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

Diff of /code/trunk/ChangeLog

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

revision 200 by ph10, Wed Aug 1 09:10:40 2007 UTC revision 1370 by ph10, Wed Oct 9 10:18:26 2013 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 7.3 05-Jul-07  Version 8.34 xx-xxxx-201x
5    --------------------------
6    
7    1.  Add pcre[16|32]_jit_free_unused_memory to forcibly free unused JIT
8        executable memory. Patch inspired by Carsten Klein.
9    
10    2.  ./configure --enable-coverage defined SUPPORT_GCOV in config.h, although
11        this macro is never tested and has no effect, because the work to support
12        coverage involves only compiling and linking options and special targets in
13        the Makefile. The comment in config.h implied that defining the macro would
14        enable coverage support, which is totally false. There was also support for
15        setting this macro in the CMake files (my fault, I just copied it from
16        configure). SUPPORT_GCOV has now been removed.
17    
18    3.  Make a small performance improvement in strlen16() and strlen32() in
19        pcretest.
20    
21    4.  Change 36 for 8.33 left some unreachable statements in pcre_exec.c,
22        detected by the Solaris compiler (gcc doesn't seem to be able to diagnose
23        these cases). There was also one in pcretest.c.
24    
25    5.  Cleaned up a "may be uninitialized" compiler warning in pcre_exec.c.
26    
27    6.  In UTF mode, the code for checking whether a group could match an empty
28        string (which is used for indefinitely repeated groups to allow for
29        breaking an infinite loop) was broken when the group contained a repeated
30        negated single-character class with a character that occupied more than one
31        data item and had a minimum repetition of zero (for example, [^\x{100}]* in
32        UTF-8 mode). The effect was undefined: the group might or might not be
33        deemed as matching an empty string, or the program might have crashed.
34    
35    7.  The code for checking whether a group could match an empty string was not
36        recognizing that \h, \H, \v, \V, and \R must match a character.
37    
38    8.  Implemented PCRE_INFO_MATCH_EMPTY, which yields 1 if the pattern can match
39        an empty string. If it can, pcretest shows this in its information output.
40    
41    9.  Fixed two related bugs that applied to Unicode extended grapheme clusters
42        that were repeated with a maximizing qualifier (e.g. \X* or \X{2,5}) when
43        matched by pcre_exec() without using JIT:
44    
45        (a) If the rest of the pattern did not match after a maximal run of
46            grapheme clusters, the code for backing up to try with fewer of them
47            did not always back up over a full grapheme when characters that do not
48            have the modifier quality were involved, e.g. Hangul syllables.
49    
50        (b) If the match point in a subject started with modifier character, and
51            there was no match, the code could incorrectly back up beyond the match
52            point, and potentially beyond the first character in the subject,
53            leading to a segfault or an incorrect match result.
54    
55    10. A conditional group with an assertion condition could lead to PCRE
56        recording an incorrect first data item for a match if no other first data
57        item was recorded. For example, the pattern (?(?=ab)ab) recorded "a" as a
58        first data item, and therefore matched "ca" after "c" instead of at the
59        start.
60    
61    11. Change 40 for 8.33 (allowing pcregrep to find empty strings) showed up a
62        bug that caused the command "echo a | ./pcregrep -M '|a'" to loop.
63    
64    12. The source of pcregrep now includes z/OS-specific code so that it can be
65        compiled for z/OS as part of the special z/OS distribution.
66    
67    13. Added the -T and -TM options to pcretest.
68    
69    14. The code in pcre_compile.c for creating the table of named capturing groups
70        has been refactored. Instead of creating the table dynamically during the
71        actual compiling pass, the information is remembered during the pre-compile
72        pass (on the stack unless there are more than 20 named groups, in which
73        case malloc() is used) and the whole table is created before the actual
74        compile happens. This has simplified the code (it is now nearly 150 lines
75        shorter) and prepared the way for better handling of references to groups
76        with duplicate names.
77    
78    15. A back reference to a named subpattern when there is more than one of the
79        same name now checks them in the order in which they appear in the pattern.
80        The first one that is set is used for the reference. Previously only the
81        first one was inspected. This change makes PCRE more compatible with Perl.
82    
83    16. Unicode character properties were updated from Unicode 6.3.0.
84    
85    17. The compile-time code for auto-possessification has been refactored, based
86        on a patch by Zoltan Herczeg. It now happens after instead of during
87        compilation. The code is cleaner, and more cases are handled. The option
88        PCRE_NO_AUTO_POSSESSIFY is added for testing purposes, and the -O and /O
89        options in pcretest are provided to set it.
90    
91    18. The character VT has been added to the set of characters that match \s and
92        are generally treated as white space, following this same change in Perl
93        5.18. There is now no difference between "Perl space" and "POSIX space".
94    
95    19. The code for checking named groups as conditions, either for being set or
96        for being recursed, has been refactored (this is related to 14 and 15
97        above). Processing unduplicated named groups should now be as fast at
98        numerical groups, and processing duplicated groups should be faster than
99        before.
100    
101    20. Two patches to the CMake build system, by Alexander Barkov:
102    
103          (1) Replace the "source" command by "." in CMakeLists.txt because
104              "source" is a bash-ism.
105    
106          (2) Add missing HAVE_STDINT_H and HAVE_INTTYPES_H to config-cmake.h.in;
107              without these the CMake build does not work on Solaris.
108    
109    21. Perl has changed its handling of \8 and \9. If there is no previously
110        encountered capturing group of those numbers, they are treated as the
111        literal characters 8 and 9 instead of a binary zero followed by the
112        literals. PCRE now does the same.
113    
114    22. Following Perl, added \o{} to specify codepoints in octal, making it
115        possible to specify values greater than 0777 and also making them
116        unambiguous.
117    
118    23. Perl now gives an error for missing closing braces after \x{... instead of
119        treating the string as literal. PCRE now does the same.
120    
121    24. RunTest used to grumble if an inappropriate test was selected explicitly,
122        but just skip it when running all tests. This make it awkward to run ranges
123        of tests when one of them was inappropriate. Now it just skips any
124        inappropriate tests, as it always did when running all tests.
125    
126    
127    
128    Version 8.33 28-May-2013
129    ------------------------
130    
131    1.  Added 'U' to some constants that are compared to unsigned integers, to
132        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
133        variables that are added to signed variables, to ensure the result is
134        signed and can be negated.
135    
136    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
137        CMake config files.
138    
139    3.  Revise the creation of config.h.generic so that all boolean macros are
140        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
141        overriding via -D on the command line possible.
142    
143    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
144        to unsigned int is reported to make a quite noticeable speed difference in
145        a specific Windows environment. Testing on Linux did also appear to show
146        some benefit (and it is clearly not harmful). Also fixed the definition of
147        Xop which should be unsigned.
148    
149    5.  Related to (4), changing the definition of the intermediate variable cc
150        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
151        improvements.
152    
153    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
154        unnecessary spaces.
155    
156    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
157        and later.
158    
159    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
160    
161    9.  Optimizing fast_forward_start_bits in JIT.
162    
163    10. Adding support for callouts in JIT, and fixing some issues revealed
164        during this work. Namely:
165    
166        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
167    
168        (b) Minimum length was not checked before the matching is started.
169    
170    11. The value of capture_last that is passed to callouts was incorrect in some
171        cases when there was a capture on one path that was subsequently abandoned
172        after a backtrack. Also, the capture_last value is now reset after a
173        recursion, since all captures are also reset in this case.
174    
175    12. The interpreter no longer returns the "too many substrings" error in the
176        case when an overflowing capture is in a branch that is subsequently
177        abandoned after a backtrack.
178    
179    13. In the pathological case when an offset vector of size 2 is used, pcretest
180        now prints out the matched string after a yield of 0 or 1.
181    
182    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
183        Only supported by the JIT compiler at the moment.
184    
185    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
186    
187    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
188        offset of the starting point of the matching process, provided the offsets
189        vector is large enough.
190    
191    17. The \A escape now records a lookbehind value of 1, though its execution
192        does not actually inspect the previous character. This is to ensure that,
193        in partial multi-segment matching, at least one character from the old
194        segment is retained when a new segment is processed. Otherwise, if there
195        are no lookbehinds in the pattern, \A might match incorrectly at the start
196        of a new segment.
197    
198    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
199    
200    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
201        gives some modest performance improvement in 8-bit mode.
202    
203    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
204        be expressed in certain programming languages using Universal Character
205        Names.
206    
207    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
208        which points out that "non characters" are not "characters that may not
209        appear in Unicode strings" but rather "characters that are reserved for
210        internal use and have only local meaning".
211    
212    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
213        there was a conditional group that depended on an assertion, if the
214        assertion was false, the callout that immediately followed the alternation
215        in the condition was skipped when pcre_exec() was used for matching.
216    
217    23. Allow an explicit callout to be inserted before an assertion that is the
218        condition for a conditional group, for compatibility with automatic
219        callouts, which always insert a callout at this point.
220    
221    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
222        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
223    
224    25. (*PRUNE) is now supported by the JIT compiler.
225    
226    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
227    
228    27. Fix the case where there are two or more SKIPs with arguments that may be
229        ignored.
230    
231    28. (*SKIP) is now supported by the JIT compiler.
232    
233    29. (*THEN) is now supported by the JIT compiler.
234    
235    30. Update RunTest with additional test selector options.
236    
237    31. The way PCRE handles backtracking verbs has been changed in two ways.
238    
239        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
240        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
241        backtracking. In some cases this makes it more Perl-compatible, but Perl's
242        rather obscure rules do not always do the same thing.
243    
244        (2) Previously, backtracking verbs were confined within assertions. This is
245        no longer the case for positive assertions, except for (*ACCEPT). Again,
246        this sometimes improves Perl compatibility, and sometimes does not.
247    
248    32. A number of tests that were in test 2 because Perl did things differently
249        have been moved to test 1, because either Perl or PCRE has changed, and
250        these tests are now compatible.
251    
252    32. Backtracking control verbs are now handled in the same way in JIT and
253        interpreter.
254    
255    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
256        contained a forward subroutine reference caused a compile error.
257    
258    34. Auto-detect and optimize limited repetitions in JIT.
259    
260    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
261        blocking (*UTF) etc.
262    
263    36. In the interpreter, maximizing pattern repetitions for characters and
264        character types now use tail recursion, which reduces stack usage.
265    
266    37. The value of the max lookbehind was not correctly preserved if a compiled
267        and saved regex was reloaded on a host of different endianness.
268    
269    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
270        of the compiled pattern block, expand the flags field from 16 to 32 bits
271        because it was almost full.
272    
273    39. Try madvise first before posix_madvise.
274    
275    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
276        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
277        original change locked out all matches of empty strings. This has been
278        changed so that one match of an empty string per line is recognized.
279        Subsequent searches on the same line (for colouring or for --only-matching,
280        for example) do not recognize empty strings.
281    
282    41. Applied a user patch to fix a number of spelling mistakes in comments.
283    
284    42. Data lines longer than 65536 caused pcretest to crash.
