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

Diff of /code/trunk/ChangeLog

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

revision 447 by ph10, Tue Sep 15 18:17:54 2009 UTC revision 1362 by chpe, Tue Oct 1 11:27:04 2013 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 8.00 ??-???-??  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    Version 8.33 28-May-2013
86    ------------------------
87    
88    1.  Added 'U' to some constants that are compared to unsigned integers, to
89        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
90        variables that are added to signed variables, to ensure the result is
91        signed and can be negated.
92    
93    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
94        CMake config files.
95    
96    3.  Revise the creation of config.h.generic so that all boolean macros are
97        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
98        overriding via -D on the command line possible.
99    
100    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
101        to unsigned int is reported to make a quite noticeable speed difference in
102        a specific Windows environment. Testing on Linux did also appear to show
103        some benefit (and it is clearly not harmful). Also fixed the definition of
104        Xop which should be unsigned.
105    
106    5.  Related to (4), changing the definition of the intermediate variable cc
107        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
108        improvements.
109    
110    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
111        unnecessary spaces.
112    
113    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
114        and later.
115    
116    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
117    
118    9.  Optimizing fast_forward_start_bits in JIT.
119    
120    10. Adding support for callouts in JIT, and fixing some issues revealed
121        during this work. Namely:
122    
123        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
124    
125        (b) Minimum length was not checked before the matching is started.
126    
127    11. The value of capture_last that is passed to callouts was incorrect in some
128        cases when there was a capture on one path that was subsequently abandoned
129        after a backtrack. Also, the capture_last value is now reset after a
130        recursion, since all captures are also reset in this case.
131    
132    12. The interpreter no longer returns the "too many substrings" error in the
133        case when an overflowing capture is in a branch that is subsequently
134        abandoned after a backtrack.
135    
136    13. In the pathological case when an offset vector of size 2 is used, pcretest
137        now prints out the matched string after a yield of 0 or 1.
138    
139    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
140        Only supported by the JIT compiler at the moment.
141    
142    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
143    
144    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
145        offset of the starting point of the matching process, provided the offsets
146        vector is large enough.
147    
148    17. The \A escape now records a lookbehind value of 1, though its execution
149        does not actually inspect the previous character. This is to ensure that,
150        in partial multi-segment matching, at least one character from the old
151        segment is retained when a new segment is processed. Otherwise, if there
152        are no lookbehinds in the pattern, \A might match incorrectly at the start
153        of a new segment.
154    
155    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
156    
157    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
158        gives some modest performance improvement in 8-bit mode.
159    
160    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
161        be expressed in certain programming languages using Universal Character
162        Names.
163    
164    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
165        which points out that "non characters" are not "characters that may not
166        appear in Unicode strings" but rather "characters that are reserved for
167        internal use and have only local meaning".
168    
169    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
170        there was a conditional group that depended on an assertion, if the
171        assertion was false, the callout that immediately followed the alternation
172        in the condition was skipped when pcre_exec() was used for matching.
173    
174    23. Allow an explicit callout to be inserted before an assertion that is the
175        condition for a conditional group, for compatibility with automatic
176        callouts, which always insert a callout at this point.
177    
178    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
179        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
180    
181    25. (*PRUNE) is now supported by the JIT compiler.
182    
183    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
184    
185    27. Fix the case where there are two or more SKIPs with arguments that may be
186        ignored.
187    
188    28. (*SKIP) is now supported by the JIT compiler.
189    
190    29. (*THEN) is now supported by the JIT compiler.
191    
192    30. Update RunTest with additional test selector options.
193    
194    31. The way PCRE handles backtracking verbs has been changed in two ways.
195    
196        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
197        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
198        backtracking. In some cases this makes it more Perl-compatible, but Perl's
199        rather obscure rules do not always do the same thing.
200    
201        (2) Previously, backtracking verbs were confined within assertions. This is
202        no longer the case for positive assertions, except for (*ACCEPT). Again,
203        this sometimes improves Perl compatibility, and sometimes does not.
204    
205    32. A number of tests that were in test 2 because Perl did things differently
206        have been moved to test 1, because either Perl or PCRE has changed, and
207        these tests are now compatible.
208    
209    32. Backtracking control verbs are now handled in the same way in JIT and
210        interpreter.
211    
212    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
213        contained a forward subroutine reference caused a compile error.
214    
215    34. Auto-detect and optimize limited repetitions in JIT.
216    
217    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
218        blocking (*UTF) etc.
219    
220    36. In the interpreter, maximizing pattern repetitions for characters and
221        character types now use tail recursion, which reduces stack usage.
222    
223    37. The value of the max lookbehind was not correctly preserved if a compiled
224        and saved regex was reloaded on a host of different endianness.
225    
226    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
227        of the compiled pattern block, expand the flags field from 16 to 32 bits
228        because it was almost full.
229    
230    39. Try madvise first before posix_madvise.
231    
232    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
233        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
234        original change locked out all matches of empty strings. This has been
235        changed so that one match of an empty string per line is recognized.
236        Subsequent searches on the same line (for colouring or for --only-matching,
237        for example) do not recognize empty strings.
238    
239    41. Applied a user patch to fix a number of spelling mistakes in comments.
240    
241    42. Data lines longer than 65536 caused pcretest to crash.
242    
243    43. Clarified the data type for length and startoffset arguments for pcre_exec
244        and pcre_dfa_exec in the function-specific man pages, where they were
245        explicitly stated to be in bytes, never having been updated. I also added
246        some clarification to the pcreapi man page.
247    
248    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
249        a segmentation fault.
250    
251    
252    Version 8.32 30-November-2012
253    -----------------------------
254    
255    1.  Improved JIT compiler optimizations for first character search and single
256        character iterators.
257    
258    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
259        Patch by Daniel Richard G.
260    
261    3.  Single character iterator optimizations in the JIT compiler.
262    
263    4.  Improved JIT compiler optimizations for character ranges.
264    
265    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
266        Reported by Giuseppe D'Angelo.
267    
268    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
269        of a line, was being set incorrectly in cases where .* appeared inside
270        atomic brackets at the start of a pattern, or where there was a subsequent
271        *PRUNE or *SKIP.
272    
273    7.  Improved instruction cache flush for POWER/PowerPC.
274        Patch by Daniel Richard G.
275    
276    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
277        grep:
278    
279        (a) There is now no limit to the number of patterns to be matched.
280    
281        (b) An error is given if a pattern is too long.
282    
283        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
284            are now supported.
285    
286        (d) --exclude-from and --include-from (multiple use) have been added.
287    
288        (e) Exclusions and inclusions now apply to all files and directories, not
289            just to those obtained from scanning a directory recursively.
290    
291        (f) Multiple uses of -f and --file-list are now supported.
292    
293        (g) In a Windows environment, the default for -d has been changed from
294            "read" (the GNU grep default) to "skip", because otherwise the presence
295            of a directory in the file list provokes an error.
296    
297        (h) The documentation has been revised and clarified in places.
298    
299    9.  Improve the matching speed of capturing brackets.
300    
301    10. Changed the meaning of \X so that it now matches a Unicode extended
302        grapheme cluster.
303    
304    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
305        out POSIX threads when JIT support is configured.
306    
307    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
308    
309    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
310        structure could go wrong in environments where size_t is not the same size
311        as int.
312    
313    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
314        set.
315    
316    15. The EBCDIC support had decayed; later updates to the code had included
317        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
318        general tidy up of EBCDIC-related issues, and the documentation was also
319        not quite right. There is now a test that can be run on ASCII systems to
320        check some of the EBCDIC-related things (but is it not a full test).
321    
322    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
323        in a small tidy to the code.
324    
325    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
326    
327    18. If the --only-matching (-o) option in pcregrep is specified multiple
328        times, each one causes appropriate output. For example, -o1 -o2 outputs the
329        substrings matched by the 1st and 2nd capturing parentheses. A separating
330        string can be specified by --om-separator (default empty).
331    
332    19. Improving the first n character searches.
333    
334    20. Turn case lists for horizontal and vertical white space into macros so that
335        they are defined only once.
