ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 214 by ph10, Wed Aug 15 14:08:10 2007 UTC revision 1480 by ph10, Tue May 27 18:02:51 2014 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4  Version 7.3 09-Aug-07  Version 8.36 xx-xxx-2014
5    ------------------------
7    1.  Got rid of some compiler warnings in the C++ modules that were shown up by
8        -Wmissing-field-initializers and -Wunused-parameter.
10    2.  The tests for quantifiers being too big (greater than 65535) were being
11        applied after reading the number, and stupidly assuming that integer
12        overflow would give a negative number. The tests are now applied as the
13        numbers are read.
15    3.  Tidy code in pcre_exec.c where two branches that used to be different are
16        now the same.
18    4.  The JIT compiler did not generate match limit checks for certain
19        bracketed expressions with quantifiers. This may lead to exponential
20        backtracking, instead of returning with PCRE_ERROR_MATCHLIMIT. This
21        issue should be resolved now.
23    5.  Fixed an issue, which occures when nested alternatives are optimized
24        with table jumps.
26    6.  Inserted two casts and changed some ints to size_t in the light of some
27        reported 64-bit compiler warnings (Bugzilla 1477).
29    7.  Fixed a bug concerned with zero-minimum possessive groups that could match
30        an empty string, which sometimes were behaving incorrectly in the
31        interpreter (though correctly in the JIT matcher). This pcretest input is
32        an example:
34          '\A(?:[^"]++|"(?:[^"]*+|"")*+")++'
37        the interpreter was reporting a match of 'NON QUOTED ' only, whereas the
38        JIT matcher and Perl both matched 'NON QUOTED "QUOT""ED" AFTER '. The test
39        for an empty string was breaking the inner loop and carrying on at a lower
40        level, when possessive repeated groups should always return to a higher
41        level as they have no backtrack points in them. The empty string test now
42        occurs at the outer level.
44    8.  Fixed a bug that was incorrectly auto-possessifying \w+ in the pattern
45        ^\w+(?>\s*)(?<=\w) which caused it not to match "test test".
48    Version 8.35 04-April-2014
49    --------------------------
51    1.  A new flag is set, when property checks are present in an XCLASS.
52        When this flag is not set, PCRE can perform certain optimizations
53        such as studying these XCLASS-es.
55    2.  The auto-possessification of character sets were improved: a normal
56        and an extended character set can be compared now. Furthermore
57        the JIT compiler optimizes more character set checks.
59    3.  Got rid of some compiler warnings for potentially uninitialized variables
60        that show up only when compiled with -O2.
62    4.  A pattern such as (?=ab\K) that uses \K in an assertion can set the start
63        of a match later then the end of the match. The pcretest program was not
64        handling the case sensibly - it was outputting from the start to the next
65        binary zero. It now reports this situation in a message, and outputs the
66        text from the end to the start.
68    5.  Fast forward search is improved in JIT. Instead of the first three
69        characters, any three characters with fixed position can be searched.
70        Search order: first, last, middle.
72    6.  Improve character range checks in JIT. Characters are read by an inprecise
73        function now, which returns with an unknown value if the character code is
74        above a certain treshold (e.g: 256). The only limitation is that the value
75        must be bigger than the treshold as well. This function is useful, when
76        the characters above the treshold are handled in the same way.
78    7.  The macros whose names start with RAWUCHAR are placeholders for a future
79        mode in which only the bottom 21 bits of 32-bit data items are used. To
80        make this more memorable for those maintaining the code, the names have
81        been changed to start with UCHAR21, and an extensive comment has been added
82        to their definition.
84    8.  Add missing (new) files sljitNativeTILEGX.c and sljitNativeTILEGX-encoder.c
85        to the export list in Makefile.am (they were accidentally omitted from the
86        8.34 tarball).
88    9.  The informational output from pcretest used the phrase "starting byte set"
89        which is inappropriate for the 16-bit and 32-bit libraries. As the output
90        for "first char" and "need char" really means "non-UTF-char", I've changed
91        "byte" to "char", and slightly reworded the output. The documentation about
92        these values has also been (I hope) clarified.
94    10. Another JIT related optimization: use table jumps for selecting the correct
95        backtracking path, when more than four alternatives are present inside a
96        bracket.
98    11. Empty match is not possible, when the minimum length is greater than zero,
99        and there is no \K in the pattern. JIT should avoid empty match checks in
100        such cases.