285    
286    43. Clarified the data type for length and startoffset arguments for pcre_exec
287        and pcre_dfa_exec in the function-specific man pages, where they were
288        explicitly stated to be in bytes, never having been updated. I also added
289        some clarification to the pcreapi man page.
290    
291    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
292        a segmentation fault.
293    
294    
295    Version 8.32 30-November-2012
296    -----------------------------
297    
298    1.  Improved JIT compiler optimizations for first character search and single
299        character iterators.
300    
301    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
302        Patch by Daniel Richard G.
303    
304    3.  Single character iterator optimizations in the JIT compiler.
305    
306    4.  Improved JIT compiler optimizations for character ranges.
307    
308    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
309        Reported by Giuseppe D'Angelo.
310    
311    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
312        of a line, was being set incorrectly in cases where .* appeared inside
313        atomic brackets at the start of a pattern, or where there was a subsequent
314        *PRUNE or *SKIP.
315    
316    7.  Improved instruction cache flush for POWER/PowerPC.
317        Patch by Daniel Richard G.
318    
319    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
320        grep:
321    
322        (a) There is now no limit to the number of patterns to be matched.
323    
324        (b) An error is given if a pattern is too long.
325    
326        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
327            are now supported.
328    
329        (d) --exclude-from and --include-from (multiple use) have been added.
330    
331        (e) Exclusions and inclusions now apply to all files and directories, not
332            just to those obtained from scanning a directory recursively.
333    
334        (f) Multiple uses of -f and --file-list are now supported.
335    
336        (g) In a Windows environment, the default for -d has been changed from
337            "read" (the GNU grep default) to "skip", because otherwise the presence
338            of a directory in the file list provokes an error.
339    
340        (h) The documentation has been revised and clarified in places.
341    
342    9.  Improve the matching speed of capturing brackets.
343    
344    10. Changed the meaning of \X so that it now matches a Unicode extended
345        grapheme cluster.
346    
347    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
348        out POSIX threads when JIT support is configured.
349    
350    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
351    
352    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
353        structure could go wrong in environments where size_t is not the same size
354        as int.
355    
356    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
357        set.
358    
359    15. The EBCDIC support had decayed; later updates to the code had included
360        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
361        general tidy up of EBCDIC-related issues, and the documentation was also
362        not quite right. There is now a test that can be run on ASCII systems to
363        check some of the EBCDIC-related things (but is it not a full test).
364    
365    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
366        in a small tidy to the code.
367    
368    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
369    
370    18. If the --only-matching (-o) option in pcregrep is specified multiple
371        times, each one causes appropriate output. For example, -o1 -o2 outputs the
372        substrings matched by the 1st and 2nd capturing parentheses. A separating
373        string can be specified by --om-separator (default empty).
374    
375    19. Improving the first n character searches.
376    
377    20. Turn case lists for horizontal and vertical white space into macros so that
378        they are defined only once.
379    
380    21. This set of changes together give more compatible Unicode case-folding
381        behaviour for characters that have more than one other case when UCP
382        support is available.
383    
384        (a) The Unicode property table now has offsets into a new table of sets of
385            three or more characters that are case-equivalent. The MultiStage2.py
386            script that generates these tables (the pcre_ucd.c file) now scans
387            CaseFolding.txt instead of UnicodeData.txt for character case
388            information.
389    
390        (b) The code for adding characters or ranges of characters to a character
391            class has been abstracted into a generalized function that also handles
392            case-independence. In UTF-mode with UCP support, this uses the new data
393            to handle characters with more than one other case.
394    
395        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
396            whose other case is greater than 256 are now correctly matched
397            caselessly. Previously, the high codepoint matched the low one, but not
398            vice versa.
399    
400        (d) The processing of \h, \H, \v, and \ in character classes now makes use
401            of the new class addition function, using character lists defined as
402            macros alongside the case definitions of 20 above.
403    
404        (e) Caseless back references now work with characters that have more than
405            one other case.
406    
407        (f) General caseless matching of characters with more than one other case
408            is supported.
409    
410    22. Unicode character properties were updated from Unicode 6.2.0
411    
412    23. Improved CMake support under Windows. Patch by Daniel Richard G.
413    
414    24. Add support for 32-bit character strings, and UTF-32
415    
416    25. Major JIT compiler update (code refactoring and bugfixing).
417        Experimental Sparc 32 support is added.
418    
419    26. Applied a modified version of Daniel Richard G's patch to create
420        pcre.h.generic and config.h.generic by "make" instead of in the
421        PrepareRelease script.
422    
423    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
424        pcre_compile.c when checking for a zero character.
425    
426    28. Introducing a native interface for JIT. Through this interface, the compiled
427        machine code can be directly executed. The purpose of this interface is to
428        provide fast pattern matching, so several sanity checks are not performed.
429        However, feature tests are still performed. The new interface provides
430        1.4x speedup compared to the old one.
431    
432    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
433        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
434        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
435    
436    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
437        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
438        the "old" RFC 2279). Instead, it ended up passing a negative length to
439        pcre_exec().
440    
441    31. Add support for GCC's visibility feature to hide internal functions.
442    
443    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
444        "unknown -C option" after outputting 0 or 1.
445    
446    33. There is now support for generating a code coverage report for the test
447        suite in environments where gcc is the compiler and lcov is installed. This
448        is mainly for the benefit of the developers.
449    
450    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
451        unaddressable using valgrind annotations, allowing valgrind to detect
452        invalid memory accesses. This is mainly for the benefit of the developers.
453    
454    25. (*UTF) can now be used to start a pattern in any of the three libraries.
455    
456    26. Give configure error if --enable-cpp but no C++ compiler found.
457    
458    
459    Version 8.31 06-July-2012
460    -------------------------
461    
462    1.  Fixing a wrong JIT test case and some compiler warnings.
463    
464    2.  Removed a bashism from the RunTest script.
465    
466    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
467        to unsigned type, result still unsigned" that was given by an MS compiler
468        on encountering the code "-sizeof(xxx)".
469    
470    4.  Partial matching support is added to the JIT compiler.
471    
472    5.  Fixed several bugs concerned with partial matching of items that consist
473        of more than one character:
474    
475        (a) /^(..)\1/ did not partially match "aba" because checking references was
476            done on an "all or nothing" basis. This also applied to repeated
477            references.
478    
479        (b) \R did not give a hard partial match if \r was found at the end of the
480            subject.
481    
482        (c) \X did not give a hard partial match after matching one or more
483            characters at the end of the subject.
484    
485        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
486            a partial match for the string "\r".
487    
488        (e) When newline was set to CRLF, the metacharacter "." did not recognize
489            a partial match for a CR character at the end of the subject string.
490    
491    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
492        running pcretest, the text "(JIT)" added to the output whenever JIT is
493        actually used to run the match.
494    
495    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
496        or /S+[+] with a digit between 1 and 7.
497    
498    8.  OP_NOT now supports any UTF character not just single-byte ones.
499    
500    9.  (*MARK) control verb is now supported by the JIT compiler.
501    
502    10. The command "./RunTest list" lists the available tests without actually
503        running any of them. (Because I keep forgetting what they all are.)
504    
505    11. Add PCRE_INFO_MAXLOOKBEHIND.
506    
507    12. Applied a (slightly modified) user-supplied patch that improves performance
508        when the heap is used for recursion (compiled with --disable-stack-for-
509        recursion). Instead of malloc and free for each heap frame each time a
510        logical recursion happens, frames are retained on a chain and re-used where
511        possible. This sometimes gives as much as 30% improvement.
512    
513    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
514        call.
515    
516    14. As documented, (*COMMIT) is now confined to within a positive assertion.
517    
518    15. It is now possible to link pcretest with libedit as an alternative to
519        libreadline.
520    
521    16. (*COMMIT) control verb is now supported by the JIT compiler.
522    
523    17. The Unicode data tables have been updated to Unicode 6.1.0.
524    
525    18. Added --file-list option to pcregrep.
526    
527    19. Added binary file support to pcregrep, including the -a, --binary-files,
528        -I, and --text options.
529    
530    20. The madvise function is renamed for posix_madvise for QNX compatibility
531        reasons. Fixed by Giuseppe D'Angelo.
532    
533    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
534    
535    22. Changed the option for creating symbolic links for 16-bit man pages from
536        -s to -sf so that re-installing does not cause issues.
537    
538    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
539    
540    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
541        matches in certain environments (the workspace was not being correctly
542        retained). Also added to pcre_dfa_exec() a simple plausibility check on
543        some of the workspace data at the beginning of a restart.
544    
545    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
546        was not doing so when it should - probably a typo introduced by SVN 528
547        (change 8.10/14).
548    
549    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
550        \w+ when the character tables indicated that \x{c4} was a word character.
551        There were several related cases, all because the tests for doing a table
552        lookup were testing for characters less than 127 instead of 255.
553    
554    27. If a pattern contains capturing parentheses that are not used in a match,
555        their slots in the ovector are set to -1. For those that are higher than
556        any matched groups, this happens at the end of processing. In the case when
557        there were back references that the ovector was too small to contain
558        (causing temporary malloc'd memory to be used during matching), and the
559        highest capturing number was not used, memory off the end of the ovector
560        was incorrectly being set to -1. (It was using the size of the temporary
561        memory instead of the true size.)
562    
563    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
564        ovector size, it uses memory at the end of the block that it has got.
565    
566    29. Check for an overlong MARK name and give an error at compile time. The
567        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
568    
569    30. JIT compiler update.
570    
571    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
572        Rill for the patch.
573    
574    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
575    
576    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
577    
578    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
579        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
580        was enabled.
581    
582    35. Improve JIT code generation for greedy plus quantifier.
583    
584    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
585        1 to "aa" instead of to an empty string. The bug affected repeated groups
586        that could potentially match an empty string.
587    
588    37. Optimizing single character iterators in JIT.
589    
590    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
591        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
592        codepoints that are too big for the mode are faulted, and in a UTF mode,
593        disallowed codepoints are also faulted.
594    
595    39. If PCRE was compiled with UTF support, in three places in the DFA
596        matcher there was code that should only have been obeyed in UTF mode, but
597        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
598        processing when bytes with values greater than 127 were present. In 16-bit
599        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
600        both cases the values are those that cannot be the first data item in a UTF
601        character. The three items that might have provoked this were recursions,
602        possessively repeated groups, and atomic groups.
603    
604    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
605        and pcregrep, because some OS require shared objects to be explicitly
606        passed to ld, causing the link step to fail if they are not.
607    
608    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
609        mode, patterns that started with \h* or \R* might be incorrectly matched.
610    
611    
612    Version 8.30 04-February-2012
613    -----------------------------
614    
615    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
616        name is defined in ctype.h.
617    
618    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
619        only in quite long subpatterns.
620    
621    3.  Removed the function pcre_info(), which has been obsolete and deprecated
622        since it was replaced by pcre_fullinfo() in February 2000.
623    
624    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
625        match a (*MARK), and the match failed at the start of the subject, a
626        reference to memory before the start of the subject could occur. This bug
627        was introduced by fix 17 of release 8.21.