336    
337    21. This set of changes together give more compatible Unicode case-folding
338        behaviour for characters that have more than one other case when UCP
339        support is available.
340    
341        (a) The Unicode property table now has offsets into a new table of sets of
342            three or more characters that are case-equivalent. The MultiStage2.py
343            script that generates these tables (the pcre_ucd.c file) now scans
344            CaseFolding.txt instead of UnicodeData.txt for character case
345            information.
346    
347        (b) The code for adding characters or ranges of characters to a character
348            class has been abstracted into a generalized function that also handles
349            case-independence. In UTF-mode with UCP support, this uses the new data
350            to handle characters with more than one other case.
351    
352        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
353            whose other case is greater than 256 are now correctly matched
354            caselessly. Previously, the high codepoint matched the low one, but not
355            vice versa.
356    
357        (d) The processing of \h, \H, \v, and \ in character classes now makes use
358            of the new class addition function, using character lists defined as
359            macros alongside the case definitions of 20 above.
360    
361        (e) Caseless back references now work with characters that have more than
362            one other case.
363    
364        (f) General caseless matching of characters with more than one other case
365            is supported.
366    
367    22. Unicode character properties were updated from Unicode 6.2.0
368    
369    23. Improved CMake support under Windows. Patch by Daniel Richard G.
370    
371    24. Add support for 32-bit character strings, and UTF-32
372    
373    25. Major JIT compiler update (code refactoring and bugfixing).
374        Experimental Sparc 32 support is added.
375    
376    26. Applied a modified version of Daniel Richard G's patch to create
377        pcre.h.generic and config.h.generic by "make" instead of in the
378        PrepareRelease script.
379    
380    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
381        pcre_compile.c when checking for a zero character.
382    
383    28. Introducing a native interface for JIT. Through this interface, the compiled
384        machine code can be directly executed. The purpose of this interface is to
385        provide fast pattern matching, so several sanity checks are not performed.
386        However, feature tests are still performed. The new interface provides
387        1.4x speedup compared to the old one.
388    
389    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
390        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
391        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
392    
393    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
394        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
395        the "old" RFC 2279). Instead, it ended up passing a negative length to
396        pcre_exec().
397    
398    31. Add support for GCC's visibility feature to hide internal functions.
399    
400    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
401        "unknown -C option" after outputting 0 or 1.
402    
403    33. There is now support for generating a code coverage report for the test
404        suite in environments where gcc is the compiler and lcov is installed. This
405        is mainly for the benefit of the developers.
406    
407    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
408        unaddressable using valgrind annotations, allowing valgrind to detect
409        invalid memory accesses. This is mainly for the benefit of the developers.
410    
411    25. (*UTF) can now be used to start a pattern in any of the three libraries.
412    
413    26. Give configure error if --enable-cpp but no C++ compiler found.
414    
415    
416    Version 8.31 06-July-2012
417    -------------------------
418    
419    1.  Fixing a wrong JIT test case and some compiler warnings.
420    
421    2.  Removed a bashism from the RunTest script.
422    
423    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
424        to unsigned type, result still unsigned" that was given by an MS compiler
425        on encountering the code "-sizeof(xxx)".
426    
427    4.  Partial matching support is added to the JIT compiler.
428    
429    5.  Fixed several bugs concerned with partial matching of items that consist
430        of more than one character:
431    
432        (a) /^(..)\1/ did not partially match "aba" because checking references was
433            done on an "all or nothing" basis. This also applied to repeated
434            references.
435    
436        (b) \R did not give a hard partial match if \r was found at the end of the
437            subject.
438    
439        (c) \X did not give a hard partial match after matching one or more
440            characters at the end of the subject.
441    
442        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
443            a partial match for the string "\r".
444    
445        (e) When newline was set to CRLF, the metacharacter "." did not recognize
446            a partial match for a CR character at the end of the subject string.
447    
448    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
449        running pcretest, the text "(JIT)" added to the output whenever JIT is
450        actually used to run the match.
451    
452    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
453        or /S+[+] with a digit between 1 and 7.
454    
455    8.  OP_NOT now supports any UTF character not just single-byte ones.
456    
457    9.  (*MARK) control verb is now supported by the JIT compiler.
458    
459    10. The command "./RunTest list" lists the available tests without actually
460        running any of them. (Because I keep forgetting what they all are.)
461    
462    11. Add PCRE_INFO_MAXLOOKBEHIND.
463    
464    12. Applied a (slightly modified) user-supplied patch that improves performance
465        when the heap is used for recursion (compiled with --disable-stack-for-
466        recursion). Instead of malloc and free for each heap frame each time a
467        logical recursion happens, frames are retained on a chain and re-used where
468        possible. This sometimes gives as much as 30% improvement.
469    
470    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
471        call.
472    
473    14. As documented, (*COMMIT) is now confined to within a positive assertion.
474    
475    15. It is now possible to link pcretest with libedit as an alternative to
476        libreadline.
477    
478    16. (*COMMIT) control verb is now supported by the JIT compiler.
479    
480    17. The Unicode data tables have been updated to Unicode 6.1.0.
481    
482    18. Added --file-list option to pcregrep.
483    
484    19. Added binary file support to pcregrep, including the -a, --binary-files,
485        -I, and --text options.
486    
487    20. The madvise function is renamed for posix_madvise for QNX compatibility
488        reasons. Fixed by Giuseppe D'Angelo.
489    
490    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
491    
492    22. Changed the option for creating symbolic links for 16-bit man pages from
493        -s to -sf so that re-installing does not cause issues.
494    
495    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
496    
497    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
498        matches in certain environments (the workspace was not being correctly
499        retained). Also added to pcre_dfa_exec() a simple plausibility check on
500        some of the workspace data at the beginning of a restart.
501    
502    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
503        was not doing so when it should - probably a typo introduced by SVN 528
504        (change 8.10/14).
505    
506    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
507        \w+ when the character tables indicated that \x{c4} was a word character.
508        There were several related cases, all because the tests for doing a table
509        lookup were testing for characters less than 127 instead of 255.
510    
511    27. If a pattern contains capturing parentheses that are not used in a match,
512        their slots in the ovector are set to -1. For those that are higher than
513        any matched groups, this happens at the end of processing. In the case when
514        there were back references that the ovector was too small to contain
515        (causing temporary malloc'd memory to be used during matching), and the
516        highest capturing number was not used, memory off the end of the ovector
517        was incorrectly being set to -1. (It was using the size of the temporary
518        memory instead of the true size.)
519    
520    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
521        ovector size, it uses memory at the end of the block that it has got.
522    
523    29. Check for an overlong MARK name and give an error at compile time. The
524        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
525    
526    30. JIT compiler update.
527    
528    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
529        Rill for the patch.
530    
531    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
532    
533    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
534    
535    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
536        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
537        was enabled.
538    
539    35. Improve JIT code generation for greedy plus quantifier.
540    
541    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
542        1 to "aa" instead of to an empty string. The bug affected repeated groups
543        that could potentially match an empty string.
544    
545    37. Optimizing single character iterators in JIT.
546    
547    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
548        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
549        codepoints that are too big for the mode are faulted, and in a UTF mode,
550        disallowed codepoints are also faulted.
551    
552    39. If PCRE was compiled with UTF support, in three places in the DFA
553        matcher there was code that should only have been obeyed in UTF mode, but
554        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
555        processing when bytes with values greater than 127 were present. In 16-bit
556        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
557        both cases the values are those that cannot be the first data item in a UTF
558        character. The three items that might have provoked this were recursions,
559        possessively repeated groups, and atomic groups.
560    
561    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
562        and pcregrep, because some OS require shared objects to be explicitly
563        passed to ld, causing the link step to fail if they are not.
564    
565    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
566        mode, patterns that started with \h* or \R* might be incorrectly matched.
567    
568    
569    Version 8.30 04-February-2012
570    -----------------------------
571    
572    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
573        name is defined in ctype.h.
574    
575    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
576        only in quite long subpatterns.
577    
578    3.  Removed the function pcre_info(), which has been obsolete and deprecated
579        since it was replaced by pcre_fullinfo() in February 2000.
