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

Diff of /code/trunk/ChangeLog

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

revision 361 by ph10, Thu Jul 10 16:03:28 2008 UTC revision 1395 by ph10, Sat Nov 9 16:54:52 2013 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 8.0 02 Jul-08  Version 8.34 xx-xxxx-201x
5    --------------------------
6    
7    1.  Add pcre[16|32]_jit_free_unused_memory to forcibly free unused JIT
8        executable memory. Patch inspired by Carsten Klein.
9    
10    2.  ./configure --enable-coverage defined SUPPORT_GCOV in config.h, although
11        this macro is never tested and has no effect, because the work to support
12        coverage involves only compiling and linking options and special targets in
13        the Makefile. The comment in config.h implied that defining the macro would
14        enable coverage support, which is totally false. There was also support for
15        setting this macro in the CMake files (my fault, I just copied it from
16        configure). SUPPORT_GCOV has now been removed.
17    
18    3.  Make a small performance improvement in strlen16() and strlen32() in
19        pcretest.
20    
21    4.  Change 36 for 8.33 left some unreachable statements in pcre_exec.c,
22        detected by the Solaris compiler (gcc doesn't seem to be able to diagnose
23        these cases). There was also one in pcretest.c.
24    
25    5.  Cleaned up a "may be uninitialized" compiler warning in pcre_exec.c.
26    
27    6.  In UTF mode, the code for checking whether a group could match an empty
28        string (which is used for indefinitely repeated groups to allow for
29        breaking an infinite loop) was broken when the group contained a repeated
30        negated single-character class with a character that occupied more than one
31        data item and had a minimum repetition of zero (for example, [^\x{100}]* in
32        UTF-8 mode). The effect was undefined: the group might or might not be
33        deemed as matching an empty string, or the program might have crashed.
34    
35    7.  The code for checking whether a group could match an empty string was not
36        recognizing that \h, \H, \v, \V, and \R must match a character.
37    
38    8.  Implemented PCRE_INFO_MATCH_EMPTY, which yields 1 if the pattern can match
39        an empty string. If it can, pcretest shows this in its information output.
40    
41    9.  Fixed two related bugs that applied to Unicode extended grapheme clusters
42        that were repeated with a maximizing qualifier (e.g. \X* or \X{2,5}) when
43        matched by pcre_exec() without using JIT:
44    
45        (a) If the rest of the pattern did not match after a maximal run of
46            grapheme clusters, the code for backing up to try with fewer of them
47            did not always back up over a full grapheme when characters that do not
48            have the modifier quality were involved, e.g. Hangul syllables.
49    
50        (b) If the match point in a subject started with modifier character, and
51            there was no match, the code could incorrectly back up beyond the match
52            point, and potentially beyond the first character in the subject,
53            leading to a segfault or an incorrect match result.
54    
55    10. A conditional group with an assertion condition could lead to PCRE
56        recording an incorrect first data item for a match if no other first data
57        item was recorded. For example, the pattern (?(?=ab)ab) recorded "a" as a
58        first data item, and therefore matched "ca" after "c" instead of at the
59        start.
60    
61    11. Change 40 for 8.33 (allowing pcregrep to find empty strings) showed up a
62        bug that caused the command "echo a | ./pcregrep -M '|a'" to loop.
63    
64    12. The source of pcregrep now includes z/OS-specific code so that it can be
65        compiled for z/OS as part of the special z/OS distribution.
66    
67    13. Added the -T and -TM options to pcretest.
68    
69    14. The code in pcre_compile.c for creating the table of named capturing groups
70        has been refactored. Instead of creating the table dynamically during the
71        actual compiling pass, the information is remembered during the pre-compile
72        pass (on the stack unless there are more than 20 named groups, in which
73        case malloc() is used) and the whole table is created before the actual
74        compile happens. This has simplified the code (it is now nearly 150 lines
75        shorter) and prepared the way for better handling of references to groups
76        with duplicate names.
77    
78    15. A back reference to a named subpattern when there is more than one of the
79        same name now checks them in the order in which they appear in the pattern.
80        The first one that is set is used for the reference. Previously only the
81        first one was inspected. This change makes PCRE more compatible with Perl.
82    
83    16. Unicode character properties were updated from Unicode 6.3.0.
84    
85    17. The compile-time code for auto-possessification has been refactored, based
86        on a patch by Zoltan Herczeg. It now happens after instead of during
87        compilation. The code is cleaner, and more cases are handled. The option
88        PCRE_NO_AUTO_POSSESSIFY is added for testing purposes, and the -O and /O
89        options in pcretest are provided to set it. It can also be set by
90        (*NO_AUTO_POSSESS) at the start of a pattern.
91    
92    18. The character VT has been added to the set of characters that match \s and
93        are generally treated as white space, following this same change in Perl
94        5.18. There is now no difference between "Perl space" and "POSIX space".
95    
96    19. The code for checking named groups as conditions, either for being set or
97        for being recursed, has been refactored (this is related to 14 and 15
98        above). Processing unduplicated named groups should now be as fast at
99        numerical groups, and processing duplicated groups should be faster than
100        before.
101    
102    20. Two patches to the CMake build system, by Alexander Barkov:
103    
104          (1) Replace the "source" command by "." in CMakeLists.txt because
105              "source" is a bash-ism.
106    
107          (2) Add missing HAVE_STDINT_H and HAVE_INTTYPES_H to config-cmake.h.in;
108              without these the CMake build does not work on Solaris.
109    
110    21. Perl has changed its handling of \8 and \9. If there is no previously
111        encountered capturing group of those numbers, they are treated as the
112        literal characters 8 and 9 instead of a binary zero followed by the
113        literals. PCRE now does the same.
114    
115    22. Following Perl, added \o{} to specify codepoints in octal, making it
116        possible to specify values greater than 0777 and also making them
117        unambiguous.
118    
119    23. Perl now gives an error for missing closing braces after \x{... instead of
120        treating the string as literal. PCRE now does the same.
121    
122    24. RunTest used to grumble if an inappropriate test was selected explicitly,
123        but just skip it when running all tests. This make it awkward to run ranges
124        of tests when one of them was inappropriate. Now it just skips any
125        inappropriate tests, as it always did when running all tests.
126    
127    25. If PCRE_AUTO_CALLOUT and PCRE_UCP were set for a pattern that contained
128        character types such as \d or \w, too many callouts were inserted, and the
129        data that they returned was rubbish.
130    
131    26. In UCP mode, \s was not matching two of the characters that Perl matches,
132        namely NEL (U+0085) and MONGOLIAN VOWEL SEPARATOR (U+180E), though they
133        were matched by \h. The code has now been refactored so that the lists of
134        the horizontal and vertical whitespace characters used for \h and \v (which
135        are defined only in one place) are now also used for \s.
136    
137    27. Add JIT support for the 64 bit TileGX architecture.
138        Patch by Jiong Wang (Tilera Corporation).
139    
140    28. Possessive quantifiers for classes (both explicit and automatically
141        generated) now use special opcodes instead of wrapping in ONCE brackets.
142    
143    29. Whereas an item such as A{4}+ ignored the possessivenes of the quantifier
144        (because it's meaningless), this was not happening when PCRE_CASELESS was
145        set. Not wrong, but inefficient.
146    
147    30. Updated perltest.pl to add /u (force Unicode mode) when /W (use Unicode
148        properties for \w, \d, etc) is present in a test regex. Otherwise if the
149        test contains no characters greater than 255, Perl doesn't realise it
150        should be using Unicode semantics.
151    
152    31. Upgraded the handling of the POSIX classes [:graph:], [:print:], and
153        [:punct:] when PCRE_UCP is set so as to include the same characters as Perl
154        does in Unicode mode.
155    
156    32. Added the "forbid" facility to pcretest so that putting tests into the
157        wrong test files can sometimes be quickly detected.
158    
159    33. There is now a limit (default 250) on the depth of nesting of parentheses.
160        This limit is imposed to control the amount of system stack used at compile
161        time. It can be changed at build time by --with-parens-nest-limit=xxx or
162        the equivalent in CMake.
163    
164    34. Character classes such as [A-\d] or [a-[:digit:]] now cause compile-time
165        errors. Perl warns for these when in warning mode, but PCRE has no facility
166        for giving warnings.
167    
168    35. Change 34 for 8.13 allowed quantifiers on assertions, because Perl does.
169        However, this was not working for (?!) because it is optimized to (*FAIL),
170        for which PCRE does not allow quantifiers. The optimization is now disabled
171        when a quantifier follows (?!). I can't see any use for this, but it makes
172        things uniform.
173    
174    36. Perl no longer allows group names to start with digits, so I have made this
175        change also in PCRE. It simplifies the code a bit.
176    
177    
178    Version 8.33 28-May-2013
179    ------------------------
180    
181    1.  Added 'U' to some constants that are compared to unsigned integers, to
182        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
183        variables that are added to signed variables, to ensure the result is
184        signed and can be negated.
185    
186    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
187        CMake config files.
188    
189    3.  Revise the creation of config.h.generic so that all boolean macros are
190        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
191        overriding via -D on the command line possible.