628    
629    5.  A reference to an unset group with zero minimum repetition was giving
630        totally wrong answers (in non-JavaScript-compatibility mode). For example,
631        /(another)?(\1?)test/ matched against "hello world test". This bug was
632        introduced in release 8.13.
633    
634    6.  Add support for 16-bit character strings (a large amount of work involving
635        many changes and refactorings).
636    
637    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
638        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
639        from a file.
640    
641    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
642        rounding is not applied in this particular case).
643    
644    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
645        if they appear, or are escaped, in patterns.
646    
647    10. Get rid of a number of -Wunused-but-set-variable warnings.
648    
649    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
650        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
651        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
652        also returns the mark "x". This bug applied to capturing parentheses,
653        non-capturing parentheses, and atomic parentheses. It also applied to some
654        assertions.
655    
656    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
657        information out of configure.ac instead of relying on pcre.h.generic, which
658        is not stored in the repository.
659    
660    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
661        -lreadline.
662    
663    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
664    
665    15. Applied Graycode's patch to put the top-level frame on the stack rather
666        than the heap when not using the stack for recursion. This gives a
667        performance improvement in many cases when recursion is not deep.
668    
669    16. Experimental code added to "pcretest -C" to output the stack frame size.
670    
671    
672    Version 8.21 12-Dec-2011
673    ------------------------
674    
675    1.  Updating the JIT compiler.
676    
677    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
678        are added as well.
679    
680    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
681        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
682        calling _pcre_jit_exec. Some extra comments are added.
683    
684    4.  (*MARK) settings inside atomic groups that do not contain any capturing
685        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
686        was introduced by change 18 for 8.20.
687    
688    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
689        ECMA-262 standard.
690    
691    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
692        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
693        This bug was probably introduced by change 9 of 8.13.
694    
695    7.  While fixing 6 above, I noticed that a number of other items were being
696        incorrectly rejected as "not fixed length". This arose partly because newer
697        opcodes had not been added to the fixed-length checking code. I have (a)
698        corrected the bug and added tests for these items, and (b) arranged for an
699        error to occur if an unknown opcode is encountered while checking for fixed
700        length instead of just assuming "not fixed length". The items that were
701        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
702        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
703        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
704    
705    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
706        being incorrectly compiled and would have given unpredicatble results.
707    
708    9.  A possessively repeated subpattern with minimum repeat count greater than
709        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
710        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
711        the first (A) could occur when it should not.
712    
713    10. Add a cast and remove a redundant test from the code.
714    
715    11. JIT should use pcre_malloc/pcre_free for allocation.
716    
717    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
718        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
719        is anything other than /usr, -L is still shown).
720    
721    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
722    
723    14. Perl does not support \N without a following name in a [] class; PCRE now
724        also gives an error.
725    
726    15. If a forward reference was repeated with an upper limit of around 2000,
727        it caused the error "internal error: overran compiling workspace". The
728        maximum number of forward references (including repeats) was limited by the
729        internal workspace, and dependent on the LINK_SIZE. The code has been
730        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
731        the default depends on LINK_SIZE. There is a new upper limit (for safety)
732        of around 200,000 forward references. While doing this, I also speeded up
733        the filling in of repeated forward references.
734    
735    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
736        incorrectly expecting the subject to contain another "a" after the start.
737    
738    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
739        in the match, the SKIP should be ignored. This was not happening; instead
740        the SKIP was being treated as NOMATCH. For patterns such as
741        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
742        tested.
743    
744    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
745        now much more compatible with Perl, in particular in cases where the result
746        is a non-match for a non-anchored pattern. For example, if
747        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
748        "m", where previously it did not return a name. A side effect of this
749        change is that for partial matches, the last encountered mark name is
750        returned, as for non matches. A number of tests that were previously not
751        Perl-compatible have been moved into the Perl-compatible test files. The
752        refactoring has had the pleasing side effect of removing one argument from
753        the match() function, thus reducing its stack requirements.
754    
755    19. If the /S+ option was used in pcretest to study a pattern using JIT,
756        subsequent uses of /S (without +) incorrectly behaved like /S+.
757    
758    21. Retrieve executable code size support for the JIT compiler and fixing
759        some warnings.
760    
761    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
762        not work when the shorter character appeared right at the end of the
763        subject string.
764    
765    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
766        systems.
767    
768    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
769        output it when the /M option is used in pcretest.
770    
771    25. The CheckMan script was not being included in the distribution. Also, added
772        an explicit "perl" to run Perl scripts from the PrepareRelease script
773        because this is reportedly needed in Windows.
774    
775    26. If study data was being save in a file and studying had not found a set of
776        "starts with" bytes for the pattern, the data written to the file (though
777        never used) was taken from uninitialized memory and so caused valgrind to
778        complain.
779    
780    27. Updated RunTest.bat as provided by Sheri Pierce.
781    
782    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
783    
784    29. Computation of memory usage for the table of capturing group names was
785        giving an unnecessarily large value.
786    
787    
788    Version 8.20 21-Oct-2011
789    ------------------------
790    
791    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
792        a POSIX class. After further experiments with Perl, which convinced me that
793        Perl has bugs and confusions, a closing square bracket is no longer allowed
794        in a POSIX name. This bug also affected patterns with classes that started
795        with full stops.
796    
797    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
798        captured substring, but while checking the failing first alternative,
799        substring 1 is temporarily captured. If the output vector supplied to
800        pcre_exec() was not big enough for this capture, the yield of the function
801        was still zero ("insufficient space for captured substrings"). This cannot
802        be totally fixed without adding another stack variable, which seems a lot
803        of expense for a edge case. However, I have improved the situation in cases
804        such as /(a)(b)x|abc/ matched against "abc", where the return code
805        indicates that fewer than the maximum number of slots in the ovector have
806        been set.
807    
808    3.  Related to (2) above: when there are more back references in a pattern than
809        slots in the output vector, pcre_exec() uses temporary memory during
810        matching, and copies in the captures as far as possible afterwards. It was
811        using the entire output vector, but this conflicts with the specification
812        that only 2/3 is used for passing back captured substrings. Now it uses
813        only the first 2/3, for compatibility. This is, of course, another edge
814        case.
815    
816    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
817        main code base, and can be used by building with --enable-jit. When this is
818        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
819        runtime --no-jit option is given.
820    
821    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
822        ovector, the return from the function was zero, implying that there were
823        other matches that did not fit. The correct "exactly full" value is now
824        returned.
825    
826    6.  If a subpattern that was called recursively or as a subroutine contained
827        (*PRUNE) or any other control that caused it to give a non-standard return,
828        invalid errors such as "Error -26 (nested recursion at the same subject
829        position)" or even infinite loops could occur.
830    
831    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
832        computing the minimum length on reaching *ACCEPT, and so ended up with the
833        wrong value of 1 rather than 0. Further investigation indicates that
834        computing a minimum subject length in the presence of *ACCEPT is difficult
835        (think back references, subroutine calls), and so I have changed the code
836        so that no minimum is registered for a pattern that contains *ACCEPT.
837    
838    8.  If (*THEN) was present in the first (true) branch of a conditional group,
839        it was not handled as intended. [But see 16 below.]
840    
841    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
842        Sheri Pierce.
843    
844    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
845        the first byte in a match must be "a".
846    
847    11. Change 17 for 8.13 increased the recursion depth for patterns like
848        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
849        pattern contains any instances of (*THEN). If it does not, the old
850        optimizations are restored. It would be nice to do this on a per-group
851        basis, but at the moment that is not feasible.
852    
853    12. In some environments, the output of pcretest -C is CRLF terminated. This
854        broke RunTest's code that checks for the link size. A single white space
855        character after the value is now allowed for.
856    
857    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
858        For "fr", it uses the Windows-specific input and output files.
859    
860    14. If (*THEN) appeared in a group that was called recursively or as a
861        subroutine, it did not work as intended. [But see next item.]
862    
863    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
864        pattern fragments (but not containing any | characters). If A and B are
865        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
866        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
867        D. In other words, Perl considers parentheses that do not contain any |
868        characters to be part of a surrounding alternative, whereas PCRE was
869        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
870        differently. PCRE now behaves in the same way as Perl, except in the case
871        of subroutine/recursion calls such as (?1) which have in any case always
872        been different (but PCRE had them first :-).
873    
874    16. Related to 15 above: Perl does not treat the | in a conditional group as
875        creating alternatives. Such a group is treated in the same way as an
876        ordinary group without any | characters when processing (*THEN). PCRE has
877        been changed to match Perl's behaviour.
878    
879    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
880        RunGrepTest script failed.
881    
882    18. Change 22 for version 13 caused atomic groups to use more stack. This is
883        inevitable for groups that contain captures, but it can lead to a lot of
884        stack use in large patterns. The old behaviour has been restored for atomic
885        groups that do not contain any capturing parentheses.
886    
887    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
888        suppress the check for a minimum subject length at run time. (If it was
889        given to pcre_exec() or pcre_dfa_exec() it did work.)
890    
891    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
892        fail to work when decoding hex characters in data strings in EBCDIC
893        environments.
894    
895    21. It appears that in at least one Mac OS environment, the isxdigit() function
896        is implemented as a macro that evaluates to its argument more than once,
897        contravening the C 90 Standard (I haven't checked a later standard). There
898        was an instance in pcretest which caused it to go wrong when processing
899        \x{...} escapes in subject strings. The has been rewritten to avoid using
900        things like p++ in the argument of isxdigit().
901    
902    
903    Version 8.13 16-Aug-2011
904    ------------------------
905    
906    1.  The Unicode data tables have been updated to Unicode 6.0.0.
907    
908    2.  Two minor typos in pcre_internal.h have been fixed.
909    
910    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
911        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
912        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
913    
914    4.  There were a number of related bugs in the code for matching backrefences
915        caselessly in UTF-8 mode when codes for the characters concerned were
916        different numbers of bytes. For example, U+023A and U+2C65 are an upper
917        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
918        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
919        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
920        2-byte code at the end of the subject (it thought there wasn't enough data
921        left).
922    
923    5.  Comprehensive information about what went wrong is now returned by
924        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
925        as the output vector has at least 2 elements. The offset of the start of
926        the failing character and a reason code are placed in the vector.
927    
928    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
929        now returned is for the first byte of the failing character, instead of the
930        last byte inspected. This is an incompatible change, but I hope it is small
931        enough not to be a problem. It makes the returned offset consistent with
932        pcre_exec() and pcre_dfa_exec().
933    
934    7.  pcretest now gives a text phrase as well as the error number when
935        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
936        failure, the offset and reason code are output.
937    
938    8.  When \R was used with a maximizing quantifier it failed to skip backwards
939        over a \r\n pair if the subsequent match failed. Instead, it just skipped
940        back over a single character (\n). This seems wrong (because it treated the
941        two characters as a single entity when going forwards), conflicts with the
942        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
943        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
944        has been changed.