580    
581    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
582        match a (*MARK), and the match failed at the start of the subject, a
583        reference to memory before the start of the subject could occur. This bug
584        was introduced by fix 17 of release 8.21.
585    
586    5.  A reference to an unset group with zero minimum repetition was giving
587        totally wrong answers (in non-JavaScript-compatibility mode). For example,
588        /(another)?(\1?)test/ matched against "hello world test". This bug was
589        introduced in release 8.13.
590    
591    6.  Add support for 16-bit character strings (a large amount of work involving
592        many changes and refactorings).
593    
594    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
595        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
596        from a file.
597    
598    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
599        rounding is not applied in this particular case).
600    
601    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
602        if they appear, or are escaped, in patterns.
603    
604    10. Get rid of a number of -Wunused-but-set-variable warnings.
605    
606    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
607        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
608        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
609        also returns the mark "x". This bug applied to capturing parentheses,
610        non-capturing parentheses, and atomic parentheses. It also applied to some
611        assertions.
612    
613    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
614        information out of configure.ac instead of relying on pcre.h.generic, which
615        is not stored in the repository.
616    
617    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
618        -lreadline.
619    
620    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
621    
622    15. Applied Graycode's patch to put the top-level frame on the stack rather
623        than the heap when not using the stack for recursion. This gives a
624        performance improvement in many cases when recursion is not deep.
625    
626    16. Experimental code added to "pcretest -C" to output the stack frame size.
627    
628    
629    Version 8.21 12-Dec-2011
630    ------------------------
631    
632    1.  Updating the JIT compiler.
633    
634    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
635        are added as well.
636    
637    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
638        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
639        calling _pcre_jit_exec. Some extra comments are added.
640    
641    4.  (*MARK) settings inside atomic groups that do not contain any capturing
642        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
643        was introduced by change 18 for 8.20.
644    
645    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
646        ECMA-262 standard.
647    
648    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
649        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
650        This bug was probably introduced by change 9 of 8.13.
651    
652    7.  While fixing 6 above, I noticed that a number of other items were being
653        incorrectly rejected as "not fixed length". This arose partly because newer
654        opcodes had not been added to the fixed-length checking code. I have (a)
655        corrected the bug and added tests for these items, and (b) arranged for an
656        error to occur if an unknown opcode is encountered while checking for fixed
657        length instead of just assuming "not fixed length". The items that were
658        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
659        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
660        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
661    
662    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
663        being incorrectly compiled and would have given unpredicatble results.
664    
665    9.  A possessively repeated subpattern with minimum repeat count greater than
666        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
667        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
668        the first (A) could occur when it should not.
669    
670    10. Add a cast and remove a redundant test from the code.
671    
672    11. JIT should use pcre_malloc/pcre_free for allocation.
673    
674    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
675        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
676        is anything other than /usr, -L is still shown).
677    
678    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
679    
680    14. Perl does not support \N without a following name in a [] class; PCRE now
681        also gives an error.
682    
683    15. If a forward reference was repeated with an upper limit of around 2000,
684        it caused the error "internal error: overran compiling workspace". The
685        maximum number of forward references (including repeats) was limited by the
686        internal workspace, and dependent on the LINK_SIZE. The code has been
687        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
688        the default depends on LINK_SIZE. There is a new upper limit (for safety)
689        of around 200,000 forward references. While doing this, I also speeded up
690        the filling in of repeated forward references.
691    
692    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
693        incorrectly expecting the subject to contain another "a" after the start.
694    
695    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
696        in the match, the SKIP should be ignored. This was not happening; instead
697        the SKIP was being treated as NOMATCH. For patterns such as
698        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
699        tested.
700    
701    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
702        now much more compatible with Perl, in particular in cases where the result
703        is a non-match for a non-anchored pattern. For example, if
704        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
705        "m", where previously it did not return a name. A side effect of this
706        change is that for partial matches, the last encountered mark name is
707        returned, as for non matches. A number of tests that were previously not
708        Perl-compatible have been moved into the Perl-compatible test files. The
709        refactoring has had the pleasing side effect of removing one argument from
710        the match() function, thus reducing its stack requirements.
711    
712    19. If the /S+ option was used in pcretest to study a pattern using JIT,
713        subsequent uses of /S (without +) incorrectly behaved like /S+.
714    
715    21. Retrieve executable code size support for the JIT compiler and fixing
716        some warnings.
717    
718    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
719        not work when the shorter character appeared right at the end of the
720        subject string.
721    
722    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
723        systems.
724    
725    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
726        output it when the /M option is used in pcretest.
727    
728    25. The CheckMan script was not being included in the distribution. Also, added
729        an explicit "perl" to run Perl scripts from the PrepareRelease script
730        because this is reportedly needed in Windows.
731    
732    26. If study data was being save in a file and studying had not found a set of
733        "starts with" bytes for the pattern, the data written to the file (though
734        never used) was taken from uninitialized memory and so caused valgrind to
735        complain.
736    
737    27. Updated RunTest.bat as provided by Sheri Pierce.
738    
739    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
740    
741    29. Computation of memory usage for the table of capturing group names was
742        giving an unnecessarily large value.
743    
744    
745    Version 8.20 21-Oct-2011
746    ------------------------
747    
748    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
749        a POSIX class. After further experiments with Perl, which convinced me that
750        Perl has bugs and confusions, a closing square bracket is no longer allowed
751        in a POSIX name. This bug also affected patterns with classes that started
752        with full stops.
753    
754    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
755        captured substring, but while checking the failing first alternative,
756        substring 1 is temporarily captured. If the output vector supplied to
757        pcre_exec() was not big enough for this capture, the yield of the function
758        was still zero ("insufficient space for captured substrings"). This cannot
759        be totally fixed without adding another stack variable, which seems a lot
760        of expense for a edge case. However, I have improved the situation in cases
761        such as /(a)(b)x|abc/ matched against "abc", where the return code
762        indicates that fewer than the maximum number of slots in the ovector have
763        been set.
764    
765    3.  Related to (2) above: when there are more back references in a pattern than
766        slots in the output vector, pcre_exec() uses temporary memory during
767        matching, and copies in the captures as far as possible afterwards. It was
768        using the entire output vector, but this conflicts with the specification
769        that only 2/3 is used for passing back captured substrings. Now it uses
770        only the first 2/3, for compatibility. This is, of course, another edge
771        case.
772    
773    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
774        main code base, and can be used by building with --enable-jit. When this is
775        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
776        runtime --no-jit option is given.
777    
778    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
779        ovector, the return from the function was zero, implying that there were
780        other matches that did not fit. The correct "exactly full" value is now
781        returned.
782    
783    6.  If a subpattern that was called recursively or as a subroutine contained
784        (*PRUNE) or any other control that caused it to give a non-standard return,
785        invalid errors such as "Error -26 (nested recursion at the same subject
786        position)" or even infinite loops could occur.
787    
788    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
789        computing the minimum length on reaching *ACCEPT, and so ended up with the
790        wrong value of 1 rather than 0. Further investigation indicates that
791        computing a minimum subject length in the presence of *ACCEPT is difficult
792        (think back references, subroutine calls), and so I have changed the code
793        so that no minimum is registered for a pattern that contains *ACCEPT.
794    
795    8.  If (*THEN) was present in the first (true) branch of a conditional group,
796        it was not handled as intended. [But see 16 below.]
797    
798    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
799        Sheri Pierce.
800    
801    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
802        the first byte in a match must be "a".
803    
804    11. Change 17 for 8.13 increased the recursion depth for patterns like
805        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
806        pattern contains any instances of (*THEN). If it does not, the old
807        optimizations are restored. It would be nice to do this on a per-group
808        basis, but at the moment that is not feasible.
809    
810    12. In some environments, the output of pcretest -C is CRLF terminated. This
811        broke RunTest's code that checks for the link size. A single white space
812        character after the value is now allowed for.
813    
814    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
815        For "fr", it uses the Windows-specific input and output files.
816    
817    14. If (*THEN) appeared in a group that was called recursively or as a
818        subroutine, it did not work as intended. [But see next item.]