192    
193    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
194        to unsigned int is reported to make a quite noticeable speed difference in
195        a specific Windows environment. Testing on Linux did also appear to show
196        some benefit (and it is clearly not harmful). Also fixed the definition of
197        Xop which should be unsigned.
198    
199    5.  Related to (4), changing the definition of the intermediate variable cc
200        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
201        improvements.
202    
203    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
204        unnecessary spaces.
205    
206    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
207        and later.
208    
209    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
210    
211    9.  Optimizing fast_forward_start_bits in JIT.
212    
213    10. Adding support for callouts in JIT, and fixing some issues revealed
214        during this work. Namely:
215    
216        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
217    
218        (b) Minimum length was not checked before the matching is started.
219    
220    11. The value of capture_last that is passed to callouts was incorrect in some
221        cases when there was a capture on one path that was subsequently abandoned
222        after a backtrack. Also, the capture_last value is now reset after a
223        recursion, since all captures are also reset in this case.
224    
225    12. The interpreter no longer returns the "too many substrings" error in the
226        case when an overflowing capture is in a branch that is subsequently
227        abandoned after a backtrack.
228    
229    13. In the pathological case when an offset vector of size 2 is used, pcretest
230        now prints out the matched string after a yield of 0 or 1.
231    
232    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
233        Only supported by the JIT compiler at the moment.
234    
235    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
236    
237    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
238        offset of the starting point of the matching process, provided the offsets
239        vector is large enough.
240    
241    17. The \A escape now records a lookbehind value of 1, though its execution
242        does not actually inspect the previous character. This is to ensure that,
243        in partial multi-segment matching, at least one character from the old
244        segment is retained when a new segment is processed. Otherwise, if there
245        are no lookbehinds in the pattern, \A might match incorrectly at the start
246        of a new segment.
247    
248    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
249    
250    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
251        gives some modest performance improvement in 8-bit mode.
252    
253    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
254        be expressed in certain programming languages using Universal Character
255        Names.
256    
257    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
258        which points out that "non characters" are not "characters that may not
259        appear in Unicode strings" but rather "characters that are reserved for
260        internal use and have only local meaning".
261    
262    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
263        there was a conditional group that depended on an assertion, if the
264        assertion was false, the callout that immediately followed the alternation
265        in the condition was skipped when pcre_exec() was used for matching.
266    
267    23. Allow an explicit callout to be inserted before an assertion that is the
268        condition for a conditional group, for compatibility with automatic
269        callouts, which always insert a callout at this point.
270    
271    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
272        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
273    
274    25. (*PRUNE) is now supported by the JIT compiler.
275    
276    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
277    
278    27. Fix the case where there are two or more SKIPs with arguments that may be
279        ignored.
280    
281    28. (*SKIP) is now supported by the JIT compiler.
282    
283    29. (*THEN) is now supported by the JIT compiler.
284    
285    30. Update RunTest with additional test selector options.
286    
287    31. The way PCRE handles backtracking verbs has been changed in two ways.
288    
289        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
290        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
291        backtracking. In some cases this makes it more Perl-compatible, but Perl's
292        rather obscure rules do not always do the same thing.
293    
294        (2) Previously, backtracking verbs were confined within assertions. This is
295        no longer the case for positive assertions, except for (*ACCEPT). Again,
296        this sometimes improves Perl compatibility, and sometimes does not.
297    
298    32. A number of tests that were in test 2 because Perl did things differently
299        have been moved to test 1, because either Perl or PCRE has changed, and
300        these tests are now compatible.
301    
302    32. Backtracking control verbs are now handled in the same way in JIT and
303        interpreter.
304    
305    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
306        contained a forward subroutine reference caused a compile error.
307    
308    34. Auto-detect and optimize limited repetitions in JIT.
309    
310    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
311        blocking (*UTF) etc.
312    
313    36. In the interpreter, maximizing pattern repetitions for characters and
314        character types now use tail recursion, which reduces stack usage.
315    
316    37. The value of the max lookbehind was not correctly preserved if a compiled
317        and saved regex was reloaded on a host of different endianness.
318    
319    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
320        of the compiled pattern block, expand the flags field from 16 to 32 bits
321        because it was almost full.
322    
323    39. Try madvise first before posix_madvise.
324    
325    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
326        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
327        original change locked out all matches of empty strings. This has been
328        changed so that one match of an empty string per line is recognized.
329        Subsequent searches on the same line (for colouring or for --only-matching,
330        for example) do not recognize empty strings.
331    
332    41. Applied a user patch to fix a number of spelling mistakes in comments.
333    
334    42. Data lines longer than 65536 caused pcretest to crash.
335    
336    43. Clarified the data type for length and startoffset arguments for pcre_exec
337        and pcre_dfa_exec in the function-specific man pages, where they were
338        explicitly stated to be in bytes, never having been updated. I also added
339        some clarification to the pcreapi man page.
340    
341    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
342        a segmentation fault.
343    
344    
345    Version 8.32 30-November-2012
346    -----------------------------
347    
348    1.  Improved JIT compiler optimizations for first character search and single
349        character iterators.
350    
351    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
352        Patch by Daniel Richard G.
353    
354    3.  Single character iterator optimizations in the JIT compiler.
355    
356    4.  Improved JIT compiler optimizations for character ranges.
357    
358    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
359        Reported by Giuseppe D'Angelo.
360    
361    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
362        of a line, was being set incorrectly in cases where .* appeared inside
363        atomic brackets at the start of a pattern, or where there was a subsequent
364        *PRUNE or *SKIP.
365    
366    7.  Improved instruction cache flush for POWER/PowerPC.
367        Patch by Daniel Richard G.
368    
369    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
370        grep:
371    
372        (a) There is now no limit to the number of patterns to be matched.
373    
374        (b) An error is given if a pattern is too long.
375    
376        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
377            are now supported.
378    
379        (d) --exclude-from and --include-from (multiple use) have been added.
380    
381        (e) Exclusions and inclusions now apply to all files and directories, not
382            just to those obtained from scanning a directory recursively.
383    
384        (f) Multiple uses of -f and --file-list are now supported.
385    
386        (g) In a Windows environment, the default for -d has been changed from
387            "read" (the GNU grep default) to "skip", because otherwise the presence
388            of a directory in the file list provokes an error.
389    
390        (h) The documentation has been revised and clarified in places.
391    
392    9.  Improve the matching speed of capturing brackets.
393    
394    10. Changed the meaning of \X so that it now matches a Unicode extended
395        grapheme cluster.
396    
397    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
398        out POSIX threads when JIT support is configured.
399    
400    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
401    
402    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
403        structure could go wrong in environments where size_t is not the same size
404        as int.
405    
406    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
407        set.
408    
409    15. The EBCDIC support had decayed; later updates to the code had included
410        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
411        general tidy up of EBCDIC-related issues, and the documentation was also
412        not quite right. There is now a test that can be run on ASCII systems to
413        check some of the EBCDIC-related things (but is it not a full test).
414    
415    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
416        in a small tidy to the code.
417    
418    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
419    
420    18. If the --only-matching (-o) option in pcregrep is specified multiple
421        times, each one causes appropriate output. For example, -o1 -o2 outputs the
422        substrings matched by the 1st and 2nd capturing parentheses. A separating
423        string can be specified by --om-separator (default empty).
424    
425    19. Improving the first n character searches.
426    
427    20. Turn case lists for horizontal and vertical white space into macros so that
428        they are defined only once.
429    
430    21. This set of changes together give more compatible Unicode case-folding
431        behaviour for characters that have more than one other case when UCP
432        support is available.
433    
434        (a) The Unicode property table now has offsets into a new table of sets of
435            three or more characters that are case-equivalent. The MultiStage2.py
436            script that generates these tables (the pcre_ucd.c file) now scans
437            CaseFolding.txt instead of UnicodeData.txt for character case
438            information.
439    
440        (b) The code for adding characters or ranges of characters to a character
441            class has been abstracted into a generalized function that also handles
442            case-independence. In UTF-mode with UCP support, this uses the new data
443            to handle characters with more than one other case.
444    
445        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
446            whose other case is greater than 256 are now correctly matched
447            caselessly. Previously, the high codepoint matched the low one, but not
448            vice versa.
449    
450        (d) The processing of \h, \H, \v, and \ in character classes now makes use
451            of the new class addition function, using character lists defined as
452            macros alongside the case definitions of 20 above.
453    
454        (e) Caseless back references now work with characters that have more than
455            one other case.
456    
457        (f) General caseless matching of characters with more than one other case
458            is supported.
459    
460    22. Unicode character properties were updated from Unicode 6.2.0
461    
462    23. Improved CMake support under Windows. Patch by Daniel Richard G.
463    
464    24. Add support for 32-bit character strings, and UTF-32
465    
466    25. Major JIT compiler update (code refactoring and bugfixing).
467        Experimental Sparc 32 support is added.
468    
469    26. Applied a modified version of Daniel Richard G's patch to create
470        pcre.h.generic and config.h.generic by "make" instead of in the
471        PrepareRelease script.
472    
473    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
474        pcre_compile.c when checking for a zero character.
475    
476    28. Introducing a native interface for JIT. Through this interface, the compiled
477        machine code can be directly executed. The purpose of this interface is to
478        provide fast pattern matching, so several sanity checks are not performed.
479        However, feature tests are still performed. The new interface provides
480        1.4x speedup compared to the old one.