945    
946    9.  Some internal refactoring has changed the processing so that the handling
947        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
948        time (the PCRE_DOTALL option was changed this way some time ago: version
949        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
950        which was always a bit of a fudge. It also means that there is one less
951        argument for the match() function, which reduces its stack requirements
952        slightly. This change also fixes an incompatibility with Perl: the pattern
953        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
954    
955    10. More internal refactoring has drastically reduced the number of recursive
956        calls to match() for possessively repeated groups such as (abc)++ when
957        using pcre_exec().
958    
959    11. While implementing 10, a number of bugs in the handling of groups were
960        discovered and fixed:
961    
962        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
963        (a|)*(?1) gave a compile-time internal error.
964        ((a|)+)+  did not notice that the outer group could match an empty string.
965        (^a|^)+   was not marked as anchored.
966        (.*a|.*)+ was not marked as matching at start or after a newline.
967    
968    12. Yet more internal refactoring has removed another argument from the match()
969        function. Special calls to this function are now indicated by setting a
970        value in a variable in the "match data" data block.
971    
972    13. Be more explicit in pcre_study() instead of relying on "default" for
973        opcodes that mean there is no starting character; this means that when new
974        ones are added and accidentally left out of pcre_study(), testing should
975        pick them up.
976    
977    14. The -s option of pcretest has been documented for ages as being an old
978        synonym of -m (show memory usage). I have changed it to mean "force study
979        for every regex", that is, assume /S for every regex. This is similar to -i
980        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
981        using it. It makes it easier to run collections of tests with and without
982        study enabled, and thereby test pcre_study() more easily. All the standard
983        tests are now run with and without -s (but some patterns can be marked as
984        "never study" - see 20 below).
985    
986    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
987        restoration of the capturing data to the outer values was not happening
988        correctly.
989    
990    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
991        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
992        pattern had matched an empty string, and so incorrectly returned a no
993        match.
994    
995    17. There was optimizing code for the last branch of non-capturing parentheses,
996        and also for the obeyed branch of a conditional subexpression, which used
997        tail recursion to cut down on stack usage. Unfortunately, now that there is
998        the possibility of (*THEN) occurring in these branches, tail recursion is
999        no longer possible because the return has to be checked for (*THEN). These
1000        two optimizations have therefore been removed. [But see 8.20/11 above.]
1001    
1002    18. If a pattern containing \R was studied, it was assumed that \R always
1003        matched two bytes, thus causing the minimum subject length to be
1004        incorrectly computed because \R can also match just one byte.
1005    
1006    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
1007        was incorrectly computed.
1008    
1009    20. If /S is present twice on a test pattern in pcretest input, it now
1010        *disables* studying, thereby overriding the use of -s on the command line
1011        (see 14 above). This is necessary for one or two tests to keep the output
1012        identical in both cases.
1013    
1014    21. When (*ACCEPT) was used in an assertion that matched an empty string and
1015        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
1016    
1017    22. When an atomic group that contained a capturing parenthesis was
1018        successfully matched, but the branch in which it appeared failed, the
1019        capturing was not being forgotten if a higher numbered group was later
1020        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
1021        group 1 to "a", when in fact it should be unset. This applied to multi-
1022        branched capturing and non-capturing groups, repeated or not, and also to
1023        positive assertions (capturing in negative assertions does not happen
1024        in PCRE) and also to nested atomic groups.
1025    
1026    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
1027        subject after a captured substring, to make it easier to tell which of a
1028        number of identical substrings has been captured.
1029    
1030    24. The way atomic groups are processed by pcre_exec() has been changed so that
1031        if they are repeated, backtracking one repetition now resets captured
1032        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
1033        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
1034        "aaa". Previously, it would have been "a". As part of this code
1035        refactoring, the way recursive calls are handled has also been changed.
1036    
1037    25. If an assertion condition captured any substrings, they were not passed
1038        back unless some other capturing happened later. For example, if
1039        (?(?=(a))a) was matched against "a", no capturing was returned.
1040    
1041    26. When studying a pattern that contained subroutine calls or assertions,
1042        the code for finding the minimum length of a possible match was handling
1043        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
1044        group 1 called group 2 while simultaneously a separate group 2 called group
1045        1). A stack overflow occurred in this case. I have fixed this by limiting
1046        the recursion depth to 10.
1047    
1048    27. Updated RunTest.bat in the distribution to the version supplied by Tom
1049        Fortmann. This supports explicit test numbers on the command line, and has
1050        argument validation and error reporting.
1051    
1052    28. An instance of \X with an unlimited repeat could fail if at any point the
1053        first character it looked at was a mark character.
1054    
1055    29. Some minor code refactoring concerning Unicode properties and scripts
1056        should reduce the stack requirement of match() slightly.
1057    
1058    30. Added the '=' option to pcretest to check the setting of unused capturing
1059        slots at the end of the pattern, which are documented as being -1, but are
1060        not included in the return count.
1061    
1062    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
1063        compiled something random. Now it gives a compile-time error (as does
1064        Perl).
1065    
1066    32. A *MARK encountered during the processing of a positive assertion is now
1067        recorded and passed back (compatible with Perl).
1068    
1069    33. If --only-matching or --colour was set on a pcregrep call whose pattern
1070        had alternative anchored branches, the search for a second match in a line
1071        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
1072        matched the line "0102" twice. The same bug affected patterns that started
1073        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
1074        twice.
1075    
1076    34. Previously, PCRE did not allow quantification of assertions. However, Perl
1077        does, and because of capturing effects, quantifying parenthesized
1078        assertions may at times be useful. Quantifiers are now allowed for
1079        parenthesized assertions.
1080    
1081    35. A minor code tidy in pcre_compile() when checking options for \R usage.
1082    
1083    36. \g was being checked for fancy things in a character class, when it should
1084        just be a literal "g".
1085    
1086    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
1087        appearance of a nested POSIX class supersedes an apparent external class.
1088        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
1089        unescaped square brackets may also appear as part of class names. For
1090        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
1091        more like Perl. (But see 8.20/1 above.)
1092    
1093    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
1094        was because it thought it was \N{name}, which is not supported).
1095    
1096    39. Add minix to OS list not supporting the -S option in pcretest.
1097    
1098    40. PCRE tries to detect cases of infinite recursion at compile time, but it
1099        cannot analyze patterns in sufficient detail to catch mutual recursions
1100        such as ((?1))((?2)). There is now a runtime test that gives an error if a
1101        subgroup is called recursively as a subpattern for a second time at the
1102        same position in the subject string. In previous releases this might have
1103        been caught by the recursion limit, or it might have run out of stack.
1104    
1105    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
1106        happen only once. PCRE was, however incorrectly giving a compile time error
1107        "recursive call could loop indefinitely" because it cannot analyze the
1108        pattern in sufficient detail. The compile time test no longer happens when
1109        PCRE is compiling a conditional subpattern, but actual runaway loops are
1110        now caught at runtime (see 40 above).
1111    
1112    42. It seems that Perl allows any characters other than a closing parenthesis
1113        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
1114        has been changed to be the same.
1115    
1116    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
1117        as not to get warnings when autogen.sh is called. Also changed
1118        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
1119    
1120    44. To help people who use pcregrep to scan files containing exceedingly long
1121        lines, the following changes have been made:
1122    
1123        (a) The default value of the buffer size parameter has been increased from
1124            8K to 20K. (The actual buffer used is three times this size.)
1125    
1126        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
1127            PCRE is built.
1128    
1129        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
1130            to be set at run time.
1131    
1132        (d) Numerical values in pcregrep options can be followed by K or M, for
1133            example --buffer-size=50K.
1134    
1135        (e) If a line being scanned overflows pcregrep's buffer, an error is now
1136            given and the return code is set to 2.
1137    
1138    45. Add a pointer to the latest mark to the callout data block.
1139    
1140    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
1141        partial match of an empty string instead of no match. This was specific to
1142        the use of ".".
1143    
1144    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
1145        complete match instead of a partial match. This bug was dependent on both
1146        the PCRE_UTF8 and PCRE_DOTALL options being set.
1147    
1148    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1149        starting byte set, because \b was not being ignored.
1150    
1151    
1152    Version 8.12 15-Jan-2011
1153    ------------------------
1154    
1155    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1156        checks for such things as part of the documentation building process.
1157    
1158    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1159        --match-limit and --recursion-limit options (added for 8.11). In
1160        particular, this made one of the standard tests fail. (The integer value
1161        went into the wrong half of a long int.)
1162    
1163    3.  If the --colour option was given to pcregrep with -v (invert match), it
1164        did strange things, either producing crazy output, or crashing. It should,
1165        of course, ignore a request for colour when reporting lines that do not
1166        match.
1167    
1168    4.  Another pcregrep bug caused similar problems if --colour was specified with
1169        -M (multiline) and the pattern match finished with a line ending.
1170    
1171    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1172        matched in multiline mode, the following line was shown as part of the
1173        match. This seems wrong, so I have changed it.
1174    
1175    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1176        the check for further matches in the same line (so they could be coloured)
1177        to overrun the end of the current line. If another match was found, it was
1178        incorrectly shown (and then shown again when found in the next line).
1179    
1180    7.  If pcregrep was compiled under Windows, there was a reference to the
1181        function pcregrep_exit() before it was defined. I am assuming this was
1182        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1183        reported by a user. I've moved the definition above the reference.
1184    
1185    
1186    Version 8.11 10-Dec-2010
1187    ------------------------
1188    
1189    1.  (*THEN) was not working properly if there were untried alternatives prior
1190        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1191        backtracked to try for "b" instead of moving to the next alternative branch
1192        at the same level (in this case, to look for "c"). The Perl documentation
1193        is clear that when (*THEN) is backtracked onto, it goes to the "next
1194        alternative in the innermost enclosing group".
1195    
1196    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1197        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1198        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1199        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1200        (*THEN).
1201    
1202    3.  If \s appeared in a character class, it removed the VT character from
1203        the class, even if it had been included by some previous item, for example
1204        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1205        of \s, but is part of the POSIX "space" class.)
1206    
1207    4.  A partial match never returns an empty string (because you can always
1208        match an empty string at the end of the subject); however the checking for
1209        an empty string was starting at the "start of match" point. This has been
1210        changed to the "earliest inspected character" point, because the returned
1211        data for a partial match starts at this character. This means that, for
1212        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1213        (previously it gave "no match").
1214    
1215    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1216        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1217        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1218        has an implication that the given string is incomplete (because a partial
1219        match is preferred over a full match). For this reason, these items now
1220        give a partial match in this situation. [Aside: previously, the one case
1221        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1222        match rather than a full match, which was wrong by the old rules, but is
1223        now correct.]
1224    
1225    6.  There was a bug in the handling of #-introduced comments, recognized when
1226        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1227        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1228        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1229        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1230        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1231        places in pcre_compile().
1232    
1233    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1234        comments when looking ahead for named forward references to subpatterns,
1235        the only newline sequence it recognized was NL. It now handles newlines
1236        according to the set newline convention.
1237    
1238    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1239        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1240        cater for a lack of strerror(). These oversights have been fixed.