102    12. In a caseless character class with UCP support, when a character with more
103        than one alternative case was not the first character of a range, not all
104        the alternative cases were added to the class. For example, s and \x{17f}
105        are both alternative cases for S: the class [RST] was handled correctly,
106        but [R-T] was not.
108    13. The configure.ac file always checked for pthread support when JIT was
109        enabled. This is not used in Windows, so I have put this test inside a
110        check for the presence of windows.h (which was already tested for).
112    14. Improve pattern prefix search by a simplified Boyer-Moore algorithm in JIT.
113        The algorithm provides a way to skip certain starting offsets, and usually
114        faster than linear prefix searches.
116    15. Change 13 for 8.20 updated RunTest to check for the 'fr' locale as well
117        as for 'fr_FR' and 'french'. For some reason, however, it then used the
118        Windows-specific input and output files, which have 'french' screwed in.
119        So this could never have worked. One of the problems with locales is that
120        they aren't always the same. I have now updated RunTest so that it checks
121        the output of the locale test (test 3) against three different output
122        files, and it allows the test to pass if any one of them matches. With luck
123        this should make the test pass on some versions of Solaris where it was
124        failing. Because of the uncertainty, the script did not used to stop if
125        test 3 failed; it now does. If further versions of a French locale ever
126        come to light, they can now easily be added.
128    16. If --with-pcregrep-bufsize was given a non-integer value such as "50K",
129        there was a message during ./configure, but it did not stop. This now
130        provokes an error. The invalid example in README has been corrected.
131        If a value less than the minimum is given, the minimum value has always
132        been used, but now a warning is given.
134    17. If --enable-bsr-anycrlf was set, the special 16/32-bit test failed. This
135        was a bug in the test system, which is now fixed. Also, the list of various
136        configurations that are tested for each release did not have one with both
137        16/32 bits and --enable-bar-anycrlf. It now does.
139    18. pcretest was missing "-C bsr" for displaying the \R default setting.
141    19. Little endian PowerPC systems are supported now by the JIT compiler.
143    20. The fast forward newline mechanism could enter to an infinite loop on
144        certain invalid UTF-8 input. Although we don't support these cases
145        this issue can be fixed by a performance optimization.
147    21. Change 33 of 8.34 is not sufficient to ensure stack safety because it does
148        not take account if existing stack usage. There is now a new global
149        variable called pcre_stack_guard that can be set to point to an external
150        function to check stack availability. It is called at the start of
151        processing every parenthesized group.
153    22. A typo in the code meant that in ungreedy mode the max/min qualifier
154        behaved like a min-possessive qualifier, and, for example, /a{1,3}b/U did
155        not match "ab".
157    23. When UTF was disabled, the JIT program reported some incorrect compile
158        errors. These messages are silenced now.
160    24. Experimental support for ARM-64 and MIPS-64 has been added to the JIT
161        compiler.
163    25. Change all the temporary files used in RunGrepTest to be different to those
164        used by RunTest so that the tests can be run simultaneously, for example by
165        "make -j check".
168    Version 8.34 15-December-2013
169    -----------------------------
171    1.  Add pcre[16|32]_jit_free_unused_memory to forcibly free unused JIT
172        executable memory. Patch inspired by Carsten Klein.
174    2.  ./configure --enable-coverage defined SUPPORT_GCOV in config.h, although
175        this macro is never tested and has no effect, because the work to support
176        coverage involves only compiling and linking options and special targets in
177        the Makefile. The comment in config.h implied that defining the macro would
178        enable coverage support, which is totally false. There was also support for
179        setting this macro in the CMake files (my fault, I just copied it from
180        configure). SUPPORT_GCOV has now been removed.
182    3.  Make a small performance improvement in strlen16() and strlen32() in
183        pcretest.
185    4.  Change 36 for 8.33 left some unreachable statements in pcre_exec.c,
186        detected by the Solaris compiler (gcc doesn't seem to be able to diagnose
187        these cases). There was also one in pcretest.c.
189    5.  Cleaned up a "may be uninitialized" compiler warning in pcre_exec.c.
191    6.  In UTF mode, the code for checking whether a group could match an empty
192        string (which is used for indefinitely repeated groups to allow for
193        breaking an infinite loop) was broken when the group contained a repeated
194        negated single-character class with a character that occupied more than one
195        data item and had a minimum repetition of zero (for example, [^\x{100}]* in
196        UTF-8 mode). The effect was undefined: the group might or might not be
197        deemed as matching an empty string, or the program might have crashed.