481    
482    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
483        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
484        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
485    
486    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
487        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
488        the "old" RFC 2279). Instead, it ended up passing a negative length to
489        pcre_exec().
490    
491    31. Add support for GCC's visibility feature to hide internal functions.
492    
493    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
494        "unknown -C option" after outputting 0 or 1.
495    
496    33. There is now support for generating a code coverage report for the test
497        suite in environments where gcc is the compiler and lcov is installed. This
498        is mainly for the benefit of the developers.
499    
500    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
501        unaddressable using valgrind annotations, allowing valgrind to detect
502        invalid memory accesses. This is mainly for the benefit of the developers.
503    
504    25. (*UTF) can now be used to start a pattern in any of the three libraries.
505    
506    26. Give configure error if --enable-cpp but no C++ compiler found.
507    
508    
509    Version 8.31 06-July-2012
510    -------------------------
511    
512    1.  Fixing a wrong JIT test case and some compiler warnings.
513    
514    2.  Removed a bashism from the RunTest script.
515    
516    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
517        to unsigned type, result still unsigned" that was given by an MS compiler
518        on encountering the code "-sizeof(xxx)".
519    
520    4.  Partial matching support is added to the JIT compiler.
521    
522    5.  Fixed several bugs concerned with partial matching of items that consist
523        of more than one character:
524    
525        (a) /^(..)\1/ did not partially match "aba" because checking references was
526            done on an "all or nothing" basis. This also applied to repeated
527            references.
528    
529        (b) \R did not give a hard partial match if \r was found at the end of the
530            subject.
531    
532        (c) \X did not give a hard partial match after matching one or more
533            characters at the end of the subject.
534    
535        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
536            a partial match for the string "\r".
537    
538        (e) When newline was set to CRLF, the metacharacter "." did not recognize
539            a partial match for a CR character at the end of the subject string.
540    
541    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
542        running pcretest, the text "(JIT)" added to the output whenever JIT is
543        actually used to run the match.
544    
545    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
546        or /S+[+] with a digit between 1 and 7.
547    
548    8.  OP_NOT now supports any UTF character not just single-byte ones.
549    
550    9.  (*MARK) control verb is now supported by the JIT compiler.
551    
552    10. The command "./RunTest list" lists the available tests without actually
553        running any of them. (Because I keep forgetting what they all are.)
554    
555    11. Add PCRE_INFO_MAXLOOKBEHIND.
556    
557    12. Applied a (slightly modified) user-supplied patch that improves performance
558        when the heap is used for recursion (compiled with --disable-stack-for-
559        recursion). Instead of malloc and free for each heap frame each time a
560        logical recursion happens, frames are retained on a chain and re-used where
561        possible. This sometimes gives as much as 30% improvement.
562    
563    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
564        call.
565    
566    14. As documented, (*COMMIT) is now confined to within a positive assertion.
567    
568    15. It is now possible to link pcretest with libedit as an alternative to
569        libreadline.
570    
571    16. (*COMMIT) control verb is now supported by the JIT compiler.
572    
573    17. The Unicode data tables have been updated to Unicode 6.1.0.
574    
575    18. Added --file-list option to pcregrep.
576    
577    19. Added binary file support to pcregrep, including the -a, --binary-files,
578        -I, and --text options.
579    
580    20. The madvise function is renamed for posix_madvise for QNX compatibility
581        reasons. Fixed by Giuseppe D'Angelo.
582    
583    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
584    
585    22. Changed the option for creating symbolic links for 16-bit man pages from
586        -s to -sf so that re-installing does not cause issues.
587    
588    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
589    
590    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
591        matches in certain environments (the workspace was not being correctly
592        retained). Also added to pcre_dfa_exec() a simple plausibility check on
593        some of the workspace data at the beginning of a restart.
594    
595    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
596        was not doing so when it should - probably a typo introduced by SVN 528
597        (change 8.10/14).
598    
599    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
600        \w+ when the character tables indicated that \x{c4} was a word character.
601        There were several related cases, all because the tests for doing a table
602        lookup were testing for characters less than 127 instead of 255.
603    
604    27. If a pattern contains capturing parentheses that are not used in a match,
605        their slots in the ovector are set to -1. For those that are higher than
606        any matched groups, this happens at the end of processing. In the case when
607        there were back references that the ovector was too small to contain
608        (causing temporary malloc'd memory to be used during matching), and the
609        highest capturing number was not used, memory off the end of the ovector
610        was incorrectly being set to -1. (It was using the size of the temporary
611        memory instead of the true size.)
612    
613    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
614        ovector size, it uses memory at the end of the block that it has got.
615    
616    29. Check for an overlong MARK name and give an error at compile time. The
617        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
618    
619    30. JIT compiler update.
620    
621    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
622        Rill for the patch.
623    
624    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
625    
626    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
627    
628    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
629        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
630        was enabled.
631    
632    35. Improve JIT code generation for greedy plus quantifier.
633    
634    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
635        1 to "aa" instead of to an empty string. The bug affected repeated groups
636        that could potentially match an empty string.
637    
638    37. Optimizing single character iterators in JIT.
639    
640    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
641        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
642        codepoints that are too big for the mode are faulted, and in a UTF mode,
643        disallowed codepoints are also faulted.
644    
645    39. If PCRE was compiled with UTF support, in three places in the DFA
646        matcher there was code that should only have been obeyed in UTF mode, but
647        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
648        processing when bytes with values greater than 127 were present. In 16-bit
649        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
650        both cases the values are those that cannot be the first data item in a UTF
651        character. The three items that might have provoked this were recursions,
652        possessively repeated groups, and atomic groups.
653    
654    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
655        and pcregrep, because some OS require shared objects to be explicitly
656        passed to ld, causing the link step to fail if they are not.
657    
658    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
659        mode, patterns that started with \h* or \R* might be incorrectly matched.
660    
661    
662    Version 8.30 04-February-2012
663    -----------------------------
664    
665    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
666        name is defined in ctype.h.
667    
668    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
669        only in quite long subpatterns.
670    
671    3.  Removed the function pcre_info(), which has been obsolete and deprecated
672        since it was replaced by pcre_fullinfo() in February 2000.
673    
674    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
675        match a (*MARK), and the match failed at the start of the subject, a
676        reference to memory before the start of the subject could occur. This bug
677        was introduced by fix 17 of release 8.21.
678    
679    5.  A reference to an unset group with zero minimum repetition was giving
680        totally wrong answers (in non-JavaScript-compatibility mode). For example,
681        /(another)?(\1?)test/ matched against "hello world test". This bug was
682        introduced in release 8.13.
683    
684    6.  Add support for 16-bit character strings (a large amount of work involving
685        many changes and refactorings).
686    
687    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
688        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
689        from a file.
690    
691    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
692        rounding is not applied in this particular case).
693    
694    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
695        if they appear, or are escaped, in patterns.
696    
697    10. Get rid of a number of -Wunused-but-set-variable warnings.
698    
699    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
700        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
701        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
702        also returns the mark "x". This bug applied to capturing parentheses,
703        non-capturing parentheses, and atomic parentheses. It also applied to some
704        assertions.
705    
706    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
707        information out of configure.ac instead of relying on pcre.h.generic, which
708        is not stored in the repository.
709    
710    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
711        -lreadline.
712    
713    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
714    
715    15. Applied Graycode's patch to put the top-level frame on the stack rather
716        than the heap when not using the stack for recursion. This gives a
717        performance improvement in many cases when recursion is not deep.
718    
719    16. Experimental code added to "pcretest -C" to output the stack frame size.
720    
721    
722    Version 8.21 12-Dec-2011
723    ------------------------
724    
725    1.  Updating the JIT compiler.
726    
727    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
728        are added as well.
729    
730    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
731        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
732        calling _pcre_jit_exec. Some extra comments are added.
733    
734    4.  (*MARK) settings inside atomic groups that do not contain any capturing
735        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
736        was introduced by change 18 for 8.20.
737    
738    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
739        ECMA-262 standard.
740    
741    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
742        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
743        This bug was probably introduced by change 9 of 8.13.
744    
745    7.  While fixing 6 above, I noticed that a number of other items were being
746        incorrectly rejected as "not fixed length". This arose partly because newer
747        opcodes had not been added to the fixed-length checking code. I have (a)
748        corrected the bug and added tests for these items, and (b) arranged for an
749        error to occur if an unknown opcode is encountered while checking for fixed
750        length instead of just assuming "not fixed length". The items that were
751        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
752        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
753        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
754    
755    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
756        being incorrectly compiled and would have given unpredicatble results.
757    
758    9.  A possessively repeated subpattern with minimum repeat count greater than
759        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
760        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
761        the first (A) could occur when it should not.
762    
763    10. Add a cast and remove a redundant test from the code.
764    
765    11. JIT should use pcre_malloc/pcre_free for allocation.
766    
767    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
768        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
769        is anything other than /usr, -L is still shown).
770    
771    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
772    
773    14. Perl does not support \N without a following name in a [] class; PCRE now
774        also gives an error.