1241    
1242    9.  Added --match-limit and --recursion-limit to pcregrep.
1243    
1244    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1245    
1246    11. When the -o option was used, pcregrep was setting a return code of 1, even
1247        when matches were found, and --line-buffered was not being honoured.
1248    
1249    12. Added an optional parentheses number to the -o and --only-matching options
1250        of pcregrep.
1251    
1252    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1253        can match an empty string. The code to do it in pcretest and pcredemo
1254        needed fixing:
1255    
1256        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1257            only one byte after an empty string match just before CRLF (this case
1258            just got forgotten; "any" and "anycrlf" were OK).
1259    
1260        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1261            mode when an empty string match preceded an ASCII character followed by
1262            a non-ASCII character. (The code for advancing by one character rather
1263            than one byte was nonsense.)
1264    
1265        (c) The pcredemo.c sample program did not have any code at all to handle
1266            the cases when CRLF is a valid newline sequence.
1267    
1268    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1269        as a starting offset was within the subject string. There is now a new
1270        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1271        negative or greater than the length of the string. In order to test this,
1272        pcretest is extended to allow the setting of negative starting offsets.
1273    
1274    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1275        starting offset points to the beginning of a UTF-8 character was
1276        unnecessarily clumsy. I tidied it up.
1277    
1278    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1279        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1280    
1281    17. Nobody had reported that the --include_dir option, which was added in
1282        release 7.7 should have been called --include-dir (hyphen, not underscore)
1283        for compatibility with GNU grep. I have changed it to --include-dir, but
1284        left --include_dir as an undocumented synonym, and the same for
1285        --exclude-dir, though that is not available in GNU grep, at least as of
1286        release 2.5.4.
1287    
1288    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1289        characters from a string of bytes) have been redefined so as not to use
1290        loops, in order to improve performance in some environments. At the same
1291        time, I abstracted some of the common code into auxiliary macros to save
1292        repetition (this should not affect the compiled code).
1293    
1294    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1295        compile-time error is now given if \c is not followed by an ASCII
1296        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1297        different, and any byte value is allowed.)
1298    
1299    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1300        START_OPTIMIZE option, which is now allowed at compile time - but just
1301        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1302        to pcregrep and other applications that have no direct access to PCRE
1303        options. The new /Y option in pcretest sets this option when calling
1304        pcre_compile().
1305    
1306    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1307        back references. Groups containing recursive back references were forced to
1308        be atomic by that change, but in the case of named groups, the amount of
1309        memory required was incorrectly computed, leading to "Failed: internal
1310        error: code overflow". This has been fixed.
1311    
1312    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1313        pcretest.c, to avoid build problems in some Borland environments.
1314    
1315    
1316    Version 8.10 25-Jun-2010
1317    ------------------------
1318    
1319    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1320        THEN.
1321    
1322    2.  (*ACCEPT) was not working when inside an atomic group.
1323    
1324    3.  Inside a character class, \B is treated as a literal by default, but
1325        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1326        causes the error). The code is unchanged, but I tidied the documentation.
1327    
1328    4.  Inside a character class, PCRE always treated \R and \X as literals,
1329        whereas Perl faults them if its -w option is set. I have changed PCRE so
1330        that it faults them when PCRE_EXTRA is set.
1331    
1332    5.  Added support for \N, which always matches any character other than
1333        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1334    
1335    6.  When compiling pcregrep with newer versions of gcc which may have
1336        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1337        declared with attribute warn_unused_result" were given. Just casting the
1338        result to (void) does not stop the warnings; a more elaborate fudge is
1339        needed. I've used a macro to implement this.
1340    
1341    7.  Minor change to pcretest.c to avoid a compiler warning.
1342    
1343    8.  Added four artifical Unicode properties to help with an option to make
1344        \s etc use properties (see next item). The new properties are: Xan
1345        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1346    
1347    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1348        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1349        this option. Modified pcretest to add /W to test this facility. Added
1350        REG_UCP to make it available via the POSIX interface.
1351    
1352    10. Added --line-buffered to pcregrep.
1353    
1354    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1355        studied, and the match started with a letter with a code point greater than
1356        127 whose first byte was different to the first byte of the other case of
1357        the letter, the other case of this starting letter was not recognized
1358        (#976).
1359    
1360    12. If a pattern that was studied started with a repeated Unicode property
1361        test, for example, \p{Nd}+, there was the theoretical possibility of
1362        setting up an incorrect bitmap of starting bytes, but fortunately it could
1363        not have actually happened in practice until change 8 above was made (it
1364        added property types that matched character-matching opcodes).
1365    
1366    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1367        possible starting bytes for non-anchored patterns.
1368    
1369    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1370        \R, and also a number of cases that involve Unicode properties, both
1371        explicit and implicit when PCRE_UCP is set.
1372    
1373    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1374        input, it could crash or give wrong results if characters with values
1375        greater than 0xc0 were present in the subject string. (Detail: it assumed
1376        UTF-8 input when processing these items.)
1377    
1378    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1379        size_t is 64-bit (#991).
1380    
1381    17. Added a check for running out of memory when PCRE is compiled with
1382        --disable-stack-for-recursion (#990).
1383    
1384    18. If the last data line in a file for pcretest does not have a newline on
1385        the end, a newline was missing in the output.
1386    
1387    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1388        less than 128) in its various bitmaps. However, there is a facility for
1389        generating tables according to the current locale when PCRE is compiled. It
1390        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1391        characters, are recognized by isspace() and therefore were getting set in
1392        these tables, and indeed these tables seem to approximate to ISO 8859. This
1393        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1394        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1395        which of course cannot start UTF-8 characters. I have changed the code so
1396        that only real ASCII characters (less than 128) and the correct starting
1397        bytes for UTF-8 encodings are set for characters greater than 127 when in
1398        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1399        altogether.)
1400    
1401    20. Added the /T option to pcretest so as to be able to run tests with non-
1402        standard character tables, thus making it possible to include the tests
1403        used for 19 above in the standard set of tests.
1404    
1405    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1406        reference to a subpattern the other side of a comment that contains an
1407        opening parenthesis caused either an internal compiling error, or a
1408        reference to the wrong subpattern.
1409    
1410    
1411    Version 8.02 19-Mar-2010
1412    ------------------------
1413    
1414    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1415    
1416    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1417        configured.
1418    
1419    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1420        original author of that file, following a query about its status.
1421    
1422    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1423        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1424    
1425    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1426        quantifier applied to a forward-referencing subroutine call, could compile
1427        incorrect code or give the error "internal error: previously-checked
1428        referenced subpattern not found".
1429    
1430    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1431        variables to point to external functions. For these systems, therefore,
1432        pcre_malloc etc. are now initialized to local functions that call the
1433        relevant global functions.
1434    
1435    7.  There were two entries missing in the vectors called coptable and poptable
1436        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1437        I've fixed the data, and added a kludgy way of testing at compile time that
1438        the lengths are correct (equal to the number of opcodes).
1439    
1440    8.  Following on from 7, I added a similar kludge to check the length of the
1441        eint vector in pcreposix.c.
1442    
1443    9.  Error texts for pcre_compile() are held as one long string to avoid too
1444        much relocation at load time. To find a text, the string is searched,
1445        counting zeros. There was no check for running off the end of the string,
1446        which could happen if a new error number was added without updating the
1447        string.
1448    
1449    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1450    
1451    11. \K was not working if it appeared in an atomic group or in a group that
1452        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1453        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1454        the assertion is positive, but not if it is negative.
1455    
1456    12. Change 11 fortuitously reduced the size of the stack frame used in the
1457        "match()" function of pcre_exec.c by one pointer. Forthcoming
1458        implementation of support for (*MARK) will need an extra pointer on the
1459        stack; I have reserved it now, so that the stack frame size does not
1460        decrease.
1461    
1462    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1463        item in branch that calls a recursion is a subroutine call - as in the
1464        second branch in the above example - was incorrectly given the compile-
1465        time error "recursive call could loop indefinitely" because pcre_compile()
1466        was not correctly checking the subroutine for matching a non-empty string.
1467    
1468    14. The checks for overrunning compiling workspace could trigger after an
1469        overrun had occurred. This is a "should never occur" error, but it can be
1470        triggered by pathological patterns such as hundreds of nested parentheses.
1471        The checks now trigger 100 bytes before the end of the workspace.
1472    
1473    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1474    
1475    
1476    Version 8.01 19-Jan-2010
1477    ------------------------
1478    
1479    1.  If a pattern contained a conditional subpattern with only one branch (in
1480        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1481        computed the wrong minimum data length (which is of course zero for such
1482        subpatterns). This could cause incorrect "no match" results.
1483    
1484    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1485        the pattern is reset in the first branch, pcre_compile() failed with
1486        "internal error: code overflow at offset...". This happened only when
1487        the reset was to the original external option setting. (An optimization
1488        abstracts leading options settings into an external setting, which was the
1489        cause of this.)
1490    
1491    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1492        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1493        assertion pattern did not match (meaning that the assertion was true), it
1494        was incorrectly treated as false if the SKIP had been reached during the
1495        matching. This also applied to assertions used as conditions.
1496    
1497    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1498        assertion subpattern, including such a pattern used as a condition,
1499        unpredictable results occurred, instead of the error return
1500        PCRE_ERROR_DFA_UITEM.
1501    
1502    5.  The C++ GlobalReplace function was not working like Perl for the special
1503        situation when an empty string is matched. It now does the fancy magic
1504        stuff that is necessary.
1505    
1506    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1507        removed. (These were left over from very, very early versions of PCRE.)
1508    
1509    7.  Some cosmetic changes to the code to make life easier when compiling it
1510        as part of something else:
1511    
1512        (a) Change DEBUG to PCRE_DEBUG.
1513    
1514        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1515            called "current" as "current_branch", to prevent a collision with the
1516            Linux macro when compiled as a kernel module.
1517    
1518        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1519            prevent a collision with the Linux macro when compiled as a kernel
1520            module.
1521    
1522    8.  In pcre_compile() there are some checks for integer overflows that used to
1523        cast potentially large values to (double). This has been changed to that
1524        when building, a check for int64_t is made, and if it is found, it is used
1525        instead, thus avoiding the use of floating point arithmetic. (There is no
1526        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1527        double.
1528    
1529    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1530        2005 (difference between two addresses compared to an unsigned value).
1531    
1532    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1533        custom one, because of the following reported problem in Windows:
1534    
1535          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1536              under Win32.
1537          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1538              therefore missing the function definition.
1539          - The compiler thus generates a "C" signature for the test function.
1540          - The linker fails to find the "C" function.
1541          - PCRE fails to configure if asked to do so against libbz2.
1542    
1543    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1544        messages were output:
1545    
1546          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1547          rerunning libtoolize, to keep the correct libtool macros in-tree.
1548          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1549    
1550        I have done both of these things.
1551    
1552    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1553        most of the time, it *can* run out if it is given a pattern that contains a
1554        runaway infinite recursion. I updated the discussion in the pcrestack man
1555        page.