819    
820    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
821        pattern fragments (but not containing any | characters). If A and B are
822        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
823        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
824        D. In other words, Perl considers parentheses that do not contain any |
825        characters to be part of a surrounding alternative, whereas PCRE was
826        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
827        differently. PCRE now behaves in the same way as Perl, except in the case
828        of subroutine/recursion calls such as (?1) which have in any case always
829        been different (but PCRE had them first :-).
830    
831    16. Related to 15 above: Perl does not treat the | in a conditional group as
832        creating alternatives. Such a group is treated in the same way as an
833        ordinary group without any | characters when processing (*THEN). PCRE has
834        been changed to match Perl's behaviour.
835    
836    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
837        RunGrepTest script failed.
838    
839    18. Change 22 for version 13 caused atomic groups to use more stack. This is
840        inevitable for groups that contain captures, but it can lead to a lot of
841        stack use in large patterns. The old behaviour has been restored for atomic
842        groups that do not contain any capturing parentheses.
843    
844    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
845        suppress the check for a minimum subject length at run time. (If it was
846        given to pcre_exec() or pcre_dfa_exec() it did work.)
847    
848    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
849        fail to work when decoding hex characters in data strings in EBCDIC
850        environments.
851    
852    21. It appears that in at least one Mac OS environment, the isxdigit() function
853        is implemented as a macro that evaluates to its argument more than once,
854        contravening the C 90 Standard (I haven't checked a later standard). There
855        was an instance in pcretest which caused it to go wrong when processing
856        \x{...} escapes in subject strings. The has been rewritten to avoid using
857        things like p++ in the argument of isxdigit().
858    
859    
860    Version 8.13 16-Aug-2011
861    ------------------------
862    
863    1.  The Unicode data tables have been updated to Unicode 6.0.0.
864    
865    2.  Two minor typos in pcre_internal.h have been fixed.
866    
867    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
868        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
869        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
870    
871    4.  There were a number of related bugs in the code for matching backrefences
872        caselessly in UTF-8 mode when codes for the characters concerned were
873        different numbers of bytes. For example, U+023A and U+2C65 are an upper
874        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
875        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
876        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
877        2-byte code at the end of the subject (it thought there wasn't enough data
878        left).
879    
880    5.  Comprehensive information about what went wrong is now returned by
881        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
882        as the output vector has at least 2 elements. The offset of the start of
883        the failing character and a reason code are placed in the vector.
884    
885    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
886        now returned is for the first byte of the failing character, instead of the
887        last byte inspected. This is an incompatible change, but I hope it is small
888        enough not to be a problem. It makes the returned offset consistent with
889        pcre_exec() and pcre_dfa_exec().
890    
891    7.  pcretest now gives a text phrase as well as the error number when
892        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
893        failure, the offset and reason code are output.
894    
895    8.  When \R was used with a maximizing quantifier it failed to skip backwards
896        over a \r\n pair if the subsequent match failed. Instead, it just skipped
897        back over a single character (\n). This seems wrong (because it treated the
898        two characters as a single entity when going forwards), conflicts with the
899        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
900        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
901        has been changed.
902    
903    9.  Some internal refactoring has changed the processing so that the handling
904        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
905        time (the PCRE_DOTALL option was changed this way some time ago: version
906        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
907        which was always a bit of a fudge. It also means that there is one less
908        argument for the match() function, which reduces its stack requirements
909        slightly. This change also fixes an incompatibility with Perl: the pattern
910        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
911    
912    10. More internal refactoring has drastically reduced the number of recursive
913        calls to match() for possessively repeated groups such as (abc)++ when
914        using pcre_exec().
915    
916    11. While implementing 10, a number of bugs in the handling of groups were
917        discovered and fixed:
918    
919        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
920        (a|)*(?1) gave a compile-time internal error.
921        ((a|)+)+  did not notice that the outer group could match an empty string.
922        (^a|^)+   was not marked as anchored.
923        (.*a|.*)+ was not marked as matching at start or after a newline.
924    
925    12. Yet more internal refactoring has removed another argument from the match()
926        function. Special calls to this function are now indicated by setting a
927        value in a variable in the "match data" data block.
928    
929    13. Be more explicit in pcre_study() instead of relying on "default" for
930        opcodes that mean there is no starting character; this means that when new
931        ones are added and accidentally left out of pcre_study(), testing should
932        pick them up.
933    
934    14. The -s option of pcretest has been documented for ages as being an old
935        synonym of -m (show memory usage). I have changed it to mean "force study
936        for every regex", that is, assume /S for every regex. This is similar to -i
937        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
938        using it. It makes it easier to run collections of tests with and without
939        study enabled, and thereby test pcre_study() more easily. All the standard
940        tests are now run with and without -s (but some patterns can be marked as
941        "never study" - see 20 below).
942    
943    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
944        restoration of the capturing data to the outer values was not happening
945        correctly.
946    
947    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
948        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
949        pattern had matched an empty string, and so incorrectly returned a no
950        match.
951    
952    17. There was optimizing code for the last branch of non-capturing parentheses,
953        and also for the obeyed branch of a conditional subexpression, which used
954        tail recursion to cut down on stack usage. Unfortunately, now that there is
955        the possibility of (*THEN) occurring in these branches, tail recursion is
956        no longer possible because the return has to be checked for (*THEN). These
957        two optimizations have therefore been removed. [But see 8.20/11 above.]
958    
959    18. If a pattern containing \R was studied, it was assumed that \R always
960        matched two bytes, thus causing the minimum subject length to be
961        incorrectly computed because \R can also match just one byte.
962    
963    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
964        was incorrectly computed.
965    
966    20. If /S is present twice on a test pattern in pcretest input, it now
967        *disables* studying, thereby overriding the use of -s on the command line
968        (see 14 above). This is necessary for one or two tests to keep the output
969        identical in both cases.
970    
971    21. When (*ACCEPT) was used in an assertion that matched an empty string and
972        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
973    
974    22. When an atomic group that contained a capturing parenthesis was
975        successfully matched, but the branch in which it appeared failed, the
976        capturing was not being forgotten if a higher numbered group was later
977        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
978        group 1 to "a", when in fact it should be unset. This applied to multi-
979        branched capturing and non-capturing groups, repeated or not, and also to
980        positive assertions (capturing in negative assertions does not happen
981        in PCRE) and also to nested atomic groups.
982    
983    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
984        subject after a captured substring, to make it easier to tell which of a
985        number of identical substrings has been captured.
986    
987    24. The way atomic groups are processed by pcre_exec() has been changed so that
988        if they are repeated, backtracking one repetition now resets captured
989        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
990        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
991        "aaa". Previously, it would have been "a". As part of this code
992        refactoring, the way recursive calls are handled has also been changed.
993    
994    25. If an assertion condition captured any substrings, they were not passed
995        back unless some other capturing happened later. For example, if
996        (?(?=(a))a) was matched against "a", no capturing was returned.
997    
998    26. When studying a pattern that contained subroutine calls or assertions,
999        the code for finding the minimum length of a possible match was handling
1000        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
1001        group 1 called group 2 while simultaneously a separate group 2 called group
1002        1). A stack overflow occurred in this case. I have fixed this by limiting
1003        the recursion depth to 10.
1004    
1005    27. Updated RunTest.bat in the distribution to the version supplied by Tom
1006        Fortmann. This supports explicit test numbers on the command line, and has
1007        argument validation and error reporting.
1008    
1009    28. An instance of \X with an unlimited repeat could fail if at any point the
1010        first character it looked at was a mark character.
1011    
1012    29. Some minor code refactoring concerning Unicode properties and scripts
1013        should reduce the stack requirement of match() slightly.
1014    
1015    30. Added the '=' option to pcretest to check the setting of unused capturing
1016        slots at the end of the pattern, which are documented as being -1, but are
1017        not included in the return count.
1018    
1019    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
1020        compiled something random. Now it gives a compile-time error (as does
1021        Perl).
1022    
1023    32. A *MARK encountered during the processing of a positive assertion is now
1024        recorded and passed back (compatible with Perl).