775    
776    15. If a forward reference was repeated with an upper limit of around 2000,
777        it caused the error "internal error: overran compiling workspace". The
778        maximum number of forward references (including repeats) was limited by the
779        internal workspace, and dependent on the LINK_SIZE. The code has been
780        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
781        the default depends on LINK_SIZE. There is a new upper limit (for safety)
782        of around 200,000 forward references. While doing this, I also speeded up
783        the filling in of repeated forward references.
784    
785    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
786        incorrectly expecting the subject to contain another "a" after the start.
787    
788    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
789        in the match, the SKIP should be ignored. This was not happening; instead
790        the SKIP was being treated as NOMATCH. For patterns such as
791        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
792        tested.
793    
794    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
795        now much more compatible with Perl, in particular in cases where the result
796        is a non-match for a non-anchored pattern. For example, if
797        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
798        "m", where previously it did not return a name. A side effect of this
799        change is that for partial matches, the last encountered mark name is
800        returned, as for non matches. A number of tests that were previously not
801        Perl-compatible have been moved into the Perl-compatible test files. The
802        refactoring has had the pleasing side effect of removing one argument from
803        the match() function, thus reducing its stack requirements.
804    
805    19. If the /S+ option was used in pcretest to study a pattern using JIT,
806        subsequent uses of /S (without +) incorrectly behaved like /S+.
807    
808    21. Retrieve executable code size support for the JIT compiler and fixing
809        some warnings.
810    
811    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
812        not work when the shorter character appeared right at the end of the
813        subject string.
814    
815    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
816        systems.
817    
818    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
819        output it when the /M option is used in pcretest.
820    
821    25. The CheckMan script was not being included in the distribution. Also, added
822        an explicit "perl" to run Perl scripts from the PrepareRelease script
823        because this is reportedly needed in Windows.
824    
825    26. If study data was being save in a file and studying had not found a set of
826        "starts with" bytes for the pattern, the data written to the file (though
827        never used) was taken from uninitialized memory and so caused valgrind to
828        complain.
829    
830    27. Updated RunTest.bat as provided by Sheri Pierce.
831    
832    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
833    
834    29. Computation of memory usage for the table of capturing group names was
835        giving an unnecessarily large value.
836    
837    
838    Version 8.20 21-Oct-2011
839    ------------------------
840    
841    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
842        a POSIX class. After further experiments with Perl, which convinced me that
843        Perl has bugs and confusions, a closing square bracket is no longer allowed
844        in a POSIX name. This bug also affected patterns with classes that started
845        with full stops.
846    
847    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
848        captured substring, but while checking the failing first alternative,
849        substring 1 is temporarily captured. If the output vector supplied to
850        pcre_exec() was not big enough for this capture, the yield of the function
851        was still zero ("insufficient space for captured substrings"). This cannot
852        be totally fixed without adding another stack variable, which seems a lot
853        of expense for a edge case. However, I have improved the situation in cases
854        such as /(a)(b)x|abc/ matched against "abc", where the return code
855        indicates that fewer than the maximum number of slots in the ovector have
856        been set.
857    
858    3.  Related to (2) above: when there are more back references in a pattern than
859        slots in the output vector, pcre_exec() uses temporary memory during
860        matching, and copies in the captures as far as possible afterwards. It was
861        using the entire output vector, but this conflicts with the specification
862        that only 2/3 is used for passing back captured substrings. Now it uses
863        only the first 2/3, for compatibility. This is, of course, another edge
864        case.
865    
866    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
867        main code base, and can be used by building with --enable-jit. When this is
868        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
869        runtime --no-jit option is given.
870    
871    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
872        ovector, the return from the function was zero, implying that there were
873        other matches that did not fit. The correct "exactly full" value is now
874        returned.
875    
876    6.  If a subpattern that was called recursively or as a subroutine contained
877        (*PRUNE) or any other control that caused it to give a non-standard return,
878        invalid errors such as "Error -26 (nested recursion at the same subject
879        position)" or even infinite loops could occur.
880    
881    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
882        computing the minimum length on reaching *ACCEPT, and so ended up with the
883        wrong value of 1 rather than 0. Further investigation indicates that
884        computing a minimum subject length in the presence of *ACCEPT is difficult
885        (think back references, subroutine calls), and so I have changed the code
886        so that no minimum is registered for a pattern that contains *ACCEPT.
887    
888    8.  If (*THEN) was present in the first (true) branch of a conditional group,
889        it was not handled as intended. [But see 16 below.]
890    
891    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
892        Sheri Pierce.
893    
894    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
895        the first byte in a match must be "a".
896    
897    11. Change 17 for 8.13 increased the recursion depth for patterns like
898        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
899        pattern contains any instances of (*THEN). If it does not, the old
900        optimizations are restored. It would be nice to do this on a per-group
901        basis, but at the moment that is not feasible.
902    
903    12. In some environments, the output of pcretest -C is CRLF terminated. This
904        broke RunTest's code that checks for the link size. A single white space
905        character after the value is now allowed for.
906    
907    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
908        For "fr", it uses the Windows-specific input and output files.
909    
910    14. If (*THEN) appeared in a group that was called recursively or as a
911        subroutine, it did not work as intended. [But see next item.]
912    
913    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
914        pattern fragments (but not containing any | characters). If A and B are
915        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
916        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
917        D. In other words, Perl considers parentheses that do not contain any |
918        characters to be part of a surrounding alternative, whereas PCRE was
919        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
920        differently. PCRE now behaves in the same way as Perl, except in the case
921        of subroutine/recursion calls such as (?1) which have in any case always
922        been different (but PCRE had them first :-).
923    
924    16. Related to 15 above: Perl does not treat the | in a conditional group as
925        creating alternatives. Such a group is treated in the same way as an
926        ordinary group without any | characters when processing (*THEN). PCRE has
927        been changed to match Perl's behaviour.
928    
929    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
930        RunGrepTest script failed.
931    
932    18. Change 22 for version 13 caused atomic groups to use more stack. This is
933        inevitable for groups that contain captures, but it can lead to a lot of
934        stack use in large patterns. The old behaviour has been restored for atomic
935        groups that do not contain any capturing parentheses.
936    
937    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
938        suppress the check for a minimum subject length at run time. (If it was
939        given to pcre_exec() or pcre_dfa_exec() it did work.)
940    
941    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
942        fail to work when decoding hex characters in data strings in EBCDIC
943        environments.
944    
945    21. It appears that in at least one Mac OS environment, the isxdigit() function
946        is implemented as a macro that evaluates to its argument more than once,
947        contravening the C 90 Standard (I haven't checked a later standard). There
948        was an instance in pcretest which caused it to go wrong when processing
949        \x{...} escapes in subject strings. The has been rewritten to avoid using
950        things like p++ in the argument of isxdigit().
951    
952    
953    Version 8.13 16-Aug-2011
954    ------------------------
955    
956    1.  The Unicode data tables have been updated to Unicode 6.0.0.
957    
958    2.  Two minor typos in pcre_internal.h have been fixed.
959    
960    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
961        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
962        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
963    
964    4.  There were a number of related bugs in the code for matching backrefences
965        caselessly in UTF-8 mode when codes for the characters concerned were
966        different numbers of bytes. For example, U+023A and U+2C65 are an upper
967        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
968        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
969        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
970        2-byte code at the end of the subject (it thought there wasn't enough data
971        left).
972    
973    5.  Comprehensive information about what went wrong is now returned by
974        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
975        as the output vector has at least 2 elements. The offset of the start of
976        the failing character and a reason code are placed in the vector.
977    
978    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
979        now returned is for the first byte of the failing character, instead of the
980        last byte inspected. This is an incompatible change, but I hope it is small
981        enough not to be a problem. It makes the returned offset consistent with
982        pcre_exec() and pcre_dfa_exec().
983    
984    7.  pcretest now gives a text phrase as well as the error number when
985        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
986        failure, the offset and reason code are output.
987    
988    8.  When \R was used with a maximizing quantifier it failed to skip backwards
989        over a \r\n pair if the subsequent match failed. Instead, it just skipped
990        back over a single character (\n). This seems wrong (because it treated the
991        two characters as a single entity when going forwards), conflicts with the
992        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
993        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
994        has been changed.
995    
996    9.  Some internal refactoring has changed the processing so that the handling
997        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
998        time (the PCRE_DOTALL option was changed this way some time ago: version
999        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
1000        which was always a bit of a fudge. It also means that there is one less
1001        argument for the match() function, which reduces its stack requirements
1002        slightly. This change also fixes an incompatibility with Perl: the pattern
1003        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
1004    
1005    10. More internal refactoring has drastically reduced the number of recursive
1006        calls to match() for possessively repeated groups such as (abc)++ when
1007        using pcre_exec().
1008    
1009    11. While implementing 10, a number of bugs in the handling of groups were
1010        discovered and fixed:
1011    
1012        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
1013        (a|)*(?1) gave a compile-time internal error.
1014        ((a|)+)+  did not notice that the outer group could match an empty string.
1015        (^a|^)+   was not marked as anchored.
1016        (.*a|.*)+ was not marked as matching at start or after a newline.
1017    
1018    12. Yet more internal refactoring has removed another argument from the match()
1019        function. Special calls to this function are now indicated by setting a
1020        value in a variable in the "match data" data block.
1021    
1022    13. Be more explicit in pcre_study() instead of relying on "default" for
1023        opcodes that mean there is no starting character; this means that when new
1024        ones are added and accidentally left out of pcre_study(), testing should
1025        pick them up.