1556    
1557    13. Now that we have gone to the x.xx style of version numbers, the minor
1558        version may start with zero. Using 08 or 09 is a bad idea because users
1559        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1560        interpreted as invalid octal numbers. I've updated the previous comment in
1561        configure.ac, and also added a check that gives an error if 08 or 09 are
1562        used.
1563    
1564    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1565        causing partial matching to fail when the end of the subject matched \W
1566        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1567    
1568    15. There were some discrepancies between the declarations in pcre_internal.h
1569        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1570        their definitions. The declarations used "const uschar *" and the
1571        definitions used USPTR. Even though USPTR is normally defined as "const
1572        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1573        reported that: "This difference in casting confuses some C++ compilers, for
1574        example, SunCC recognizes above declarations as different functions and
1575        generates broken code for hbpcre." I have changed the declarations to use
1576        USPTR.
1577    
1578    16. GNU libtool is named differently on some systems. The autogen.sh script now
1579        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1580        (FreeBSD).
1581    
1582    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1583        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1584        comment: "Figure out how to create a longlong from a string: strtoll and
1585        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1586        instance, but it only takes 2 args instead of 3!"
1587    
1588    18. A subtle bug concerned with back references has been fixed by a change of
1589        specification, with a corresponding code fix. A pattern such as
1590        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1591        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1592        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1593        same bug. Such groups have to be quantified to be useful, or contained
1594        inside another quantified group. (If there's no repetition, the reference
1595        can never match.) The problem arises because, having left the group and
1596        moved on to the rest of the pattern, a later failure that backtracks into
1597        the group uses the captured value from the final iteration of the group
1598        rather than the correct earlier one. I have fixed this in PCRE by forcing
1599        any group that contains a reference to itself to be an atomic group; that
1600        is, there cannot be any backtracking into it once it has completed. This is
1601        similar to recursive and subroutine calls.
1602    
1603    
1604    Version 8.00 19-Oct-09
1605    ----------------------
1606    
1607    1.  The table for translating pcre_compile() error codes into POSIX error codes
1608        was out-of-date, and there was no check on the pcre_compile() error code
1609        being within the table. This could lead to an OK return being given in
1610        error.
1611    
1612    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1613        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1614        in a Windows environment.
1615    
1616    3.  The pcregrep --count option prints the count for each file even when it is
1617        zero, as does GNU grep. However, pcregrep was also printing all files when
1618        --files-with-matches was added. Now, when both options are given, it prints
1619        counts only for those files that have at least one match. (GNU grep just
1620        prints the file name in this circumstance, but including the count seems
1621        more useful - otherwise, why use --count?) Also ensured that the
1622        combination -clh just lists non-zero counts, with no names.
1623    
1624    4.  The long form of the pcregrep -F option was incorrectly implemented as
1625        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1626        but it seems right to fix it, and I didn't think it was worth preserving
1627        the old behaviour.
1628    
1629    5.  The command line items --regex=pattern and --regexp=pattern were not
1630        recognized by pcregrep, which required --regex pattern or --regexp pattern
1631        (with a space rather than an '='). The man page documented the '=' forms,
1632        which are compatible with GNU grep; these now work.
1633    
1634    6.  No libpcreposix.pc file was created for pkg-config; there was just
1635        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1636    
1637    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1638        when UCP support is not needed, by modifying the Python script that
1639        generates it from Unicode data files. This should not matter if the module
1640        is correctly used as a library, but I received one complaint about 50K of
1641        unwanted data. My guess is that the person linked everything into his
1642        program rather than using a library. Anyway, it does no harm.
1643    
1644    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1645        was a minimum greater than 1 for a wide character in a possessive
1646        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1647        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1648        character. Chaos in the form of incorrect output or a compiling loop could
1649        result.
1650    
1651    9.  The restrictions on what a pattern can contain when partial matching is
1652        requested for pcre_exec() have been removed. All patterns can now be
1653        partially matched by this function. In addition, if there are at least two
1654        slots in the offset vector, the offset of the earliest inspected character
1655        for the match and the offset of the end of the subject are set in them when
1656        PCRE_ERROR_PARTIAL is returned.
1657    
1658    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1659        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1660        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1661        and may be more useful for multi-segment matching.
1662    
1663    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1664        used to be given if ever the end of the subject was reached; now it is
1665        given only if matching could not proceed because another character was
1666        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1667        string "Z", which now yields "no match" instead of "partial match". In the
1668        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1669        final character ended with (*FAIL).
1670    
1671    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1672        if the pattern had a "must contain" character that was already found in the
1673        earlier partial match, unless partial matching was again requested. For
1674        example, with the pattern /dog.(body)?/, the "must contain" character is
1675        "g". If the first part-match was for the string "dog", restarting with
1676        "sbody" failed. This bug has been fixed.
1677    
1678    13. The string returned by pcre_dfa_exec() after a partial match has been
1679        changed so that it starts at the first inspected character rather than the
1680        first character of the match. This makes a difference only if the pattern
1681        starts with a lookbehind assertion or \b or \B (\K is not supported by
1682        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1683        matching functions compatible, and I think it's the right thing to do.
1684    
1685    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1686        so that the demonstration program is easily available in environments where
1687        PCRE has not been installed from source.
1688    
1689    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1690        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1691        library.
1692    
1693    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1694        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1695        is not the first non-POSIX option to be added. Clearly some people find
1696        these options useful.
1697    
1698    17. If a caller to the POSIX matching function regexec() passes a non-zero
1699        value for nmatch with a NULL value for pmatch, the value of
1700        nmatch is forced to zero.
1701    
1702    18. RunGrepTest did not have a test for the availability of the -u option of
1703        the diff command, as RunTest does. It now checks in the same way as
1704        RunTest, and also checks for the -b option.
1705    
1706    19. If an odd number of negated classes containing just a single character
1707        interposed, within parentheses, between a forward reference to a named
1708        subpattern and the definition of the subpattern, compilation crashed with
1709        an internal error, complaining that it could not find the referenced
1710        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1711        [The bug was that it was starting one character too far in when skipping
1712        over the character class, thus treating the ] as data rather than
1713        terminating the class. This meant it could skip too much.]
1714    
1715    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1716        /g option in pcretest when the pattern contains \K, which makes it possible
1717        to have an empty string match not at the start, even when the pattern is
1718        anchored. Updated pcretest and pcredemo to use this option.
1719    
1720    21. If the maximum number of capturing subpatterns in a recursion was greater
1721        than the maximum at the outer level, the higher number was returned, but
1722        with unset values at the outer level. The correct (outer level) value is
1723        now given.
1724    
1725    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1726        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1727        make it do so. The string so far is captured, making this feature
1728        compatible with Perl.
1729    
1730    23. The tests have been re-organized, adding tests 11 and 12, to make it
1731        possible to check the Perl 5.10 features against Perl 5.10.
1732    
1733    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1734        pattern matches a fixed length string. PCRE did not allow this; now it
1735        does. Neither allows recursion.
1736    
1737    25. I finally figured out how to implement a request to provide the minimum
1738        length of subject string that was needed in order to match a given pattern.
1739        (It was back references and recursion that I had previously got hung up
1740        on.) This code has now been added to pcre_study(); it finds a lower bound
1741        to the length of subject needed. It is not necessarily the greatest lower
1742        bound, but using it to avoid searching strings that are too short does give
1743        some useful speed-ups. The value is available to calling programs via
1744        pcre_fullinfo().
1745    
1746    26. While implementing 25, I discovered to my embarrassment that pcretest had
1747        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1748        study optimizations had never been tested with that matching function.
1749        Oops. What is worse, even when it was passed study data, there was a bug in
1750        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1751        were also very few tests of studied patterns with pcre_dfa_exec().
1752    
1753    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1754        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1755        on the other side of the coin, they are no longer allowed to have different
1756        names, because these cannot be distinguished in PCRE, and this has caused
1757        confusion. (This is a difference from Perl.)
1758    
1759    28. When duplicate subpattern names are present (necessarily with different
1760        numbers, as required by 27 above), and a test is made by name in a
1761        conditional pattern, either for a subpattern having been matched, or for
1762        recursion in such a pattern, all the associated numbered subpatterns are
1763        tested, and the overall condition is true if the condition is true for any
1764        one of them. This is the way Perl works, and is also more like the way
1765        testing by number works.
1766    
1767    
1768    Version 7.9 11-Apr-09
1769    ---------------------
1770    
1771    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1772        (pcretest), all targets were linked against these libraries. This included
1773        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1774        libraries. This caused unwanted dependencies to be created. This problem
1775        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1776        pcretest is linked with readline.
1777    
1778    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1779        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1780        moved outside it again, because FALSE and TRUE are already defined in AIX,
1781        but BOOL is not.
1782    
1783    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1784        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1785    
1786    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1787        hyphens) following file names and line numbers when outputting matching
1788        lines. This is not true; no spaces are inserted. I have also clarified the
1789        wording for the --colour (or --color) option.
1790    
1791    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1792        was not coloured; this is different to GNU grep, so I have changed it to be
1793        the same.
1794    
1795    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1796        each matching line was coloured. Now it goes on to look for further matches
1797        of any of the test patterns, which is the same behaviour as GNU grep.
1798    
1799    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1800        doesn't make sense to accept an empty string match in pcregrep, so I have
1801        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1802        seems to be how GNU grep behaves. [But see later change 40 for release
1803        8.33.]
1804    
1805    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1806        start or after a newline", because the conditional assertion was not being
1807        correctly handled. The rule now is that both the assertion and what follows
1808        in the first alternative must satisfy the test.
1809    
1810    9.  If auto-callout was enabled in a pattern with a conditional group whose
1811        condition was an assertion, PCRE could crash during matching, both with
1812        pcre_exec() and pcre_dfa_exec().
1813    
1814    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1815        used for matching.
1816    
1817    11. Unicode property support in character classes was not working for
1818        characters (bytes) greater than 127 when not in UTF-8 mode.
1819    
1820    12. Added the -M command line option to pcretest.
1821    
1822    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1823    
1824    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1825    
1826    16. Added comments and documentation about mis-use of no_arg in the C++
1827        wrapper.
1828    
1829    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1830        from Martin Jerabek that uses macro names for all relevant character and
1831        string constants.
1832    
1833    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1834        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1835        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1836        these, but not everybody uses configure.
1837    
1838    19. A conditional group that had only one branch was not being correctly
1839        recognized as an item that could match an empty string. This meant that an
1840        enclosing group might also not be so recognized, causing infinite looping
1841        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1842        with the subject "ab", where knowledge that the repeated group can match
1843        nothing is needed in order to break the loop.
1844    
1845    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1846        exec(), but without supplying a callout function, matching went wrong.
1847    
1848    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1849        leak if the size of the offset vector was greater than 30. When the vector
1850        is smaller, the saved offsets during recursion go onto a local stack
1851        vector, but for larger vectors malloc() is used. It was failing to free
1852        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1853        error, in fact).
1854    
1855    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1856        heapframe that is used only when UTF-8 support is enabled. This caused no
1857        problem, but was untidy.