1025    
1026    33. If --only-matching or --colour was set on a pcregrep call whose pattern
1027        had alternative anchored branches, the search for a second match in a line
1028        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
1029        matched the line "0102" twice. The same bug affected patterns that started
1030        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
1031        twice.
1032    
1033    34. Previously, PCRE did not allow quantification of assertions. However, Perl
1034        does, and because of capturing effects, quantifying parenthesized
1035        assertions may at times be useful. Quantifiers are now allowed for
1036        parenthesized assertions.
1037    
1038    35. A minor code tidy in pcre_compile() when checking options for \R usage.
1039    
1040    36. \g was being checked for fancy things in a character class, when it should
1041        just be a literal "g".
1042    
1043    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
1044        appearance of a nested POSIX class supersedes an apparent external class.
1045        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
1046        unescaped square brackets may also appear as part of class names. For
1047        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
1048        more like Perl. (But see 8.20/1 above.)
1049    
1050    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
1051        was because it thought it was \N{name}, which is not supported).
1052    
1053    39. Add minix to OS list not supporting the -S option in pcretest.
1054    
1055    40. PCRE tries to detect cases of infinite recursion at compile time, but it
1056        cannot analyze patterns in sufficient detail to catch mutual recursions
1057        such as ((?1))((?2)). There is now a runtime test that gives an error if a
1058        subgroup is called recursively as a subpattern for a second time at the
1059        same position in the subject string. In previous releases this might have
1060        been caught by the recursion limit, or it might have run out of stack.
1061    
1062    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
1063        happen only once. PCRE was, however incorrectly giving a compile time error
1064        "recursive call could loop indefinitely" because it cannot analyze the
1065        pattern in sufficient detail. The compile time test no longer happens when
1066        PCRE is compiling a conditional subpattern, but actual runaway loops are
1067        now caught at runtime (see 40 above).
1068    
1069    42. It seems that Perl allows any characters other than a closing parenthesis
1070        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
1071        has been changed to be the same.
1072    
1073    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
1074        as not to get warnings when autogen.sh is called. Also changed
1075        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
1076    
1077    44. To help people who use pcregrep to scan files containing exceedingly long
1078        lines, the following changes have been made:
1079    
1080        (a) The default value of the buffer size parameter has been increased from
1081            8K to 20K. (The actual buffer used is three times this size.)
1082    
1083        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
1084            PCRE is built.
1085    
1086        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
1087            to be set at run time.
1088    
1089        (d) Numerical values in pcregrep options can be followed by K or M, for
1090            example --buffer-size=50K.
1091    
1092        (e) If a line being scanned overflows pcregrep's buffer, an error is now
1093            given and the return code is set to 2.
1094    
1095    45. Add a pointer to the latest mark to the callout data block.
1096    
1097    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
1098        partial match of an empty string instead of no match. This was specific to
1099        the use of ".".
1100    
1101    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
1102        complete match instead of a partial match. This bug was dependent on both
1103        the PCRE_UTF8 and PCRE_DOTALL options being set.
1104    
1105    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1106        starting byte set, because \b was not being ignored.
1107    
1108    
1109    Version 8.12 15-Jan-2011
1110    ------------------------
1111    
1112    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1113        checks for such things as part of the documentation building process.
1114    
1115    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1116        --match-limit and --recursion-limit options (added for 8.11). In
1117        particular, this made one of the standard tests fail. (The integer value
1118        went into the wrong half of a long int.)
1119    
1120    3.  If the --colour option was given to pcregrep with -v (invert match), it
1121        did strange things, either producing crazy output, or crashing. It should,
1122        of course, ignore a request for colour when reporting lines that do not
1123        match.
1124    
1125    4.  Another pcregrep bug caused similar problems if --colour was specified with
1126        -M (multiline) and the pattern match finished with a line ending.
1127    
1128    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1129        matched in multiline mode, the following line was shown as part of the
1130        match. This seems wrong, so I have changed it.
1131    
1132    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1133        the check for further matches in the same line (so they could be coloured)
1134        to overrun the end of the current line. If another match was found, it was
1135        incorrectly shown (and then shown again when found in the next line).
1136    
1137    7.  If pcregrep was compiled under Windows, there was a reference to the
1138        function pcregrep_exit() before it was defined. I am assuming this was
1139        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1140        reported by a user. I've moved the definition above the reference.
1141    
1142    
1143    Version 8.11 10-Dec-2010
1144    ------------------------
1145    
1146    1.  (*THEN) was not working properly if there were untried alternatives prior
1147        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1148        backtracked to try for "b" instead of moving to the next alternative branch
1149        at the same level (in this case, to look for "c"). The Perl documentation
1150        is clear that when (*THEN) is backtracked onto, it goes to the "next
1151        alternative in the innermost enclosing group".
1152    
1153    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1154        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1155        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1156        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1157        (*THEN).
1158    
1159    3.  If \s appeared in a character class, it removed the VT character from
1160        the class, even if it had been included by some previous item, for example
1161        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1162        of \s, but is part of the POSIX "space" class.)
1163    
1164    4.  A partial match never returns an empty string (because you can always
1165        match an empty string at the end of the subject); however the checking for
1166        an empty string was starting at the "start of match" point. This has been
1167        changed to the "earliest inspected character" point, because the returned
1168        data for a partial match starts at this character. This means that, for
1169        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1170        (previously it gave "no match").
1171    
1172    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1173        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1174        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1175        has an implication that the given string is incomplete (because a partial
1176        match is preferred over a full match). For this reason, these items now
1177        give a partial match in this situation. [Aside: previously, the one case
1178        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1179        match rather than a full match, which was wrong by the old rules, but is
1180        now correct.]
1181    
1182    6.  There was a bug in the handling of #-introduced comments, recognized when
1183        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1184        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1185        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1186        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1187        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1188        places in pcre_compile().
1189    
1190    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1191        comments when looking ahead for named forward references to subpatterns,
1192        the only newline sequence it recognized was NL. It now handles newlines
1193        according to the set newline convention.
1194    
1195    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1196        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1197        cater for a lack of strerror(). These oversights have been fixed.
1198    
1199    9.  Added --match-limit and --recursion-limit to pcregrep.
1200    
1201    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1202    
1203    11. When the -o option was used, pcregrep was setting a return code of 1, even
1204        when matches were found, and --line-buffered was not being honoured.
1205    
1206    12. Added an optional parentheses number to the -o and --only-matching options
1207        of pcregrep.
1208    
1209    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1210        can match an empty string. The code to do it in pcretest and pcredemo
1211        needed fixing:
1212    
1213        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1214            only one byte after an empty string match just before CRLF (this case
1215            just got forgotten; "any" and "anycrlf" were OK).
1216    
1217        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1218            mode when an empty string match preceded an ASCII character followed by
1219            a non-ASCII character. (The code for advancing by one character rather
1220            than one byte was nonsense.)
1221    
1222        (c) The pcredemo.c sample program did not have any code at all to handle
1223            the cases when CRLF is a valid newline sequence.
1224    
1225    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1226        as a starting offset was within the subject string. There is now a new
1227        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1228        negative or greater than the length of the string. In order to test this,
1229        pcretest is extended to allow the setting of negative starting offsets.
1230    
1231    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1232        starting offset points to the beginning of a UTF-8 character was
1233        unnecessarily clumsy. I tidied it up.
1234    
1235    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1236        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1237    
1238    17. Nobody had reported that the --include_dir option, which was added in
1239        release 7.7 should have been called --include-dir (hyphen, not underscore)
1240        for compatibility with GNU grep. I have changed it to --include-dir, but
1241        left --include_dir as an undocumented synonym, and the same for
1242        --exclude-dir, though that is not available in GNU grep, at least as of
1243        release 2.5.4.
1244    
1245    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1246        characters from a string of bytes) have been redefined so as not to use
1247        loops, in order to improve performance in some environments. At the same
1248        time, I abstracted some of the common code into auxiliary macros to save
1249        repetition (this should not affect the compiled code).
1250    
1251    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1252        compile-time error is now given if \c is not followed by an ASCII
1253        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1254        different, and any byte value is allowed.)