1026    
1027    14. The -s option of pcretest has been documented for ages as being an old
1028        synonym of -m (show memory usage). I have changed it to mean "force study
1029        for every regex", that is, assume /S for every regex. This is similar to -i
1030        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
1031        using it. It makes it easier to run collections of tests with and without
1032        study enabled, and thereby test pcre_study() more easily. All the standard
1033        tests are now run with and without -s (but some patterns can be marked as
1034        "never study" - see 20 below).
1035    
1036    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
1037        restoration of the capturing data to the outer values was not happening
1038        correctly.
1039    
1040    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
1041        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
1042        pattern had matched an empty string, and so incorrectly returned a no
1043        match.
1044    
1045    17. There was optimizing code for the last branch of non-capturing parentheses,
1046        and also for the obeyed branch of a conditional subexpression, which used
1047        tail recursion to cut down on stack usage. Unfortunately, now that there is
1048        the possibility of (*THEN) occurring in these branches, tail recursion is
1049        no longer possible because the return has to be checked for (*THEN). These
1050        two optimizations have therefore been removed. [But see 8.20/11 above.]
1051    
1052    18. If a pattern containing \R was studied, it was assumed that \R always
1053        matched two bytes, thus causing the minimum subject length to be
1054        incorrectly computed because \R can also match just one byte.
1055    
1056    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
1057        was incorrectly computed.
1058    
1059    20. If /S is present twice on a test pattern in pcretest input, it now
1060        *disables* studying, thereby overriding the use of -s on the command line
1061        (see 14 above). This is necessary for one or two tests to keep the output
1062        identical in both cases.
1063    
1064    21. When (*ACCEPT) was used in an assertion that matched an empty string and
1065        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
1066    
1067    22. When an atomic group that contained a capturing parenthesis was
1068        successfully matched, but the branch in which it appeared failed, the
1069        capturing was not being forgotten if a higher numbered group was later
1070        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
1071        group 1 to "a", when in fact it should be unset. This applied to multi-
1072        branched capturing and non-capturing groups, repeated or not, and also to
1073        positive assertions (capturing in negative assertions does not happen
1074        in PCRE) and also to nested atomic groups.
1075    
1076    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
1077        subject after a captured substring, to make it easier to tell which of a
1078        number of identical substrings has been captured.
1079    
1080    24. The way atomic groups are processed by pcre_exec() has been changed so that
1081        if they are repeated, backtracking one repetition now resets captured
1082        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
1083        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
1084        "aaa". Previously, it would have been "a". As part of this code
1085        refactoring, the way recursive calls are handled has also been changed.
1086    
1087    25. If an assertion condition captured any substrings, they were not passed
1088        back unless some other capturing happened later. For example, if
1089        (?(?=(a))a) was matched against "a", no capturing was returned.
1090    
1091    26. When studying a pattern that contained subroutine calls or assertions,
1092        the code for finding the minimum length of a possible match was handling
1093        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
1094        group 1 called group 2 while simultaneously a separate group 2 called group
1095        1). A stack overflow occurred in this case. I have fixed this by limiting
1096        the recursion depth to 10.
1097    
1098    27. Updated RunTest.bat in the distribution to the version supplied by Tom
1099        Fortmann. This supports explicit test numbers on the command line, and has
1100        argument validation and error reporting.
1101    
1102    28. An instance of \X with an unlimited repeat could fail if at any point the
1103        first character it looked at was a mark character.
1104    
1105    29. Some minor code refactoring concerning Unicode properties and scripts
1106        should reduce the stack requirement of match() slightly.
1107    
1108    30. Added the '=' option to pcretest to check the setting of unused capturing
1109        slots at the end of the pattern, which are documented as being -1, but are
1110        not included in the return count.
1111    
1112    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
1113        compiled something random. Now it gives a compile-time error (as does
1114        Perl).
1115    
1116    32. A *MARK encountered during the processing of a positive assertion is now
1117        recorded and passed back (compatible with Perl).
1118    
1119    33. If --only-matching or --colour was set on a pcregrep call whose pattern
1120        had alternative anchored branches, the search for a second match in a line
1121        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
1122        matched the line "0102" twice. The same bug affected patterns that started
1123        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
1124        twice.
1125    
1126    34. Previously, PCRE did not allow quantification of assertions. However, Perl
1127        does, and because of capturing effects, quantifying parenthesized
1128        assertions may at times be useful. Quantifiers are now allowed for
1129        parenthesized assertions.
1130    
1131    35. A minor code tidy in pcre_compile() when checking options for \R usage.
1132    
1133    36. \g was being checked for fancy things in a character class, when it should
1134        just be a literal "g".
1135    
1136    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
1137        appearance of a nested POSIX class supersedes an apparent external class.
1138        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
1139        unescaped square brackets may also appear as part of class names. For
1140        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
1141        more like Perl. (But see 8.20/1 above.)
1142    
1143    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
1144        was because it thought it was \N{name}, which is not supported).
1145    
1146    39. Add minix to OS list not supporting the -S option in pcretest.
1147    
1148    40. PCRE tries to detect cases of infinite recursion at compile time, but it
1149        cannot analyze patterns in sufficient detail to catch mutual recursions
1150        such as ((?1))((?2)). There is now a runtime test that gives an error if a
1151        subgroup is called recursively as a subpattern for a second time at the
1152        same position in the subject string. In previous releases this might have
1153        been caught by the recursion limit, or it might have run out of stack.
1154    
1155    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
1156        happen only once. PCRE was, however incorrectly giving a compile time error
1157        "recursive call could loop indefinitely" because it cannot analyze the
1158        pattern in sufficient detail. The compile time test no longer happens when
1159        PCRE is compiling a conditional subpattern, but actual runaway loops are
1160        now caught at runtime (see 40 above).
1161    
1162    42. It seems that Perl allows any characters other than a closing parenthesis
1163        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
1164        has been changed to be the same.
1165    
1166    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
1167        as not to get warnings when autogen.sh is called. Also changed
1168        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
1169    
1170    44. To help people who use pcregrep to scan files containing exceedingly long
1171        lines, the following changes have been made:
1172    
1173        (a) The default value of the buffer size parameter has been increased from
1174            8K to 20K. (The actual buffer used is three times this size.)
1175    
1176        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
1177            PCRE is built.
1178    
1179        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
1180            to be set at run time.
1181    
1182        (d) Numerical values in pcregrep options can be followed by K or M, for
1183            example --buffer-size=50K.
1184    
1185        (e) If a line being scanned overflows pcregrep's buffer, an error is now
1186            given and the return code is set to 2.
1187    
1188    45. Add a pointer to the latest mark to the callout data block.
1189    
1190    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
1191        partial match of an empty string instead of no match. This was specific to
1192        the use of ".".
1193    
1194    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
1195        complete match instead of a partial match. This bug was dependent on both
1196        the PCRE_UTF8 and PCRE_DOTALL options being set.
1197    
1198    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1199        starting byte set, because \b was not being ignored.
1200    
1201    
1202    Version 8.12 15-Jan-2011
1203    ------------------------
1204    
1205    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1206        checks for such things as part of the documentation building process.
1207    
1208    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1209        --match-limit and --recursion-limit options (added for 8.11). In
1210        particular, this made one of the standard tests fail. (The integer value
1211        went into the wrong half of a long int.)
1212    
1213    3.  If the --colour option was given to pcregrep with -v (invert match), it
1214        did strange things, either producing crazy output, or crashing. It should,
1215        of course, ignore a request for colour when reporting lines that do not
1216        match.
1217    
1218    4.  Another pcregrep bug caused similar problems if --colour was specified with
1219        -M (multiline) and the pattern match finished with a line ending.
1220    
1221    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1222        matched in multiline mode, the following line was shown as part of the
1223        match. This seems wrong, so I have changed it.
1224    
1225    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1226        the check for further matches in the same line (so they could be coloured)
1227        to overrun the end of the current line. If another match was found, it was
1228        incorrectly shown (and then shown again when found in the next line).
1229    
1230    7.  If pcregrep was compiled under Windows, there was a reference to the
1231        function pcregrep_exit() before it was defined. I am assuming this was
1232        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1233        reported by a user. I've moved the definition above the reference.
1234    
1235    
1236    Version 8.11 10-Dec-2010
1237    ------------------------
1238    
1239    1.  (*THEN) was not working properly if there were untried alternatives prior
1240        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1241        backtracked to try for "b" instead of moving to the next alternative branch
1242        at the same level (in this case, to look for "c"). The Perl documentation
1243        is clear that when (*THEN) is backtracked onto, it goes to the "next
1244        alternative in the innermost enclosing group".
1245    
1246    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1247        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1248        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1249        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1250        (*THEN).
1251    
1252    3.  If \s appeared in a character class, it removed the VT character from
1253        the class, even if it had been included by some previous item, for example
1254        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1255        of \s, but is part of the POSIX "space" class.)
1256    
1257    4.  A partial match never returns an empty string (because you can always
1258        match an empty string at the end of the subject); however the checking for
1259        an empty string was starting at the "start of match" point. This has been
1260        changed to the "earliest inspected character" point, because the returned
1261        data for a partial match starts at this character. This means that, for
1262        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1263        (previously it gave "no match").