1858    
1859    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1860        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1861        included within another project.
1862    
1863    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1864        slightly modified by me:
1865    
1866          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1867              not building pcregrep.
1868    
1869          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1870              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1871    
1872    25. Forward references, both numeric and by name, in patterns that made use of
1873        duplicate group numbers, could behave incorrectly or give incorrect errors,
1874        because when scanning forward to find the reference group, PCRE was not
1875        taking into account the duplicate group numbers. A pattern such as
1876        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1877    
1878    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1879        the feature of a custom pointer more persuasive (as requested by a user).
1880    
1881    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1882        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1883        pre-defined.
1884    
1885    28. Added support for (*UTF8) at the start of a pattern.
1886    
1887    29. Arrange for flags added by the "release type" setting in CMake to be shown
1888        in the configuration summary.
1889    
1890    
1891    Version 7.8 05-Sep-08
1892    ---------------------
1893    
1894    1.  Replaced UCP searching code with optimized version as implemented for Ad
1895        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1896        stage table and inline lookup instead of a function, giving speed ups of 2
1897        to 5 times on some simple patterns that I tested. Permission was given to
1898        distribute the MultiStage2.py script that generates the tables (it's not in
1899        the tarball, but is in the Subversion repository).
1900    
1901    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1902        scripts.
1903    
1904    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1905        a group with a zero qualifier. The result of the study could be incorrect,
1906        or the function might crash, depending on the pattern.
1907    
1908    4.  Caseless matching was not working for non-ASCII characters in back
1909        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1910        It now works when Unicode Property Support is available.
1911    
1912    5.  In pcretest, an escape such as \x{de} in the data was always generating
1913        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1914        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1915        truncation.
1916    
1917    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1918    
1919    7.  Added two (int) casts to pcregrep when printing the difference of two
1920        pointers, in case they are 64-bit values.
1921    
1922    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1923        test 2 if it fails.
1924    
1925    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1926        and a #define of that name to empty if it is not externally set. This is to
1927        allow users of MSVC to set it if necessary.
1928    
1929    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1930        the convenience functions in the pcre_get.c source file.
1931    
1932    11. An option change at the start of a pattern that had top-level alternatives
1933        could cause overwriting and/or a crash. This command provoked a crash in
1934        some environments:
1935    
1936          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1937    
1938        This potential security problem was recorded as CVE-2008-2371.
1939    
1940    12. For a pattern where the match had to start at the beginning or immediately
1941        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1942        pcre_dfa_exec() could read past the end of the passed subject if there was
1943        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1944        pcretest so that it places the subject at the end of its malloc-ed buffer.
1945    
1946    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1947        exec() might read past the end of the data buffer in UTF-8 mode.
1948    
1949    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1950        the data contained the byte 0x85 as part of a UTF-8 character within its
1951        first line. This applied both to normal and DFA matching.
1952    
1953    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1954        /^[^d]*?$/8 failed to match "abc".
1955    
1956    16. Added a missing copyright notice to pcrecpp_internal.h.
1957    
1958    17. Make it more clear in the documentation that values returned from
1959        pcre_exec() in ovector are byte offsets, not character counts.
1960    
1961    18. Tidied a few places to stop certain compilers from issuing warnings.
1962    
1963    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1964        supplied by Stefan Weber. I made a further small update for 7.8 because
1965        there is a change of source arrangements: the pcre_searchfuncs.c module is
1966        replaced by pcre_ucd.c.
1967    
1968    
1969    Version 7.7 07-May-08
1970    ---------------------
1971    
1972    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1973        a string to a long long, pretend we don't even have a long long." This is
1974        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1975    
1976    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1977        pre-7.6 versions, which defined a global no_arg variable instead of putting
1978        it in the RE class. (See also #8 below.)
1979    
1980    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1981        Lopes.
1982    
1983    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1984    
1985        (1) The include/exclude patterns were being applied to the whole pathnames
1986            of files, instead of just to the final components.
1987    
1988        (2) If there was more than one level of directory, the subdirectories were
1989            skipped unless they satisfied the include/exclude conditions. This is
1990            inconsistent with GNU grep (and could even be seen as contrary to the
1991            pcregrep specification - which I improved to make it absolutely clear).
1992            The action now is always to scan all levels of directory, and just
1993            apply the include/exclude patterns to regular files.
1994    
1995    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1996        --exclude_dir in the tests to avoid scanning .svn directories.
1997    
1998    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1999        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
2000        doesn't support NULs in patterns.
2001    
2002    7.  Added some missing "const"s to declarations of static tables in
2003        pcre_compile.c and pcre_dfa_exec.c.
2004    
2005    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
2006        caused by fix #2  above. (Subsequently also a second patch to fix the
2007        first patch. And a third patch - this was a messy problem.)
2008    
2009    9.  Applied Craig's patch to remove the use of push_back().
2010    
2011    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
2012        matching function regexec().
2013    
2014    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
2015        which, however, unlike Perl's \g{...}, are subroutine calls, not back
2016        references. PCRE supports relative numbers with this syntax (I don't think
2017        Oniguruma does).
2018    
2019    12. Previously, a group with a zero repeat such as (...){0} was completely
2020        omitted from the compiled regex. However, this means that if the group
2021        was called as a subroutine from elsewhere in the pattern, things went wrong
2022        (an internal error was given). Such groups are now left in the compiled
2023        pattern, with a new opcode that causes them to be skipped at execution
2024        time.
2025    
2026    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
2027        to the way PCRE behaves:
2028    
2029        (a) A lone ] character is dis-allowed (Perl treats it as data).
2030    
2031        (b) A back reference to an unmatched subpattern matches an empty string
2032            (Perl fails the current match path).
2033    
2034        (c) A data ] in a character class must be notated as \] because if the
2035            first data character in a class is ], it defines an empty class. (In
2036            Perl it is not possible to have an empty class.) The empty class []
2037            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
2038            The negative empty class [^] matches any one character, independently
2039            of the DOTALL setting.
2040    
2041    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
2042        non-existent subpattern following a character class starting with ']' and
2043        containing () gave an internal compiling error instead of "reference to
2044        non-existent subpattern". Fortunately, when the pattern did exist, the
2045        compiled code was correct. (When scanning forwards to check for the
2046        existence of the subpattern, it was treating the data ']' as terminating
2047        the class, so got the count wrong. When actually compiling, the reference
2048        was subsequently set up correctly.)
2049    
2050    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
2051        it was being rejected as not supported by pcre_dfa_exec(), even though
2052        other assertions are supported. I have made pcre_dfa_exec() support
2053        (*FAIL).
2054    
2055    16. The implementation of 13c above involved the invention of a new opcode,
2056        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
2057        cannot be changed at match time, I realized I could make a small
2058        improvement to matching performance by compiling OP_ALLANY instead of
2059        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
2060        on the OP_ANY path.
2061    
2062    17. Compiling pcretest on Windows with readline support failed without the
2063        following two fixes: (1) Make the unistd.h include conditional on
2064        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
2065    
2066    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
2067        ncurses library to be included for pcretest when ReadLine support is
2068        requested, but also to allow for it to be overridden. This patch came from
2069        Daniel Bergström.
2070    
2071    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
2072        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
2073        any errors with the current Unicode tables. Thanks to Peter Kankowski for
2074        spotting this.
2075    
2076    
2077    Version 7.6 28-Jan-08
2078    ---------------------
2079    
2080    1.  A character class containing a very large number of characters with
2081        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
2082        overflow.
2083    
2084    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
2085        HAVE_LONG_LONG is not defined.
2086    
2087    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
2088        bring them up to date and include new features. This patch includes:
2089    
2090        - Fixed PH's badly added libz and libbz2 support.
2091        - Fixed a problem with static linking.
2092        - Added pcredemo. [But later removed - see 7 below.]
2093        - Fixed dftables problem and added an option.
2094        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
2095            HAVE_LONG_LONG.
2096        - Added readline support for pcretest.
2097        - Added an listing of the option settings after cmake has run.
2098    
2099    4.  A user submitted a patch to Makefile that makes it easy to create
2100        "pcre.dll" under mingw when using Configure/Make. I added stuff to
2101        Makefile.am that cause it to include this special target, without
2102        affecting anything else. Note that the same mingw target plus all
2103        the other distribution libraries and programs are now supported
2104        when configuring with CMake (see 6 below) instead of with
2105        Configure/Make.
2106    
2107    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
2108        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
2109        exported in the Windows port". It has not yet been confirmed that the patch
2110        solves the problem, but it does no harm.
2111    
2112    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
2113        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
2114        with CMake, and also correct the comment about stack recursion.
2115    
2116    7.  Remove the automatic building of pcredemo from the ./configure system and
2117        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
2118        of a program that users should build themselves after PCRE is installed, so
2119        building it automatically is not really right. What is more, it gave
2120        trouble in some build environments.
2121    
2122    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
2123    
2124    
2125    Version 7.5 10-Jan-08
2126    ---------------------
2127    
2128    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
2129        values in parens when parsing an RE using the C++ wrapper."
2130    
2131    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
2132        Characters greater than 255 were excluded from the class instead of being
2133        included.
2134    
2135    3.  The same bug as (2) above applied to negated POSIX classes such as
2136        [:^space:].
2137    
2138    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
2139        defined or documented. It seems to have been a typo for PCRE_STATIC, so
2140        I have changed it.
2141    
2142    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
2143        first named subpattern) and a construct such as (?&a) would reference the
2144        first named subpattern whose name started with "a" (in other words, the
2145        length check was missing). Both these problems are fixed. "Subpattern name
2146        expected" is now given for (?&) (a zero-length name), and this patch also
2147        makes it give the same error for \k'' (previously it complained that that
2148        was a reference to a non-existent subpattern).
2149    
2150    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
2151        this is right because (?- can be followed by option settings as well as by
2152        digits. I have, however, made the messages clearer.
2153    
2154    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
2155        than the number used in the conditional) now cause a compile-time error.
2156        This is actually not compatible with Perl, which accepts such patterns, but
2157        treats the conditional as always being FALSE (as PCRE used to), but it
2158        seems to me that giving a diagnostic is better.
2159    
2160    8.  Change "alphameric" to the more common word "alphanumeric" in comments
2161        and messages.
2162    
2163    9.  Fix two occurrences of "backslash" in comments that should have been
2164        "backspace".
2165    
2166    10. Remove two redundant lines of code that can never be obeyed (their function
2167        was moved elsewhere).
2168    
2169    11. The program that makes PCRE's Unicode character property table had a bug
2170        which caused it to generate incorrect table entries for sequences of
2171        characters that have the same character type, but are in different scripts.