1255    
1256    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1257        START_OPTIMIZE option, which is now allowed at compile time - but just
1258        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1259        to pcregrep and other applications that have no direct access to PCRE
1260        options. The new /Y option in pcretest sets this option when calling
1261        pcre_compile().
1262    
1263    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1264        back references. Groups containing recursive back references were forced to
1265        be atomic by that change, but in the case of named groups, the amount of
1266        memory required was incorrectly computed, leading to "Failed: internal
1267        error: code overflow". This has been fixed.
1268    
1269    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1270        pcretest.c, to avoid build problems in some Borland environments.
1271    
1272    
1273    Version 8.10 25-Jun-2010
1274    ------------------------
1275    
1276    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1277        THEN.
1278    
1279    2.  (*ACCEPT) was not working when inside an atomic group.
1280    
1281    3.  Inside a character class, \B is treated as a literal by default, but
1282        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1283        causes the error). The code is unchanged, but I tidied the documentation.
1284    
1285    4.  Inside a character class, PCRE always treated \R and \X as literals,
1286        whereas Perl faults them if its -w option is set. I have changed PCRE so
1287        that it faults them when PCRE_EXTRA is set.
1288    
1289    5.  Added support for \N, which always matches any character other than
1290        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1291    
1292    6.  When compiling pcregrep with newer versions of gcc which may have
1293        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1294        declared with attribute warn_unused_result" were given. Just casting the
1295        result to (void) does not stop the warnings; a more elaborate fudge is
1296        needed. I've used a macro to implement this.
1297    
1298    7.  Minor change to pcretest.c to avoid a compiler warning.
1299    
1300    8.  Added four artifical Unicode properties to help with an option to make
1301        \s etc use properties (see next item). The new properties are: Xan
1302        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1303    
1304    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1305        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1306        this option. Modified pcretest to add /W to test this facility. Added
1307        REG_UCP to make it available via the POSIX interface.
1308    
1309    10. Added --line-buffered to pcregrep.
1310    
1311    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1312        studied, and the match started with a letter with a code point greater than
1313        127 whose first byte was different to the first byte of the other case of
1314        the letter, the other case of this starting letter was not recognized
1315        (#976).
1316    
1317    12. If a pattern that was studied started with a repeated Unicode property
1318        test, for example, \p{Nd}+, there was the theoretical possibility of
1319        setting up an incorrect bitmap of starting bytes, but fortunately it could
1320        not have actually happened in practice until change 8 above was made (it
1321        added property types that matched character-matching opcodes).
1322    
1323    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1324        possible starting bytes for non-anchored patterns.
1325    
1326    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1327        \R, and also a number of cases that involve Unicode properties, both
1328        explicit and implicit when PCRE_UCP is set.
1329    
1330    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1331        input, it could crash or give wrong results if characters with values
1332        greater than 0xc0 were present in the subject string. (Detail: it assumed
1333        UTF-8 input when processing these items.)
1334    
1335    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1336        size_t is 64-bit (#991).
1337    
1338    17. Added a check for running out of memory when PCRE is compiled with
1339        --disable-stack-for-recursion (#990).
1340    
1341    18. If the last data line in a file for pcretest does not have a newline on
1342        the end, a newline was missing in the output.
1343    
1344    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1345        less than 128) in its various bitmaps. However, there is a facility for
1346        generating tables according to the current locale when PCRE is compiled. It
1347        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1348        characters, are recognized by isspace() and therefore were getting set in
1349        these tables, and indeed these tables seem to approximate to ISO 8859. This
1350        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1351        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1352        which of course cannot start UTF-8 characters. I have changed the code so
1353        that only real ASCII characters (less than 128) and the correct starting
1354        bytes for UTF-8 encodings are set for characters greater than 127 when in
1355        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1356        altogether.)
1357    
1358    20. Added the /T option to pcretest so as to be able to run tests with non-
1359        standard character tables, thus making it possible to include the tests
1360        used for 19 above in the standard set of tests.
1361    
1362    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1363        reference to a subpattern the other side of a comment that contains an
1364        opening parenthesis caused either an internal compiling error, or a
1365        reference to the wrong subpattern.
1366    
1367    
1368    Version 8.02 19-Mar-2010
1369    ------------------------
1370    
1371    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1372    
1373    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1374        configured.
1375    
1376    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1377        original author of that file, following a query about its status.
1378    
1379    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1380        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1381    
1382    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1383        quantifier applied to a forward-referencing subroutine call, could compile
1384        incorrect code or give the error "internal error: previously-checked
1385        referenced subpattern not found".
1386    
1387    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1388        variables to point to external functions. For these systems, therefore,
1389        pcre_malloc etc. are now initialized to local functions that call the
1390        relevant global functions.
1391    
1392    7.  There were two entries missing in the vectors called coptable and poptable
1393        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1394        I've fixed the data, and added a kludgy way of testing at compile time that
1395        the lengths are correct (equal to the number of opcodes).
1396    
1397    8.  Following on from 7, I added a similar kludge to check the length of the
1398        eint vector in pcreposix.c.
1399    
1400    9.  Error texts for pcre_compile() are held as one long string to avoid too
1401        much relocation at load time. To find a text, the string is searched,
1402        counting zeros. There was no check for running off the end of the string,
1403        which could happen if a new error number was added without updating the
1404        string.
1405    
1406    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1407    
1408    11. \K was not working if it appeared in an atomic group or in a group that
1409        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1410        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1411        the assertion is positive, but not if it is negative.
1412    
1413    12. Change 11 fortuitously reduced the size of the stack frame used in the
1414        "match()" function of pcre_exec.c by one pointer. Forthcoming
1415        implementation of support for (*MARK) will need an extra pointer on the
1416        stack; I have reserved it now, so that the stack frame size does not
1417        decrease.
1418    
1419    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1420        item in branch that calls a recursion is a subroutine call - as in the
1421        second branch in the above example - was incorrectly given the compile-
1422        time error "recursive call could loop indefinitely" because pcre_compile()
1423        was not correctly checking the subroutine for matching a non-empty string.
1424    
1425    14. The checks for overrunning compiling workspace could trigger after an
1426        overrun had occurred. This is a "should never occur" error, but it can be
1427        triggered by pathological patterns such as hundreds of nested parentheses.
1428        The checks now trigger 100 bytes before the end of the workspace.
1429    
1430    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1431    
1432    
1433    Version 8.01 19-Jan-2010
1434    ------------------------
1435    
1436    1.  If a pattern contained a conditional subpattern with only one branch (in
1437        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1438        computed the wrong minimum data length (which is of course zero for such
1439        subpatterns). This could cause incorrect "no match" results.
1440    
1441    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1442        the pattern is reset in the first branch, pcre_compile() failed with
1443        "internal error: code overflow at offset...". This happened only when
1444        the reset was to the original external option setting. (An optimization
1445        abstracts leading options settings into an external setting, which was the
1446        cause of this.)
1447    
1448    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1449        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1450        assertion pattern did not match (meaning that the assertion was true), it
1451        was incorrectly treated as false if the SKIP had been reached during the
1452        matching. This also applied to assertions used as conditions.
1453    
1454    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1455        assertion subpattern, including such a pattern used as a condition,
1456        unpredictable results occurred, instead of the error return
1457        PCRE_ERROR_DFA_UITEM.
1458    
1459    5.  The C++ GlobalReplace function was not working like Perl for the special
1460        situation when an empty string is matched. It now does the fancy magic
1461        stuff that is necessary.
1462    
1463    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1464        removed. (These were left over from very, very early versions of PCRE.)
1465    
1466    7.  Some cosmetic changes to the code to make life easier when compiling it
1467        as part of something else:
1468    
1469        (a) Change DEBUG to PCRE_DEBUG.
1470    
1471        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1472            called "current" as "current_branch", to prevent a collision with the
1473            Linux macro when compiled as a kernel module.
1474    
1475        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1476            prevent a collision with the Linux macro when compiled as a kernel
1477            module.