199    7.  The code for checking whether a group could match an empty string was not
200        recognizing that \h, \H, \v, \V, and \R must match a character.
202    8.  Implemented PCRE_INFO_MATCH_EMPTY, which yields 1 if the pattern can match
203        an empty string. If it can, pcretest shows this in its information output.
205    9.  Fixed two related bugs that applied to Unicode extended grapheme clusters
206        that were repeated with a maximizing qualifier (e.g. \X* or \X{2,5}) when
207        matched by pcre_exec() without using JIT:
209        (a) If the rest of the pattern did not match after a maximal run of
210            grapheme clusters, the code for backing up to try with fewer of them
211            did not always back up over a full grapheme when characters that do not
212            have the modifier quality were involved, e.g. Hangul syllables.
214        (b) If the match point in a subject started with modifier character, and
215            there was no match, the code could incorrectly back up beyond the match
216            point, and potentially beyond the first character in the subject,
217            leading to a segfault or an incorrect match result.
219    10. A conditional group with an assertion condition could lead to PCRE
220        recording an incorrect first data item for a match if no other first data
221        item was recorded. For example, the pattern (?(?=ab)ab) recorded "a" as a
222        first data item, and therefore matched "ca" after "c" instead of at the
223        start.
225    11. Change 40 for 8.33 (allowing pcregrep to find empty strings) showed up a
226        bug that caused the command "echo a | ./pcregrep -M '|a'" to loop.
228    12. The source of pcregrep now includes z/OS-specific code so that it can be
229        compiled for z/OS as part of the special z/OS distribution.
231    13. Added the -T and -TM options to pcretest.
233    14. The code in pcre_compile.c for creating the table of named capturing groups
234        has been refactored. Instead of creating the table dynamically during the
235        actual compiling pass, the information is remembered during the pre-compile
236        pass (on the stack unless there are more than 20 named groups, in which
237        case malloc() is used) and the whole table is created before the actual
238        compile happens. This has simplified the code (it is now nearly 150 lines
239        shorter) and prepared the way for better handling of references to groups
240        with duplicate names.
242    15. A back reference to a named subpattern when there is more than one of the
243        same name now checks them in the order in which they appear in the pattern.
244        The first one that is set is used for the reference. Previously only the
245        first one was inspected. This change makes PCRE more compatible with Perl.
247    16. Unicode character properties were updated from Unicode 6.3.0.
249    17. The compile-time code for auto-possessification has been refactored, based
250        on a patch by Zoltan Herczeg. It now happens after instead of during
251        compilation. The code is cleaner, and more cases are handled. The option
252        PCRE_NO_AUTO_POSSESS is added for testing purposes, and the -O and /O
253        options in pcretest are provided to set it. It can also be set by
254        (*NO_AUTO_POSSESS) at the start of a pattern.
256    18. The character VT has been added to the default ("C" locale) set of
257        characters that match \s and are generally treated as white space,
258        following this same change in Perl 5.18. There is now no difference between
259        "Perl space" and "POSIX space". Whether VT is treated as white space in
260        other locales depends on the locale.
262    19. The code for checking named groups as conditions, either for being set or
263        for being recursed, has been refactored (this is related to 14 and 15
264        above). Processing unduplicated named groups should now be as fast at
265        numerical groups, and processing duplicated groups should be faster than
266        before.
268    20. Two patches to the CMake build system, by Alexander Barkov:
270          (1) Replace the "source" command by "." in CMakeLists.txt because
271              "source" is a bash-ism.
273          (2) Add missing HAVE_STDINT_H and HAVE_INTTYPES_H to config-cmake.h.in;
274              without these the CMake build does not work on Solaris.
276    21. Perl has changed its handling of \8 and \9. If there is no previously
277        encountered capturing group of those numbers, they are treated as the
278        literal characters 8 and 9 instead of a binary zero followed by the
279        literals. PCRE now does the same.
281    22. Following Perl, added \o{} to specify codepoints in octal, making it
282        possible to specify values greater than 0777 and also making them
283        unambiguous.
285    23. Perl now gives an error for missing closing braces after \x{... instead of
286        treating the string as literal. PCRE now does the same.
