a pattern is recursive, and in
2422        this case 1000 isn't so big. I have been able to remove this limit at the
2423        expense of backing off one optimization in certain circumstances. Normally,
2424        when pcre_exec() would call its internal match() function recursively and
2425        immediately return the result unconditionally, it uses a "tail recursion"
2426        feature to save stack. However, when a subpattern that can match an empty
2427        string has an unlimited repetition quantifier, it no longer makes this
2428        optimization. That gives it a stack frame in which to save the data for
2429        checking that an empty string has been matched. Previously this was taken
2430        from the 1000-entry workspace that had been reserved. So now there is no
2431        explicit limit, but more stack is used.
2432    
2433     8. Applied Daniel's patches to solve problems with the import/export magic
2434        syntax that is required for Windows, and which was going wrong for the
2435        pcreposix and pcrecpp parts of the library. These were overlooked when this
2436        problem was solved for the main library.
2437    
2438     9. There were some crude static tests to avoid integer overflow when computing
2439        the size of patterns that contain repeated groups with explicit upper
2440        limits. As the maximum quantifier is 65535, the maximum group length was
2441        set at 30,000 so that the product of these two numbers did not overflow a
2442        32-bit integer. However, it turns out that people want to use groups that
2443        are longer than 30,000 bytes (though not repeat them that many times).
2444        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
2445        made it possible to implement the integer overflow checks in a much more
2446        dynamic way, which I have now done. The artificial limitation on group
2447        length has been removed - we now have only the limit on the total length of
2448        the compiled pattern, which depends on the LINK_SIZE setting.
2449    
2450    10. Fixed a bug in the documentation for get/copy named substring when
2451        duplicate names are permitted. If none of the named substrings are set, the
2452        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
2453        empty string.
2454    
2455    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
2456        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
2457        because the ] is interpreted as the first data character and the
2458        terminating ] is not found. PCRE has been made compatible with Perl in this
2459        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
2460        cause memory overwriting.
2461    
2462    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
2463        string has been matched (to stop an infinite loop). It was not recognizing
2464        a conditional subpattern that could match an empty string if that
2465        subpattern was within another subpattern. For example, it looped when
2466        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
2467        condition was not nested. This bug has been fixed.
2468    
2469    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
2470        past the start of the subject in the presence of bytes with the top bit
2471        set, for example "\x8aBCD".
2472    
2473    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
2474        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
2475    
2476    14. Optimized (?!) to (*FAIL).
2477    
2478    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
2479        This restricts code points to be within the range 0 to 0x10FFFF, excluding
2480        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
2481        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
2482        does: it's just the validity check that is more restrictive.
2483    
2484    16. Inserted checks for integer overflows during escape sequence (backslash)
2485        processing, and also fixed erroneous offset values for syntax errors during
2486        backslash processing.
2487    
2488    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
2489        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
2490    
2491    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
2492        caused an overrun.
2493    
2494    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
2495        something other than just ASCII characters) inside a group that had an
2496        unlimited repeat caused a loop at compile time (while checking to see
2497        whether the group could match an empty string).
2498    
2499    20. Debugging a pattern containing \p or \P could cause a crash. For example,
2500        [\P{Any}] did so. (Error in the code for printing property names.)
2501    
2502    21. An orphan \E inside a character class could cause a crash.
2503    
2504    22. A repeated capturing bracket such as (A)? could cause a wild memory
2505        reference during compilation.
2506    
2507    23. There are several functions in pcre_compile() that scan along a compiled
2508        expression for various reasons (e.g. to see if it's fixed length for look
2509        behind). There were bugs in these functions when a repeated \p or \P was
2510        present in the pattern. These operators have additional parameters compared
2511        with \d, etc, and these were not being taken into account when moving along
2512        the compiled data. Specifically:
2513    
2514        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2515            length.
2516    
2517        (b) An item such as \pL+ within a repeated group could cause crashes or
2518            loops.
2519    
2520        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2521            "reference to non-existent subpattern" error.
2522    
2523        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2524    
2525    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2526        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2527    
2528    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2529    
2530    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2531        character were causing crashes (broken optimization).
2532    
2533    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2534        \p or \P) caused a compile-time loop.
2535    
2536    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2537        break. For example, the unstudied pattern [\r\n]A does not match the string
2538        "\r\nA" because change 7.0/46 below moves the current point on by two
2539        characters after failing to match at the start. However, the pattern \nA
2540        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2541        the same is true. There doesn't seem any very clean way out of this, but
2542        what I have chosen to do makes the common cases work: PCRE now takes note
2543        of whether there can be an explicit match for \r or \n anywhere in the
2544        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2545        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2546        pattern has explicit CR or LF references.
2547    
2548    29. Added (*CR) etc for changing newline setting at start of pattern.
2549    
2550    
2551    Version 7.2 19-Jun-07
2552    ---------------------
2553    
2554     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
2555        which is apparently normally available under Windows.