1478    
1479    8.  In pcre_compile() there are some checks for integer overflows that used to
1480        cast potentially large values to (double). This has been changed to that
1481        when building, a check for int64_t is made, and if it is found, it is used
1482        instead, thus avoiding the use of floating point arithmetic. (There is no
1483        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1484        double.
1485    
1486    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1487        2005 (difference between two addresses compared to an unsigned value).
1488    
1489    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1490        custom one, because of the following reported problem in Windows:
1491    
1492          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1493              under Win32.
1494          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1495              therefore missing the function definition.
1496          - The compiler thus generates a "C" signature for the test function.
1497          - The linker fails to find the "C" function.
1498          - PCRE fails to configure if asked to do so against libbz2.
1499    
1500    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1501        messages were output:
1502    
1503          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1504          rerunning libtoolize, to keep the correct libtool macros in-tree.
1505          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1506    
1507        I have done both of these things.
1508    
1509    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1510        most of the time, it *can* run out if it is given a pattern that contains a
1511        runaway infinite recursion. I updated the discussion in the pcrestack man
1512        page.
1513    
1514    13. Now that we have gone to the x.xx style of version numbers, the minor
1515        version may start with zero. Using 08 or 09 is a bad idea because users
1516        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1517        interpreted as invalid octal numbers. I've updated the previous comment in
1518        configure.ac, and also added a check that gives an error if 08 or 09 are
1519        used.
1520    
1521    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1522        causing partial matching to fail when the end of the subject matched \W
1523        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1524    
1525    15. There were some discrepancies between the declarations in pcre_internal.h
1526        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1527        their definitions. The declarations used "const uschar *" and the
1528        definitions used USPTR. Even though USPTR is normally defined as "const
1529        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1530        reported that: "This difference in casting confuses some C++ compilers, for
1531        example, SunCC recognizes above declarations as different functions and
1532        generates broken code for hbpcre." I have changed the declarations to use
1533        USPTR.
1534    
1535    16. GNU libtool is named differently on some systems. The autogen.sh script now
1536        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1537        (FreeBSD).
1538    
1539    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1540        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1541        comment: "Figure out how to create a longlong from a string: strtoll and
1542        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1543        instance, but it only takes 2 args instead of 3!"
1544    
1545    18. A subtle bug concerned with back references has been fixed by a change of
1546        specification, with a corresponding code fix. A pattern such as
1547        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1548        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1549        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1550        same bug. Such groups have to be quantified to be useful, or contained
1551        inside another quantified group. (If there's no repetition, the reference
1552        can never match.) The problem arises because, having left the group and
1553        moved on to the rest of the pattern, a later failure that backtracks into
1554        the group uses the captured value from the final iteration of the group
1555        rather than the correct earlier one. I have fixed this in PCRE by forcing
1556        any group that contains a reference to itself to be an atomic group; that
1557        is, there cannot be any backtracking into it once it has completed. This is
1558        similar to recursive and subroutine calls.
1559    
1560    
1561    Version 8.00 19-Oct-09
1562  ----------------------  ----------------------
1563    
1564  1.  The table for translating pcre_compile() error codes into POSIX error codes  1.  The table for translating pcre_compile() error codes into POSIX error codes
1565      was out-of-date, and there was no check on the pcre_compile() error code      was out-of-date, and there was no check on the pcre_compile() error code
1566      being within the table. This could lead to an OK return being given in      being within the table. This could lead to an OK return being given in
1567      error.      error.
1568    
1569  2.  Changed the call to open a subject file in pcregrep from fopen(pathname,  2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1570      "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests      "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1571      in a Windows environment.      in a Windows environment.
1572    
1573  3.  The pcregrep --count option prints the count for each file even when it is  3.  The pcregrep --count option prints the count for each file even when it is
1574      zero, as does GNU grep. However, pcregrep was also printing all files when      zero, as does GNU grep. However, pcregrep was also printing all files when
1575      --files-with-matches was added. Now, when both options are given, it prints      --files-with-matches was added. Now, when both options are given, it prints
1576      counts only for those files that have at least one match. (GNU grep just      counts only for those files that have at least one match. (GNU grep just
1577      prints the file name in this circumstance, but including the count seems      prints the file name in this circumstance, but including the count seems
1578      more useful - otherwise, why use --count?) Also ensured that the      more useful - otherwise, why use --count?) Also ensured that the
1579      combination -clh just lists non-zero counts, with no names.      combination -clh just lists non-zero counts, with no names.
1580    
1581  4.  The long form of the pcregrep -F option was incorrectly implemented as  4.  The long form of the pcregrep -F option was incorrectly implemented as
1582      --fixed_strings instead of --fixed-strings. This is an incompatible change,      --fixed_strings instead of --fixed-strings. This is an incompatible change,
1583      but it seems right to fix it, and I didn't think it was worth preserving      but it seems right to fix it, and I didn't think it was worth preserving
1584      the old behaviour.      the old behaviour.
1585    
1586  5.  The command line items --regex=pattern and --regexp=pattern were not  5.  The command line items --regex=pattern and --regexp=pattern were not
1587      recognized by pcregrep, which required --regex pattern or --regexp pattern      recognized by pcregrep, which required --regex pattern or --regexp pattern
1588      (with a space rather than an '='). The man page documented the '=' forms,      (with a space rather than an '='). The man page documented the '=' forms,
1589      which are compatible with GNU grep; these now work.      which are compatible with GNU grep; these now work.
1590    
1591  6.  No libpcreposix.pc file was created for pkg-config; there was just  6.  No libpcreposix.pc file was created for pkg-config; there was just
1592      libpcre.pc and libpcrecpp.pc. The omission has been rectified.      libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1593    
1594  7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size  7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1595      when UCP support is not needed, by modifying the Python script that      when UCP support is not needed, by modifying the Python script that
1596      generates it from Unicode data files. This should not matter if the module      generates it from Unicode data files. This should not matter if the module
1597      is correctly used as a library, but I received one complaint about 50K of      is correctly used as a library, but I received one complaint about 50K of
1598      unwanted data. My guess is that the person linked everything into his      unwanted data. My guess is that the person linked everything into his
1599      program rather than using a library. Anyway, it does no harm.      program rather than using a library. Anyway, it does no harm.
1600    
1601  8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger  8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1602      was a minimum greater than 1 for a wide character in a possessive      was a minimum greater than 1 for a wide character in a possessive
1603      repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8      repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1604      which had an unlimited repeat of a nested, fixed maximum repeat of a wide      which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1605      character. Chaos in the form of incorrect output or a compiling loop could      character. Chaos in the form of incorrect output or a compiling loop could
1606      result.      result.
1607    
1608  9.  The restrictions on what a pattern can contain when partial matching is  9.  The restrictions on what a pattern can contain when partial matching is
1609      requested for pcre_exec() have been removed. All patterns can now be      requested for pcre_exec() have been removed. All patterns can now be
1610      partially matched by this function. In addition, if there are at least two      partially matched by this function. In addition, if there are at least two
1611      slots in the offset vector, the offset of the earliest inspected character      slots in the offset vector, the offset of the earliest inspected character
1612      for the match and the offset of the end of the subject are set in them when      for the match and the offset of the end of the subject are set in them when
1613      PCRE_ERROR_PARTIAL is returned.      PCRE_ERROR_PARTIAL is returned.
1614    
1615  10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is  10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1616      synonymous with PCRE_PARTIAL, for backwards compatibility, and      synonymous with PCRE_PARTIAL, for backwards compatibility, and
1617      PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,      PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1618      and may be more useful for multi-segment matching, especially with      and may be more useful for multi-segment matching.
1619      pcre_exec().  
1620    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1621  11. Partial matching with pcre_exec() is now more intuitive. A partial match      used to be given if ever the end of the subject was reached; now it is
1622      used to be given if ever the end of the subject was reached; now it is      given only if matching could not proceed because another character was
1623      given only if matching could not proceed because another character was      needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1624      needed. This makes a difference in some odd cases such as Z(*FAIL) with the      string "Z", which now yields "no match" instead of "partial match". In the
1625      string "Z", which now yields "no match" instead of "partial match". In the      case of pcre_dfa_exec(), "no match" is given if every matching path for the
1626      case of pcre_dfa_exec(), "no match" is given if every matching path for the      final character ended with (*FAIL).