288    24. RunTest used to grumble if an inappropriate test was selected explicitly,
289        but just skip it when running all tests. This make it awkward to run ranges
290        of tests when one of them was inappropriate. Now it just skips any
291        inappropriate tests, as it always did when running all tests.
293    25. If PCRE_AUTO_CALLOUT and PCRE_UCP were set for a pattern that contained
294        character types such as \d or \w, too many callouts were inserted, and the
295        data that they returned was rubbish.
297    26. In UCP mode, \s was not matching two of the characters that Perl matches,
298        namely NEL (U+0085) and MONGOLIAN VOWEL SEPARATOR (U+180E), though they
299        were matched by \h. The code has now been refactored so that the lists of
300        the horizontal and vertical whitespace characters used for \h and \v (which
301        are defined only in one place) are now also used for \s.
303    27. Add JIT support for the 64 bit TileGX architecture.
304        Patch by Jiong Wang (Tilera Corporation).
306    28. Possessive quantifiers for classes (both explicit and automatically
307        generated) now use special opcodes instead of wrapping in ONCE brackets.
309    29. Whereas an item such as A{4}+ ignored the possessivenes of the quantifier
310        (because it's meaningless), this was not happening when PCRE_CASELESS was
311        set. Not wrong, but inefficient.
313    30. Updated perltest.pl to add /u (force Unicode mode) when /W (use Unicode
314        properties for \w, \d, etc) is present in a test regex. Otherwise if the
315        test contains no characters greater than 255, Perl doesn't realise it
316        should be using Unicode semantics.
318    31. Upgraded the handling of the POSIX classes [:graph:], [:print:], and
319        [:punct:] when PCRE_UCP is set so as to include the same characters as Perl
320        does in Unicode mode.
322    32. Added the "forbid" facility to pcretest so that putting tests into the
323        wrong test files can sometimes be quickly detected.
325    33. There is now a limit (default 250) on the depth of nesting of parentheses.
326        This limit is imposed to control the amount of system stack used at compile
327        time. It can be changed at build time by --with-parens-nest-limit=xxx or
328        the equivalent in CMake.
330    34. Character classes such as [A-\d] or [a-[:digit:]] now cause compile-time
331        errors. Perl warns for these when in warning mode, but PCRE has no facility
332        for giving warnings.
334    35. Change 34 for 8.13 allowed quantifiers on assertions, because Perl does.
335        However, this was not working for (?!) because it is optimized to (*FAIL),
336        for which PCRE does not allow quantifiers. The optimization is now disabled
337        when a quantifier follows (?!). I can't see any use for this, but it makes
338        things uniform.
340    36. Perl no longer allows group names to start with digits, so I have made this
341        change also in PCRE. It simplifies the code a bit.
343    37. In extended mode, Perl ignores spaces before a + that indicates a
344        possessive quantifier. PCRE allowed a space before the quantifier, but not
345        before the possessive +. It now does.
347    38. The use of \K (reset reported match start) within a repeated possessive
348        group such as (a\Kb)*+ was not working.
350    40. Document that the same character tables must be used at compile time and
351        run time, and that the facility to pass tables to pcre_exec() and
352        pcre_dfa_exec() is for use only with saved/restored patterns.
354    41. Applied Jeff Trawick's patch CMakeLists.txt, which "provides two new
355        features for Builds with MSVC:
357        1. Support pcre.rc and/or pcreposix.rc (as is already done for MinGW
358           builds). The .rc files can be used to set FileDescription and many other
359           attributes.
361        2. Add an option (-DINSTALL_MSVC_PDB) to enable installation of .pdb files.
362           This allows higher-level build scripts which want .pdb files to avoid
363           hard-coding the exact files needed."
365    42. Added support for [[:<:]] and [[:>:]] as used in the BSD POSIX library to
366        mean "start of word" and "end of word", respectively, as a transition aid.
368    43. A minimizing repeat of a class containing codepoints greater than 255 in
369        non-UTF 16-bit or 32-bit modes caused an internal error when PCRE was
370        compiled to use the heap for recursion.
372    44. Got rid of some compiler warnings for unused variables when UTF but not UCP
373        is configured.
376    Version 8.33 28-May-2013
377    ------------------------
379    1.  Added 'U' to some constants that are compared to unsigned integers, to
380        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
381        variables that are added to signed variables, to ensure the result is
382        signed and can be negated.