1264    
1265    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1266        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1267        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1268        has an implication that the given string is incomplete (because a partial
1269        match is preferred over a full match). For this reason, these items now
1270        give a partial match in this situation. [Aside: previously, the one case
1271        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1272        match rather than a full match, which was wrong by the old rules, but is
1273        now correct.]
1274    
1275    6.  There was a bug in the handling of #-introduced comments, recognized when
1276        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1277        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1278        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1279        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1280        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1281        places in pcre_compile().
1282    
1283    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1284        comments when looking ahead for named forward references to subpatterns,
1285        the only newline sequence it recognized was NL. It now handles newlines
1286        according to the set newline convention.
1287    
1288    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1289        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1290        cater for a lack of strerror(). These oversights have been fixed.
1291    
1292    9.  Added --match-limit and --recursion-limit to pcregrep.
1293    
1294    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1295    
1296    11. When the -o option was used, pcregrep was setting a return code of 1, even
1297        when matches were found, and --line-buffered was not being honoured.
1298    
1299    12. Added an optional parentheses number to the -o and --only-matching options
1300        of pcregrep.
1301    
1302    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1303        can match an empty string. The code to do it in pcretest and pcredemo
1304        needed fixing:
1305    
1306        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1307            only one byte after an empty string match just before CRLF (this case
1308            just got forgotten; "any" and "anycrlf" were OK).
1309    
1310        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1311            mode when an empty string match preceded an ASCII character followed by
1312            a non-ASCII character. (The code for advancing by one character rather
1313            than one byte was nonsense.)
1314    
1315        (c) The pcredemo.c sample program did not have any code at all to handle
1316            the cases when CRLF is a valid newline sequence.
1317    
1318    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1319        as a starting offset was within the subject string. There is now a new
1320        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1321        negative or greater than the length of the string. In order to test this,
1322        pcretest is extended to allow the setting of negative starting offsets.
1323    
1324    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1325        starting offset points to the beginning of a UTF-8 character was
1326        unnecessarily clumsy. I tidied it up.
1327    
1328    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1329        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1330    
1331    17. Nobody had reported that the --include_dir option, which was added in
1332        release 7.7 should have been called --include-dir (hyphen, not underscore)
1333        for compatibility with GNU grep. I have changed it to --include-dir, but
1334        left --include_dir as an undocumented synonym, and the same for
1335        --exclude-dir, though that is not available in GNU grep, at least as of
1336        release 2.5.4.
1337    
1338    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1339        characters from a string of bytes) have been redefined so as not to use
1340        loops, in order to improve performance in some environments. At the same
1341        time, I abstracted some of the common code into auxiliary macros to save
1342        repetition (this should not affect the compiled code).
1343    
1344    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1345        compile-time error is now given if \c is not followed by an ASCII
1346        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1347        different, and any byte value is allowed.)
1348    
1349    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1350        START_OPTIMIZE option, which is now allowed at compile time - but just
1351        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1352        to pcregrep and other applications that have no direct access to PCRE
1353        options. The new /Y option in pcretest sets this option when calling
1354        pcre_compile().
1355    
1356    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1357        back references. Groups containing recursive back references were forced to
1358        be atomic by that change, but in the case of named groups, the amount of
1359        memory required was incorrectly computed, leading to "Failed: internal
1360        error: code overflow". This has been fixed.
1361    
1362    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1363        pcretest.c, to avoid build problems in some Borland environments.
1364    
1365    
1366    Version 8.10 25-Jun-2010
1367    ------------------------
1368    
1369    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1370        THEN.
1371    
1372    2.  (*ACCEPT) was not working when inside an atomic group.
1373    
1374    3.  Inside a character class, \B is treated as a literal by default, but
1375        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1376        causes the error). The code is unchanged, but I tidied the documentation.
1377    
1378    4.  Inside a character class, PCRE always treated \R and \X as literals,
1379        whereas Perl faults them if its -w option is set. I have changed PCRE so
1380        that it faults them when PCRE_EXTRA is set.
1381    
1382    5.  Added support for \N, which always matches any character other than
1383        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1384    
1385    6.  When compiling pcregrep with newer versions of gcc which may have
1386        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1387        declared with attribute warn_unused_result" were given. Just casting the
1388        result to (void) does not stop the warnings; a more elaborate fudge is
1389        needed. I've used a macro to implement this.
1390    
1391    7.  Minor change to pcretest.c to avoid a compiler warning.
1392    
1393    8.  Added four artifical Unicode properties to help with an option to make
1394        \s etc use properties (see next item). The new properties are: Xan
1395        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1396    
1397    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1398        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1399        this option. Modified pcretest to add /W to test this facility. Added
1400        REG_UCP to make it available via the POSIX interface.
1401    
1402    10. Added --line-buffered to pcregrep.
1403    
1404    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1405        studied, and the match started with a letter with a code point greater than
1406        127 whose first byte was different to the first byte of the other case of
1407        the letter, the other case of this starting letter was not recognized
1408        (#976).
1409    
1410    12. If a pattern that was studied started with a repeated Unicode property
1411        test, for example, \p{Nd}+, there was the theoretical possibility of
1412        setting up an incorrect bitmap of starting bytes, but fortunately it could
1413        not have actually happened in practice until change 8 above was made (it
1414        added property types that matched character-matching opcodes).
1415    
1416    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1417        possible starting bytes for non-anchored patterns.
1418    
1419    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1420        \R, and also a number of cases that involve Unicode properties, both
1421        explicit and implicit when PCRE_UCP is set.
1422    
1423    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1424        input, it could crash or give wrong results if characters with values
1425        greater than 0xc0 were present in the subject string. (Detail: it assumed
1426        UTF-8 input when processing these items.)
1427    
1428    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1429        size_t is 64-bit (#991).
1430    
1431    17. Added a check for running out of memory when PCRE is compiled with
1432        --disable-stack-for-recursion (#990).
1433    
1434    18. If the last data line in a file for pcretest does not have a newline on
1435        the end, a newline was missing in the output.
1436    
1437    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1438        less than 128) in its various bitmaps. However, there is a facility for
1439        generating tables according to the current locale when PCRE is compiled. It
1440        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1441        characters, are recognized by isspace() and therefore were getting set in
1442        these tables, and indeed these tables seem to approximate to ISO 8859. This
1443        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1444        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1445        which of course cannot start UTF-8 characters. I have changed the code so
1446        that only real ASCII characters (less than 128) and the correct starting
1447        bytes for UTF-8 encodings are set for characters greater than 127 when in
1448        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1449        altogether.)
1450    
1451    20. Added the /T option to pcretest so as to be able to run tests with non-
1452        standard character tables, thus making it possible to include the tests
1453        used for 19 above in the standard set of tests.
1454    
1455    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1456        reference to a subpattern the other side of a comment that contains an
1457        opening parenthesis caused either an internal compiling error, or a
1458        reference to the wrong subpattern.
1459    
1460    
1461    Version 8.02 19-Mar-2010
1462    ------------------------
1463    
1464    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1465    
1466    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1467        configured.
1468    
1469    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1470        original author of that file, following a query about its status.
1471    
1472    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1473        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1474    
1475    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1476        quantifier applied to a forward-referencing subroutine call, could compile
1477        incorrect code or give the error "internal error: previously-checked
1478        referenced subpattern not found".
1479    
1480    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1481        variables to point to external functions. For these systems, therefore,
1482        pcre_malloc etc. are now initialized to local functions that call the
1483        relevant global functions.
1484    
1485    7.  There were two entries missing in the vectors called coptable and poptable
1486        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1487        I've fixed the data, and added a kludgy way of testing at compile time that
1488        the lengths are correct (equal to the number of opcodes).
1489    
1490    8.  Following on from 7, I added a similar kludge to check the length of the
1491        eint vector in pcreposix.c.
1492    
1493    9.  Error texts for pcre_compile() are held as one long string to avoid too
1494        much relocation at load time. To find a text, the string is searched,
1495        counting zeros. There was no check for running off the end of the string,
1496        which could happen if a new error number was added without updating the
1497        string.
1498    
1499    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1500    
1501    11. \K was not working if it appeared in an atomic group or in a group that
1502        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1503        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1504        the assertion is positive, but not if it is negative.
1505    
1506    12. Change 11 fortuitously reduced the size of the stack frame used in the
1507        "match()" function of pcre_exec.c by one pointer. Forthcoming
1508        implementation of support for (*MARK) will need an extra pointer on the
1509        stack; I have reserved it now, so that the stack frame size does not
1510        decrease.
1511    
1512    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1513        item in branch that calls a recursion is a subroutine call - as in the
1514        second branch in the above example - was incorrectly given the compile-
1515        time error "recursive call could loop indefinitely" because pcre_compile()
1516        was not correctly checking the subroutine for matching a non-empty string.
1517    
1518    14. The checks for overrunning compiling workspace could trigger after an
1519        overrun had occurred. This is a "should never occur" error, but it can be
1520        triggered by pathological patterns such as hundreds of nested parentheses.
1521        The checks now trigger 100 bytes before the end of the workspace.
1522    
1523    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1524    
1525    
1526    Version 8.01 19-Jan-2010
1527    ------------------------
1528    
1529    1.  If a pattern contained a conditional subpattern with only one branch (in
1530        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1531        computed the wrong minimum data length (which is of course zero for such
1532        subpatterns). This could cause incorrect "no match" results.