1627      final character ended with (*FAIL).  
   
1628  12. Restarting a match using pcre_dfa_exec() after a partial match did not work  12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1629      if the pattern had a "must contain" character that was already found in the      if the pattern had a "must contain" character that was already found in the
1630      earlier partial match, unless partial matching was again requested. For      earlier partial match, unless partial matching was again requested. For
1631      example, with the pattern /dog.(body)?/, the "must contain" character is      example, with the pattern /dog.(body)?/, the "must contain" character is
1632      "g". If the first part-match was for the string "dog", restarting with      "g". If the first part-match was for the string "dog", restarting with
1633      "sbody" failed. This bug has been fixed.      "sbody" failed. This bug has been fixed.
1634    
1635  13. The string returned by pcre_dfa_exec() after a partial match has been  13. The string returned by pcre_dfa_exec() after a partial match has been
1636      changed so that it starts at the first inspected character rather than the      changed so that it starts at the first inspected character rather than the
1637      first character of the match. This makes a difference only if the pattern      first character of the match. This makes a difference only if the pattern
1638      starts with a lookbehind assertion or \b or \B (\K is not supported by      starts with a lookbehind assertion or \b or \B (\K is not supported by
1639      pcre_dfa_exec()). It's an incompatible change, but it makes the two      pcre_dfa_exec()). It's an incompatible change, but it makes the two
1640      matching functions compatible, and I think it's the right thing to do.      matching functions compatible, and I think it's the right thing to do.
1641    
1642  14. Added a pcredemo man page, created automatically from the pcredemo.c file,  14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1643      so that the demonstration program is easily available in environments where      so that the demonstration program is easily available in environments where
1644      PCRE has not been installed from source.      PCRE has not been installed from source.
1645    
1646  15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,  15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1647      libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared      libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1648      library.      library.
1649    
1650  16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.  16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1651      It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it      It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1652      is not the first non-POSIX option to be added. Clearly some people find      is not the first non-POSIX option to be added. Clearly some people find
1653      these options useful.      these options useful.
1654    
1655  17. If a caller to the POSIX matching function regexec() passes a non-zero  17. If a caller to the POSIX matching function regexec() passes a non-zero
1656      value for nmatch with a NULL value for pmatch, the value of      value for nmatch with a NULL value for pmatch, the value of
1657      nmatch is forced to zero.      nmatch is forced to zero.
1658    
1659  18. RunGrepTest did not have a test for the availability of the -u option of  18. RunGrepTest did not have a test for the availability of the -u option of
1660      the diff command, as RunTest does. It now checks in the same way as      the diff command, as RunTest does. It now checks in the same way as
1661      RunTest, and also checks for the -b option.      RunTest, and also checks for the -b option.
1662    
1663  19. If an odd number of negated classes containing just a single character  19. If an odd number of negated classes containing just a single character
1664      interposed, within parentheses, between a forward reference to a named      interposed, within parentheses, between a forward reference to a named
1665      subpattern and the definition of the subpattern, compilation crashed with      subpattern and the definition of the subpattern, compilation crashed with
1666      an internal error, complaining that it could not find the referenced      an internal error, complaining that it could not find the referenced
1667      subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.      subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1668      [The bug was that it was starting one character too far in when skipping      [The bug was that it was starting one character too far in when skipping
1669      over the character class, thus treating the ] as data rather than      over the character class, thus treating the ] as data rather than
1670      terminating the class. This meant it could skip too much.]      terminating the class. This meant it could skip too much.]
1671    
1672  20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the  20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1673      /g option in pcretest when the pattern contains \K, which makes it possible      /g option in pcretest when the pattern contains \K, which makes it possible
1674      to have an empty string match not at the start, even when the pattern is      to have an empty string match not at the start, even when the pattern is
1675      anchored. Updated pcretest and pcredemo to use this option.      anchored. Updated pcretest and pcredemo to use this option.
1676    
1677  21. If the maximum number of capturing subpatterns in a recursion was greater  21. If the maximum number of capturing subpatterns in a recursion was greater
1678      than the maximum at the outer level, the higher number was returned, but      than the maximum at the outer level, the higher number was returned, but
1679      with unset values at the outer level. The correct (outer level) value is      with unset values at the outer level. The correct (outer level) value is
1680      now given.      now given.
1681    
1682  22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of  22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1683      PCRE did not set those parentheses (unlike Perl). I have now found a way to      PCRE did not set those parentheses (unlike Perl). I have now found a way to
1684      make it do so. The string so far is captured, making this feature      make it do so. The string so far is captured, making this feature
1685      compatible with Perl.      compatible with Perl.
1686    
1687    23. The tests have been re-organized, adding tests 11 and 12, to make it
1688        possible to check the Perl 5.10 features against Perl 5.10.
1689    
1690    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1691        pattern matches a fixed length string. PCRE did not allow this; now it
1692        does. Neither allows recursion.
1693    
1694    25. I finally figured out how to implement a request to provide the minimum
1695        length of subject string that was needed in order to match a given pattern.
1696        (It was back references and recursion that I had previously got hung up
1697        on.) This code has now been added to pcre_study(); it finds a lower bound
1698        to the length of subject needed. It is not necessarily the greatest lower
1699        bound, but using it to avoid searching strings that are too short does give
1700        some useful speed-ups. The value is available to calling programs via
1701        pcre_fullinfo().
1702    
1703    26. While implementing 25, I discovered to my embarrassment that pcretest had
1704        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1705        study optimizations had never been tested with that matching function.
1706        Oops. What is worse, even when it was passed study data, there was a bug in
1707        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1708        were also very few tests of studied patterns with pcre_dfa_exec().
1709    
1710    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1711        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1712        on the other side of the coin, they are no longer allowed to have different
1713        names, because these cannot be distinguished in PCRE, and this has caused
1714        confusion. (This is a difference from Perl.)
1715    
1716    28. When duplicate subpattern names are present (necessarily with different
1717        numbers, as required by 27 above), and a test is made by name in a
1718        conditional pattern, either for a subpattern having been matched, or for
1719        recursion in such a pattern, all the associated numbered subpatterns are
1720        tested, and the overall condition is true if the condition is true for any
1721        one of them. This is the way Perl works, and is also more like the way
1722        testing by number works.
1723    
1724    
1725  Version 7.9 11-Apr-09  Version 7.9 11-Apr-09
1726  ---------------------  ---------------------
# Line 163  Version 7.9 11-Apr-09 Line 1756  Version 7.9 11-Apr-09
1756  7.  A pattern that could match an empty string could cause pcregrep to loop; it  7.  A pattern that could match an empty string could cause pcregrep to loop; it
1757      doesn't make sense to accept an empty string match in pcregrep, so I have      doesn't make sense to accept an empty string match in pcregrep, so I have
1758      locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this      locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1759      seems to be how GNU grep behaves.      seems to be how GNU grep behaves. [But see later change 40 for release
1760        8.33.]
1761    
1762  8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at  8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1763      start or after a newline", because the conditional assertion was not being      start or after a newline", because the conditional assertion was not being
# Line 406  Version 7.7 07-May-08 Line 2000  Version 7.7 07-May-08
2000      containing () gave an internal compiling error instead of "reference to      containing () gave an internal compiling error instead of "reference to
2001      non-existent subpattern". Fortunately, when the pattern did exist, the      non-existent subpattern". Fortunately, when the pattern did exist, the
2002      compiled code was correct. (When scanning forwards to check for the      compiled code was correct. (When scanning forwards to check for the
2003      existencd of the subpattern, it was treating the data ']' as terminating      existence of the subpattern, it was treating the data ']' as terminating
2004      the class, so got the count wrong. When actually compiling, the reference      the class, so got the count wrong. When actually compiling, the reference
2005      was subsequently set up correctly.)      was subsequently set up correctly.)
2006    

Legend:
Removed from v.447  
changed lines
  Added in v.1362

  ViewVC Help
Powered by ViewVC 1.1.5