2556    
2557     2. Re-jig the pcregrep tests with different newline settings in an attempt
2558        to make them independent of the local environment's newline setting.
2559    
2560     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
2561    
2562     4. Some of the "internals" tests were previously cut out when the link size
2563        was not 2, because the output contained actual offsets. The recent new
2564        "Z" feature of pcretest means that these can be cut out, making the tests
2565        usable with all link sizes.
2566    
2567     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
2568        stack recursion. This gives a massive performance boost under BSD, but just
2569        a small improvement under Linux. However, it saves one field in the frame
2570        in all cases.
2571    
2572     6. Added more features from the forthcoming Perl 5.10:
2573    
2574        (a) (?-n) (where n is a string of digits) is a relative subroutine or
2575            recursion call. It refers to the nth most recently opened parentheses.
2576    
2577        (b) (?+n) is also a relative subroutine call; it refers to the nth next
2578            to be opened parentheses.
2579    
2580        (c) Conditions that refer to capturing parentheses can be specified
2581            relatively, for example, (?(-2)... or (?(+3)...
2582    
2583        (d) \K resets the start of the current match so that everything before
2584            is not part of it.
2585    
2586        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2587    
2588        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2589            reference syntax.
2590    
2591        (g) (?| introduces a group in which the numbering of parentheses in each
2592            alternative starts with the same number.
2593    
2594        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2595    
2596     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2597        PCRE_INFO_JCHANGED.
2598    
2599     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2600        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2601        for detecting groups that can match an empty string.
2602    
2603     9. A pattern with a very large number of alternatives (more than several
2604        hundred) was running out of internal workspace during the pre-compile
2605        phase, where pcre_compile() figures out how much memory will be needed. A
2606        bit of new cunning has reduced the workspace needed for groups with
2607        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2608        workspace instead of running out of the 4096 that are available.
2609    
2610    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2611    
2612    11. Applied patch from Google to remove an optimization that didn't quite work.
2613        The report of the bug said:
2614    
2615          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2616          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2617          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2618    
2619    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2620        it matched the wrong number of bytes.
2621    
2622    
2623    Version 7.1 24-Apr-07
2624  ---------------------  ---------------------
2625    
2626   1. Applied Bob Rossi and Daniel G's patches to convert the build system to one   1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
2627      that is more "standard", making use of automake and other autotools. There      that is more "standard", making use of automake and other Autotools. There
2628      is some re-arrangement of the files and adjustment of comments consequent      is some re-arrangement of the files and adjustment of comments consequent
2629      on this.      on this.
2630    
# Line 26  Version 7.1 05-Mar-07 Line 2645  Version 7.1 05-Mar-07
2645      .br or .in.      .br or .in.
2646    
2647   4. Updated comments in configure.ac that get placed in config.h.in and also   4. Updated comments in configure.ac that get placed in config.h.in and also
2648      arranged for config.h to be included in the distribution, with the name      arranged for config.h to be included in the distribution, with the name
2649      config.h.generic, for the benefit of those who have to compile without      config.h.generic, for the benefit of those who have to compile without
2650      Autotools (compare pcre.h, which is now distributed as pcre.h.generic).      Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
2651    
2652   5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan   5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
2653      Weber: (1) pcre_internal.h was missing some function renames; (2) updated      Weber: (1) pcre_internal.h was missing some function renames; (2) updated
2654      makevp.bat for the current PCRE, using the additional files !compile.txt,      makevp.bat for the current PCRE, using the additional files
2655      !linklib.txt, and pcregexp.pas.      makevp_c.txt, makevp_l.txt, and pcregexp.pas.
2656    
2657   6. A Windows user reported a minor discrepancy with test 2, which turned out   6. A Windows user reported a minor discrepancy with test 2, which turned out
2658      to be caused by a trailing space on an input line that had got lost in his      to be caused by a trailing space on an input line that had got lost in his
# Line 63  Version 7.1 05-Mar-07 Line 2682  Version 7.1 05-Mar-07
2682  10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:  10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
2683    
2684      (a) It was defining its arguments as char * instead of void *.      (a) It was defining its arguments as char * instead of void *.
2685    
2686      (b) It was assuming that all moves were upwards in memory; this was true      (b) It was assuming that all moves were upwards in memory; this was true
2687          a long time ago when I wrote it, but is no longer the case.          a long time ago when I wrote it, but is no longer the case.
2688    
2689      The emulated memove() is provided for those environments that have neither      The emulated memove() is provided for those environments that have neither
2690      memmove() nor bcopy(). I didn't think anyone used it these days, but that      memmove() nor bcopy(). I didn't think anyone used it these days, but that
2691      is clearly not the case, as these two bugs were recently reported.      is clearly not the case, as these two bugs were recently reported.
2692    
2693  11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,  11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
2694      and Detrail to create the HTML documentation, the .txt form of the man      and Detrail to create the HTML documentation, the .txt form of the man
2695      pages, and it removes trailing spaces from listed files. It also creates      pages, and it removes trailing spaces from listed files. It also creates
2696      pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter      pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
2697      case, it wraps all the #defines with #ifndefs. This script should be run      case, it wraps all the #defines with #ifndefs. This script should be run
2698      before "make dist".      before "make dist".