2172        It amalgamated them into a single range, with the script of the first of
2173        them. In other words, some characters were in the wrong script. There were
2174        thirteen such cases, affecting characters in the following ranges:
2175    
2176          U+002b0 - U+002c1
2177          U+0060c - U+0060d
2178          U+0061e - U+00612
2179          U+0064b - U+0065e
2180          U+0074d - U+0076d
2181          U+01800 - U+01805
2182          U+01d00 - U+01d77
2183          U+01d9b - U+01dbf
2184          U+0200b - U+0200f
2185          U+030fc - U+030fe
2186          U+03260 - U+0327f
2187          U+0fb46 - U+0fbb1
2188          U+10450 - U+1049d
2189    
2190    12. The -o option (show only the matching part of a line) for pcregrep was not
2191        compatible with GNU grep in that, if there was more than one match in a
2192        line, it showed only the first of them. It now behaves in the same way as
2193        GNU grep.
2194    
2195    13. If the -o and -v options were combined for pcregrep, it printed a blank
2196        line for every non-matching line. GNU grep prints nothing, and pcregrep now
2197        does the same. The return code can be used to tell if there were any
2198        non-matching lines.
2199    
2200    14. Added --file-offsets and --line-offsets to pcregrep.
2201    
2202    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
2203        infinitely looping recursion. The bug was that positive lookaheads were not
2204        being skipped when checking for a possible empty match (negative lookaheads
2205        and both kinds of lookbehind were skipped).
2206    
2207    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
2208        inclusion of <windows.h> to before rather than after the definition of
2209        INVALID_FILE_ATTRIBUTES (patch from David Byron).
2210    
2211    17. Specifying a possessive quantifier with a specific limit for a Unicode
2212        character property caused pcre_compile() to compile bad code, which led at
2213        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
2214        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
2215        caused the error; without that there was no problem.
2216    
2217    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
2218    
2219    19. Added --enable-pcretest-libreadline.
2220    
2221    20. In pcrecpp.cc, the variable 'count' was incremented twice in
2222        RE::GlobalReplace(). As a result, the number of replacements returned was
2223        double what it should be. I removed one of the increments, but Craig sent a
2224        later patch that removed the other one (the right fix) and added unit tests
2225        that check the return values (which was not done before).
2226    
2227    21. Several CMake things:
2228    
2229        (1) Arranged that, when cmake is used on Unix, the libraries end up with
2230            the names libpcre and libpcreposix, not just pcre and pcreposix.
2231    
2232        (2) The above change means that pcretest and pcregrep are now correctly
2233            linked with the newly-built libraries, not previously installed ones.
2234    
2235        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
2236    
2237    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
2238        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
2239        UTF-8 newline character). The key issue is that the pattern starts .*;
2240        this means that the match must be either at the beginning, or after a
2241        newline. The bug was in the code for advancing after a failed match and
2242        checking that the new position followed a newline. It was not taking
2243        account of UTF-8 characters correctly.
2244    
2245    23. PCRE was behaving differently from Perl in the way it recognized POSIX
2246        character classes. PCRE was not treating the sequence [:...:] as a
2247        character class unless the ... were all letters. Perl, however, seems to
2248        allow any characters between [: and :], though of course it rejects as
2249        unknown any "names" that contain non-letters, because all the known class
2250        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
2251        for example, whereas PCRE did not - it did not recognize a POSIX character
2252        class. This seemed a bit dangerous, so the code has been changed to be
2253        closer to Perl. The behaviour is not identical to Perl, because PCRE will
2254        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
2255        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
2256        Perl does, and where it didn't before.
2257    
2258    24. Rewrite so as to remove the single use of %n from pcregrep because in some
2259        Windows environments %n is disabled by default.
2260    
2261    
2262    Version 7.4 21-Sep-07
2263    ---------------------
2264    
2265    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
2266        means that a class such as [\s] counted as "explicit reference to CR or
2267        LF". That isn't really right - the whole point of the change was to try to
2268        help when there was an actual mention of one of the two characters. So now
2269        the change happens only if \r or \n (or a literal CR or LF) character is
2270        encountered.
2271    
2272    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
2273        of both had grown to the point where there were only 3 bits left.
2274        Fortunately, there was spare space in the data structure, and so I have
2275        moved the internal flags into a new 16-bit field to free up more option
2276        bits.
2277    
2278    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
2279        but did not set the internal JCHANGED flag - either of these is enough to
2280        control the way the "get" function works - but the PCRE_INFO_JCHANGED
2281        facility is supposed to tell if (?J) was ever used, so now (?J) at the
2282        start sets both bits.
2283    
2284    4.  Added options (at build time, compile time, exec time) to change \R from
2285        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
2286    
2287    5.  doc/pcresyntax.html was missing from the distribution.
2288    
2289    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
2290        compatibility, even though it is no longer used.
2291    
2292    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
2293        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
2294        windows.h file is present (where different names are used). [This was
2295        reversed later after testing - see 16 below.]
2296    
2297    8.  Changed all #include <config.h> to #include "config.h". There were also
2298        some further <pcre.h> cases that I changed to "pcre.h".
2299    
2300    9.  When pcregrep was used with the --colour option, it missed the line ending
2301        sequence off the lines that it output.
2302    
2303    10. It was pointed out to me that arrays of string pointers cause lots of
2304        relocations when a shared library is dynamically loaded. A technique of
2305        using a single long string with a table of offsets can drastically reduce
2306        these. I have refactored PCRE in four places to do this. The result is
2307        dramatic:
2308    
2309          Originally:                          290
2310          After changing UCP table:            187
2311          After changing error message table:   43
2312          After changing table of "verbs"       36
2313          After changing table of Posix names   22
2314    
2315        Thanks to the folks working on Gregex for glib for this insight.
2316    
2317    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
2318        unicode-properties was also set.
2319    
2320    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
2321    
2322    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
2323        checked only for CRLF.
2324    
2325    14. Added casts to pcretest.c to avoid compiler warnings.
2326    
2327    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
2328    
2329    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
2330        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2331        entirely. This removes changes made in 7 above.
2332    
2333    17. The CMake files have been updated, and there is now more information about
2334        building with CMake in the NON-UNIX-USE document.
2335    
2336    
2337    Version 7.3 28-Aug-07
2338  ---------------------  ---------------------
2339    
2340   1. In the rejigging of the build system that eventually resulted in 7.1, the   1. In the rejigging of the build system that eventually resulted in 7.1, the
# Line 29  Version 7.3 05-Jul-07 Line 2362  Version 7.3 05-Jul-07
2362   4. Debugging: adjusted the names of opcodes for different kinds of parentheses   4. Debugging: adjusted the names of opcodes for different kinds of parentheses
2363      in debug output.      in debug output.
2364    
2365   5. Arrange to use "%I64d" instead of "%lld" for long printing in the pcrecpp   5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
2366      unittest when running under MinGW.      long printing in the pcrecpp unittest when running under MinGW.
2367    
2368   6. ESC_K was left out of the EBCDIC table.   6. ESC_K was left out of the EBCDIC table.
2369    
# Line 53  Version 7.3 05-Jul-07 Line 2386  Version 7.3 05-Jul-07
2386      pcreposix and pcrecpp parts of the library. These were overlooked when this      pcreposix and pcrecpp parts of the library. These were overlooked when this
2387      problem was solved for the main library.      problem was solved for the main library.
2388    
2389     9. There were some crude static tests to avoid integer overflow when computing
2390        the size of patterns that contain repeated groups with explicit upper
2391        limits. As the maximum quantifier is 65535, the maximum group length was
2392        set at 30,000 so that the product of these two numbers did not overflow a
2393        32-bit integer. However, it turns out that people want to use groups that
2394        are longer than 30,000 bytes (though not repeat them that many times).
2395        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
2396        made it possible to implement the integer overflow checks in a much more
2397        dynamic way, which I have now done. The artificial limitation on group
2398        length has been removed - we now have only the limit on the total length of
2399        the compiled pattern, which depends on the LINK_SIZE setting.
2400    
2401    10. Fixed a bug in the documentation for get/copy named substring when
2402        duplicate names are permitted. If none of the named substrings are set, the
2403        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
2404        empty string.
2405    
2406    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
2407        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
2408        because the ] is interpreted as the first data character and the
2409        terminating ] is not found. PCRE has been made compatible with Perl in this
2410        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
2411        cause memory overwriting.
2412    
2413    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
2414        string has been matched (to stop an infinite loop). It was not recognizing
2415        a conditional subpattern that could match an empty string if that
2416        subpattern was within another subpattern. For example, it looped when
2417        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
2418        condition was not nested. This bug has been fixed.
2419    
2420    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
2421        past the start of the subject in the presence of bytes with the top bit
2422        set, for example "\x8aBCD".
2423    
2424    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
2425        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
2426    
2427    14. Optimized (?!) to (*FAIL).
2428    
2429    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
2430        This restricts code points to be within the range 0 to 0x10FFFF, excluding
2431        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
2432        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
2433        does: it's just the validity check that is more restrictive.
2434    
2435    16. Inserted checks for integer overflows during escape sequence (backslash)
2436        processing, and also fixed erroneous offset values for syntax errors during
2437        backslash processing.
2438    
2439    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
2440        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
2441    
2442    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
2443        caused an overrun.
2444    
2445    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
2446        something other than just ASCII characters) inside a group that had an
2447        unlimited repeat caused a loop at compile time (while checking to see
2448        whether the group could match an empty string).
2449    
2450    20. Debugging a pattern containing \p or \P could cause a crash. For example,
2451        [\P{Any}] did so. (Error in the code for printing property names.)
2452    
2453    21. An orphan \E inside a character class could cause a crash.
2454    
2455    22. A repeated capturing bracket such as (A)? could cause a wild memory
2456        reference during compilation.
2457    
2458    23. There are several functions in pcre_compile() that scan along a compiled
2459        expression for various reasons (e.g. to see if it's fixed length for look
2460        behind). There were bugs in these functions when a repeated \p or \P was
2461        present in the pattern. These operators have additional parameters compared
2462        with \d, etc, and these were not being taken into account when moving along
2463        the compiled data. Specifically:
2464    
2465        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2466            length.
2467    
2468        (b) An item such as \pL+ within a repeated group could cause crashes or
2469            loops.
2470    
2471        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2472            "reference to non-existent subpattern" error.
2473    
2474        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2475    
2476    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2477        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2478    
2479    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2480    
2481    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2482        character were causing crashes (broken optimization).
2483    
2484    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2485        \p or \P) caused a compile-time loop.
2486    
2487    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2488        break. For example, the unstudied pattern [\r\n]A does not match the string
2489        "\r\nA" because change 7.0/46 below moves the current point on by two
2490        characters after failing to match at the start. However, the pattern \nA
2491        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2492        the same is true. There doesn't seem any very clean way out of this, but
2493        what I have chosen to do makes the common cases work: PCRE now takes note
2494        of whether there can be an explicit match for \r or \n anywhere in the
2495        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2496        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2497        pattern has explicit CR or LF references.
2498    
2499    29. Added (*CR) etc for changing newline setting at start of pattern.
2500    
2501    
2502  Version 7.2 19-Jun-07  Version 7.2 19-Jun-07
2503  ---------------------  ---------------------

Legend:
Removed from v.200  
changed lines
  Added in v.1370

  ViewVC Help
Powered by ViewVC 1.1.5