1533    
1534    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1535        the pattern is reset in the first branch, pcre_compile() failed with
1536        "internal error: code overflow at offset...". This happened only when
1537        the reset was to the original external option setting. (An optimization
1538        abstracts leading options settings into an external setting, which was the
1539        cause of this.)
1540    
1541    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1542        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1543        assertion pattern did not match (meaning that the assertion was true), it
1544        was incorrectly treated as false if the SKIP had been reached during the
1545        matching. This also applied to assertions used as conditions.
1546    
1547    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1548        assertion subpattern, including such a pattern used as a condition,
1549        unpredictable results occurred, instead of the error return
1550        PCRE_ERROR_DFA_UITEM.
1551    
1552    5.  The C++ GlobalReplace function was not working like Perl for the special
1553        situation when an empty string is matched. It now does the fancy magic
1554        stuff that is necessary.
1555    
1556    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1557        removed. (These were left over from very, very early versions of PCRE.)
1558    
1559    7.  Some cosmetic changes to the code to make life easier when compiling it
1560        as part of something else:
1561    
1562        (a) Change DEBUG to PCRE_DEBUG.
1563    
1564        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1565            called "current" as "current_branch", to prevent a collision with the
1566            Linux macro when compiled as a kernel module.
1567    
1568        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1569            prevent a collision with the Linux macro when compiled as a kernel
1570            module.
1571    
1572    8.  In pcre_compile() there are some checks for integer overflows that used to
1573        cast potentially large values to (double). This has been changed to that
1574        when building, a check for int64_t is made, and if it is found, it is used
1575        instead, thus avoiding the use of floating point arithmetic. (There is no
1576        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1577        double.
1578    
1579    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1580        2005 (difference between two addresses compared to an unsigned value).
1581    
1582    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1583        custom one, because of the following reported problem in Windows:
1584    
1585          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1586              under Win32.
1587          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1588              therefore missing the function definition.
1589          - The compiler thus generates a "C" signature for the test function.
1590          - The linker fails to find the "C" function.
1591          - PCRE fails to configure if asked to do so against libbz2.
1592    
1593    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1594        messages were output:
1595    
1596          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1597          rerunning libtoolize, to keep the correct libtool macros in-tree.
1598          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1599    
1600        I have done both of these things.
1601    
1602    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1603        most of the time, it *can* run out if it is given a pattern that contains a
1604        runaway infinite recursion. I updated the discussion in the pcrestack man
1605        page.
1606    
1607    13. Now that we have gone to the x.xx style of version numbers, the minor
1608        version may start with zero. Using 08 or 09 is a bad idea because users
1609        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1610        interpreted as invalid octal numbers. I've updated the previous comment in
1611        configure.ac, and also added a check that gives an error if 08 or 09 are
1612        used.
1613    
1614    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1615        causing partial matching to fail when the end of the subject matched \W
1616        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1617    
1618    15. There were some discrepancies between the declarations in pcre_internal.h
1619        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1620        their definitions. The declarations used "const uschar *" and the
1621        definitions used USPTR. Even though USPTR is normally defined as "const
1622        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1623        reported that: "This difference in casting confuses some C++ compilers, for
1624        example, SunCC recognizes above declarations as different functions and
1625        generates broken code for hbpcre." I have changed the declarations to use
1626        USPTR.
1627    
1628    16. GNU libtool is named differently on some systems. The autogen.sh script now
1629        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1630        (FreeBSD).
1631    
1632    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1633        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1634        comment: "Figure out how to create a longlong from a string: strtoll and
1635        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1636        instance, but it only takes 2 args instead of 3!"
1637    
1638    18. A subtle bug concerned with back references has been fixed by a change of
1639        specification, with a corresponding code fix. A pattern such as
1640        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1641        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1642        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1643        same bug. Such groups have to be quantified to be useful, or contained
1644        inside another quantified group. (If there's no repetition, the reference
1645        can never match.) The problem arises because, having left the group and
1646        moved on to the rest of the pattern, a later failure that backtracks into
1647        the group uses the captured value from the final iteration of the group
1648        rather than the correct earlier one. I have fixed this in PCRE by forcing
1649        any group that contains a reference to itself to be an atomic group; that
1650        is, there cannot be any backtracking into it once it has completed. This is
1651        similar to recursive and subroutine calls.
1652    
1653    
1654    Version 8.00 19-Oct-09
1655    ----------------------
1656    
1657    1.  The table for translating pcre_compile() error codes into POSIX error codes
1658        was out-of-date, and there was no check on the pcre_compile() error code
1659        being within the table. This could lead to an OK return being given in
1660        error.
1661    
1662    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1663        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1664        in a Windows environment.
1665    
1666    3.  The pcregrep --count option prints the count for each file even when it is
1667        zero, as does GNU grep. However, pcregrep was also printing all files when
1668        --files-with-matches was added. Now, when both options are given, it prints
1669        counts only for those files that have at least one match. (GNU grep just
1670        prints the file name in this circumstance, but including the count seems
1671        more useful - otherwise, why use --count?) Also ensured that the
1672        combination -clh just lists non-zero counts, with no names.
1673    
1674    4.  The long form of the pcregrep -F option was incorrectly implemented as
1675        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1676        but it seems right to fix it, and I didn't think it was worth preserving
1677        the old behaviour.
1678    
1679    5.  The command line items --regex=pattern and --regexp=pattern were not
1680        recognized by pcregrep, which required --regex pattern or --regexp pattern
1681        (with a space rather than an '='). The man page documented the '=' forms,
1682        which are compatible with GNU grep; these now work.
1683    
1684    6.  No libpcreposix.pc file was created for pkg-config; there was just
1685        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1686    
1687    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1688        when UCP support is not needed, by modifying the Python script that
1689        generates it from Unicode data files. This should not matter if the module
1690        is correctly used as a library, but I received one complaint about 50K of
1691        unwanted data. My guess is that the person linked everything into his
1692        program rather than using a library. Anyway, it does no harm.
1693    
1694    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1695        was a minimum greater than 1 for a wide character in a possessive
1696        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1697        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1698        character. Chaos in the form of incorrect output or a compiling loop could
1699        result.
1700    
1701    9.  The restrictions on what a pattern can contain when partial matching is
1702        requested for pcre_exec() have been removed. All patterns can now be
1703        partially matched by this function. In addition, if there are at least two
1704        slots in the offset vector, the offset of the earliest inspected character
1705        for the match and the offset of the end of the subject are set in them when
1706        PCRE_ERROR_PARTIAL is returned.
1707    
1708    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1709        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1710        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1711        and may be more useful for multi-segment matching.
1712    
1713    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1714        used to be given if ever the end of the subject was reached; now it is
1715        given only if matching could not proceed because another character was
1716        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1717        string "Z", which now yields "no match" instead of "partial match". In the
1718        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1719        final character ended with (*FAIL).
1720    
1721    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1722        if the pattern had a "must contain" character that was already found in the
1723        earlier partial match, unless partial matching was again requested. For
1724        example, with the pattern /dog.(body)?/, the "must contain" character is
1725        "g". If the first part-match was for the string "dog", restarting with
1726        "sbody" failed. This bug has been fixed.
1727    
1728    13. The string returned by pcre_dfa_exec() after a partial match has been
1729        changed so that it starts at the first inspected character rather than the
1730        first character of the match. This makes a difference only if the pattern
1731        starts with a lookbehind assertion or \b or \B (\K is not supported by
1732        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1733        matching functions compatible, and I think it's the right thing to do.
1734    
1735    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1736        so that the demonstration program is easily available in environments where
1737        PCRE has not been installed from source.
1738    
1739    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1740        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1741        library.
1742    
1743    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1744        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1745        is not the first non-POSIX option to be added. Clearly some people find
1746        these options useful.
1747    
1748    17. If a caller to the POSIX matching function regexec() passes a non-zero
1749        value for nmatch with a NULL value for pmatch, the value of
1750        nmatch is forced to zero.
1751    
1752    18. RunGrepTest did not have a test for the availability of the -u option of
1753        the diff command, as RunTest does. It now checks in the same way as
1754        RunTest, and also checks for the -b option.
1755    
1756    19. If an odd number of negated classes containing just a single character
1757        interposed, within parentheses, between a forward reference to a named
1758        subpattern and the definition of the subpattern, compilation crashed with
1759        an internal error, complaining that it could not find the referenced
1760        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1761        [The bug was that it was starting one character too far in when skipping
1762        over the character class, thus treating the ] as data rather than
1763        terminating the class. This meant it could skip too much.]
1764    
1765    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1766        /g option in pcretest when the pattern contains \K, which makes it possible
1767        to have an empty string match not at the start, even when the pattern is
1768        anchored. Updated pcretest and pcredemo to use this option.
1769    
1770    21. If the maximum number of capturing subpatterns in a recursion was greater
1771        than the maximum at the outer level, the higher number was returned, but
1772        with unset values at the outer level. The correct (outer level) value is
1773        now given.
1774    
1775    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1776        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1777        make it do so. The string so far is captured, making this feature
1778        compatible with Perl.