2699    
2700  12. Fixed two fairly obscure bugs concerned with quantified caseless matching  12. Fixed two fairly obscure bugs concerned with quantified caseless matching
2701      with Unicode property support.      with Unicode property support.
2702    
2703      (a) For a maximizing quantifier, if the two different cases of the      (a) For a maximizing quantifier, if the two different cases of the
2704          character were of different lengths in their UTF-8 codings (there are          character were of different lengths in their UTF-8 codings (there are
2705          some cases like this - I found 11), and the matching function had to          some cases like this - I found 11), and the matching function had to
2706          back up over a mixture of the two cases, it incorrectly assumed they          back up over a mixture of the two cases, it incorrectly assumed they
2707          were both the same length.          were both the same length.
2708    
2709      (b) When PCRE was configured to use the heap rather than the stack for      (b) When PCRE was configured to use the heap rather than the stack for
2710          recursion during matching, it was not correctly preserving the data for          recursion during matching, it was not correctly preserving the data for
2711          the other case of a UTF-8 character when checking ahead for a match          the other case of a UTF-8 character when checking ahead for a match
2712          while processing a minimizing repeat. If the check also involved          while processing a minimizing repeat. If the check also involved
2713          matching a wide character, but failed, corruption could cause an          matching a wide character, but failed, corruption could cause an
2714          erroneous result when trying to check for a repeat of the original          erroneous result when trying to check for a repeat of the original
2715          character.          character.
2716    
2717  13. Some tidying changes to the testing mechanism:  13. Some tidying changes to the testing mechanism:
2718    
2719      (a) The RunTest script now detects the internal link size and whether there      (a) The RunTest script now detects the internal link size and whether there
2720          is UTF-8 and UCP support by running ./pcretest -C instead of relying on          is UTF-8 and UCP support by running ./pcretest -C instead of relying on
2721          values substituted by "configure". (The RunGrepTest script already did          values substituted by "configure". (The RunGrepTest script already did
2722          this for UTF-8.) The configure.ac script no longer substitutes the          this for UTF-8.) The configure.ac script no longer substitutes the
2723          relevant variables.          relevant variables.
2724    
2725      (b) The debugging options /B and /D in pcretest show the compiled bytecode      (b) The debugging options /B and /D in pcretest show the compiled bytecode
2726          with length and offset values. This means that the output is different          with length and offset values. This means that the output is different
2727          for different internal link sizes. Test 2 is skipped for link sizes          for different internal link sizes. Test 2 is skipped for link sizes
2728          other than 2 because of this, bypassing the problem. Unfortunately,          other than 2 because of this, bypassing the problem. Unfortunately,
2729          there was also a test in test 3 (the locale tests) that used /B and          there was also a test in test 3 (the locale tests) that used /B and
2730          failed for link sizes other than 2. Rather than cut the whole test out,          failed for link sizes other than 2. Rather than cut the whole test out,
2731          I have added a new /Z option to pcretest that replaces the length and          I have added a new /Z option to pcretest that replaces the length and
2732          offset values with spaces. This is now used to make test 3 independent          offset values with spaces. This is now used to make test 3 independent
2733          of link size.          of link size. (Test 2 will be tidied up later.)
2734    
2735    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
2736        segmentation fault instead of returning the appropriate error message.
2737    
2738    15. In multiline mode when the newline sequence was set to "any", the pattern
2739        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
2740        This doesn't seem right; it now treats the CRLF combination as the line
2741        ending, and so does not match in that case. It's only a pattern such as ^$
2742        that would hit this one: something like ^ABC$ would have failed after \r
2743        and then tried again after \r\n.
2744    
2745    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
2746        in an attempt to make files that differ only in their line terminators
2747        compare equal. This works on Linux.
2748    
2749    17. Under certain error circumstances pcregrep might try to free random memory
2750        as it exited. This is now fixed, thanks to valgrind.
2751    
2752    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
2753        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
2754        was because its rules for how to advance for /g after matching an empty
2755        string at the end of a line did not allow for this case. They now check for
2756        it specially.
2757    
2758    20. pcretest is supposed to handle patterns and data of any length, by
2759        extending its buffers when necessary. It was getting this wrong when the
2760        buffer for a data line had to be extended.
2761    
2762    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
2763        CRLF as a newline sequence.
2764    
2765    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
2766        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
2767        I have nevertheless tidied it up.
2768    
2769    23. Added some casts to kill warnings from HP-UX ia64 compiler.
2770    
2771    24. Added a man page for pcre-config.
2772    
2773    
2774  Version 7.0 19-Dec-06  Version 7.0 19-Dec-06
2775  ---------------------  ---------------------

Legend:
Removed from v.118  
changed lines
  Added in v.1394

  ViewVC Help
Powered by ViewVC 1.1.5