1779    
1780    23. The tests have been re-organized, adding tests 11 and 12, to make it
1781        possible to check the Perl 5.10 features against Perl 5.10.
1782    
1783    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1784        pattern matches a fixed length string. PCRE did not allow this; now it
1785        does. Neither allows recursion.
1786    
1787    25. I finally figured out how to implement a request to provide the minimum
1788        length of subject string that was needed in order to match a given pattern.
1789        (It was back references and recursion that I had previously got hung up
1790        on.) This code has now been added to pcre_study(); it finds a lower bound
1791        to the length of subject needed. It is not necessarily the greatest lower
1792        bound, but using it to avoid searching strings that are too short does give
1793        some useful speed-ups. The value is available to calling programs via
1794        pcre_fullinfo().
1795    
1796    26. While implementing 25, I discovered to my embarrassment that pcretest had
1797        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1798        study optimizations had never been tested with that matching function.
1799        Oops. What is worse, even when it was passed study data, there was a bug in
1800        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1801        were also very few tests of studied patterns with pcre_dfa_exec().
1802    
1803    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1804        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1805        on the other side of the coin, they are no longer allowed to have different
1806        names, because these cannot be distinguished in PCRE, and this has caused
1807        confusion. (This is a difference from Perl.)
1808    
1809    28. When duplicate subpattern names are present (necessarily with different
1810        numbers, as required by 27 above), and a test is made by name in a
1811        conditional pattern, either for a subpattern having been matched, or for
1812        recursion in such a pattern, all the associated numbered subpatterns are
1813        tested, and the overall condition is true if the condition is true for any
1814        one of them. This is the way Perl works, and is also more like the way
1815        testing by number works.
1816    
1817    
1818    Version 7.9 11-Apr-09
1819    ---------------------
1820    
1821    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1822        (pcretest), all targets were linked against these libraries. This included
1823        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1824        libraries. This caused unwanted dependencies to be created. This problem
1825        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1826        pcretest is linked with readline.
1827    
1828    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1829        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1830        moved outside it again, because FALSE and TRUE are already defined in AIX,
1831        but BOOL is not.
1832    
1833    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1834        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1835    
1836    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1837        hyphens) following file names and line numbers when outputting matching
1838        lines. This is not true; no spaces are inserted. I have also clarified the
1839        wording for the --colour (or --color) option.
1840    
1841    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1842        was not coloured; this is different to GNU grep, so I have changed it to be
1843        the same.
1844    
1845    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1846        each matching line was coloured. Now it goes on to look for further matches
1847        of any of the test patterns, which is the same behaviour as GNU grep.
1848    
1849    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1850        doesn't make sense to accept an empty string match in pcregrep, so I have
1851        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1852        seems to be how GNU grep behaves. [But see later change 40 for release
1853        8.33.]
1854    
1855    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1856        start or after a newline", because the conditional assertion was not being
1857        correctly handled. The rule now is that both the assertion and what follows
1858        in the first alternative must satisfy the test.
1859    
1860    9.  If auto-callout was enabled in a pattern with a conditional group whose
1861        condition was an assertion, PCRE could crash during matching, both with
1862        pcre_exec() and pcre_dfa_exec().
1863    
1864    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1865        used for matching.
1866    
1867    11. Unicode property support in character classes was not working for
1868        characters (bytes) greater than 127 when not in UTF-8 mode.
1869    
1870    12. Added the -M command line option to pcretest.
1871    
1872    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1873    
1874    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1875    
1876    16. Added comments and documentation about mis-use of no_arg in the C++
1877        wrapper.
1878    
1879    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1880        from Martin Jerabek that uses macro names for all relevant character and
1881        string constants.
1882    
1883    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1884        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1885        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1886        these, but not everybody uses configure.
1887    
1888    19. A conditional group that had only one branch was not being correctly
1889        recognized as an item that could match an empty string. This meant that an
1890        enclosing group might also not be so recognized, causing infinite looping
1891        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1892        with the subject "ab", where knowledge that the repeated group can match
1893        nothing is needed in order to break the loop.
1894    
1895    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1896        exec(), but without supplying a callout function, matching went wrong.
1897    
1898    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1899        leak if the size of the offset vector was greater than 30. When the vector
1900        is smaller, the saved offsets during recursion go onto a local stack
1901        vector, but for larger vectors malloc() is used. It was failing to free
1902        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1903        error, in fact).
1904    
1905    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1906        heapframe that is used only when UTF-8 support is enabled. This caused no
1907        problem, but was untidy.
1908    
1909    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1910        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1911        included within another project.
1912    
1913    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1914        slightly modified by me:
1915    
1916          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1917              not building pcregrep.
1918    
1919          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1920              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1921    
1922    25. Forward references, both numeric and by name, in patterns that made use of
1923        duplicate group numbers, could behave incorrectly or give incorrect errors,
1924        because when scanning forward to find the reference group, PCRE was not
1925        taking into account the duplicate group numbers. A pattern such as
1926        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1927    
1928    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1929        the feature of a custom pointer more persuasive (as requested by a user).
1930    
1931    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1932        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1933        pre-defined.
1934    
1935    28. Added support for (*UTF8) at the start of a pattern.
1936    
1937    29. Arrange for flags added by the "release type" setting in CMake to be shown
1938        in the configuration summary.
1939    
1940    
1941    Version 7.8 05-Sep-08
1942  ---------------------  ---------------------
1943    
1944  1.  Replaced UCP searching code with optimized version as implemented for Ad  1.  Replaced UCP searching code with optimized version as implemented for Ad
# Line 31  Version 8.0 02 Jul-08 Line 1968  Version 8.0 02 Jul-08
1968    
1969  7.  Added two (int) casts to pcregrep when printing the difference of two  7.  Added two (int) casts to pcregrep when printing the difference of two
1970      pointers, in case they are 64-bit values.      pointers, in case they are 64-bit values.
1971    
1972  8.  Added comments about Mac OS X stack usage to the pcrestack man page and to  8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1973      test 2 if it fails.      test 2 if it fails.
1974    
1975  9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,  9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1976      and a #define of that name to empty if it is not externally set. This is to      and a #define of that name to empty if it is not externally set. This is to
1977      allow users of MSVC to set it if necessary.      allow users of MSVC to set it if necessary.
1978    
1979  10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from  10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1980      the convenience functions in the pcre_get.c source file.      the convenience functions in the pcre_get.c source file.
1981    
1982  11. An option change at the start of a pattern that had top-level alternatives  11. An option change at the start of a pattern that had top-level alternatives
1983      could cause overwriting and/or a crash. This command provoked a crash in      could cause overwriting and/or a crash. This command provoked a crash in
1984      some environments:      some environments:
1985    
1986        printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest        printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1987    
1988      This potential security problem was recorded as CVE-2008-2371.      This potential security problem was recorded as CVE-2008-2371.
1989    
1990  12. For a pattern where the match had to start at the beginning or immediately  12. For a pattern where the match had to start at the beginning or immediately
1991      after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and      after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1992      pcre_dfa_exec() could read past the end of the passed subject if there was      pcre_dfa_exec() could read past the end of the passed subject if there was
1993      no match. To help with detecting such bugs (e.g. with valgrind), I modified      no match. To help with detecting such bugs (e.g. with valgrind), I modified
1994      pcretest so that it places the subject at the end of its malloc-ed buffer.      pcretest so that it places the subject at the end of its malloc-ed buffer.
1995    
1996    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1997        exec() might read past the end of the data buffer in UTF-8 mode.
1998    
1999    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
2000        the data contained the byte 0x85 as part of a UTF-8 character within its
2001        first line. This applied both to normal and DFA matching.
2002    
2003    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
2004        /^[^d]*?$/8 failed to match "abc".
2005    
2006    16. Added a missing copyright notice to pcrecpp_internal.h.
2007    
2008    17. Make it more clear in the documentation that values returned from
2009        pcre_exec() in ovector are byte offsets, not character counts.
2010    
2011    18. Tidied a few places to stop certain compilers from issuing warnings.
2012    
2013    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
2014        supplied by Stefan Weber. I made a further small update for 7.8 because
2015        there is a change of source arrangements: the pcre_searchfuncs.c module is
2016        replaced by pcre_ucd.c.
2017    
2018    
2019  Version 7.7 07-May-08  Version 7.7 07-May-08
2020  ---------------------  ---------------------
# Line 134  Version 7.7 07-May-08 Line 2093  Version 7.7 07-May-08
2093      containing () gave an internal compiling error instead of "reference to      containing () gave an internal compiling error instead of "reference to
2094      non-existent subpattern". Fortunately, when the pattern did exist, the      non-existent subpattern". Fortunately, when the pattern did exist, the
2095      compiled code was correct. (When scanning forwards to check for the      compiled code was correct. (When scanning forwards to check for the
2096      existencd of the subpattern, it was treating the data ']' as terminating      existence of the subpattern, it was treating the data ']' as terminating
2097      the class, so got the count wrong. When actually compiling, the reference      the class, so got the count wrong. When actually compiling, the reference
2098      was subsequently set up correctly.)      was subsequently set up correctly.)
2099    

Legend:
Removed from v.361  
changed lines
  Added in v.1395

  ViewVC Help
Powered by ViewVC 1.1.5