384    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
385        CMake config files.
387    3.  Revise the creation of config.h.generic so that all boolean macros are
388        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
389        overriding via -D on the command line possible.
391    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
392        to unsigned int is reported to make a quite noticeable speed difference in
393        a specific Windows environment. Testing on Linux did also appear to show
394        some benefit (and it is clearly not harmful). Also fixed the definition of
395        Xop which should be unsigned.
397    5.  Related to (4), changing the definition of the intermediate variable cc
398        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
399        improvements.
401    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
402        unnecessary spaces.
404    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
405        and later.
407    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
409    9.  Optimizing fast_forward_start_bits in JIT.
411    10. Adding support for callouts in JIT, and fixing some issues revealed
412        during this work. Namely:
414        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
416        (b) Minimum length was not checked before the matching is started.
418    11. The value of capture_last that is passed to callouts was incorrect in some
419        cases when there was a capture on one path that was subsequently abandoned
420        after a backtrack. Also, the capture_last value is now reset after a
421        recursion, since all captures are also reset in this case.
423    12. The interpreter no longer returns the "too many substrings" error in the
424        case when an overflowing capture is in a branch that is subsequently
425        abandoned after a backtrack.
427    13. In the pathological case when an offset vector of size 2 is used, pcretest
428        now prints out the matched string after a yield of 0 or 1.
430    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
431        Only supported by the JIT compiler at the moment.
433    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
435    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
436        offset of the starting point of the matching process, provided the offsets
437        vector is large enough.
439    17. The \A escape now records a lookbehind value of 1, though its execution
440        does not actually inspect the previous character. This is to ensure that,
441        in partial multi-segment matching, at least one character from the old
442        segment is retained when a new segment is processed. Otherwise, if there
443        are no lookbehinds in the pattern, \A might match incorrectly at the start
444        of a new segment.
446    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
448    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
449        gives some modest performance improvement in 8-bit mode.
451    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
452        be expressed in certain programming languages using Universal Character
453        Names.
455    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
456        which points out that "non characters" are not "characters that may not
457        appear in Unicode strings" but rather "characters that are reserved for
458        internal use and have only local meaning".
460    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
461        there was a conditional group that depended on an assertion, if the
462        assertion was false, the callout that immediately followed the alternation
463        in the condition was skipped when pcre_exec() was used for matching.
465    23. Allow an explicit callout to be inserted before an assertion that is the
466        condition for a conditional group, for compatibility with automatic
467        callouts, which always insert a callout at this point.
469    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
470        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
472    25. (*PRUNE) is now supported by the JIT compiler.
474    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
476    27. Fix the case where there are two or more SKIPs with arguments that may be
477        ignored.
479    28. (*SKIP) is now supported by the JIT compiler.
481    29. (*THEN) is now supported by the JIT compiler.
483    30. Update RunTest with additional test selector options.
485    31. The way PCRE handles backtracking verbs has been changed in two ways.
487        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
488        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
489        backtracking. In some cases this makes it more Perl-compatible, but Perl's
490        rather obscure rules do not always do the same thing.
492        (2) Previously, backtracking verbs were confined within assertions. This is
493        no longer the case for positive assertions, except for (*ACCEPT). Again,
494        this sometimes improves Perl compatibility, and sometimes does not.
496    32. A number of tests that were in test 2 because Perl did things differently
497        have been moved to test 1, because either Perl or PCRE has changed, and
498        these tests are now compatible.
500    32. Backtracking control verbs are now handled in the same way in JIT and
501        interpreter.
503    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
504        contained a forward subroutine reference caused a compile error.
506    34. Auto-detect and optimize limited repetitions in JIT.
508    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
509        blocking (*UTF) etc.
511    36. In the interpreter, maximizing pattern repetitions for characters and
512        character types now use tail recursion, which reduces stack usage.
514    37. The value of the max lookbehind was not correctly preserved if a compiled
515        and saved regex was reloaded on a host of different endianness.
517    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
518        of the compiled pattern block, expand the flags field from 16 to 32 bits
519        because it was almost full.
521    39. Try madvise first before posix_madvise.
523    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
524        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
525        original change locked out all matches of empty strings. This has been
526        changed so that one match of an empty string per line is recognized.
527        Subsequent searches on the same line (for colouring or for --only-matching,
528        for example) do not recognize empty strings.
530    41. Applied a user patch to fix a number of spelling mistakes in comments.
532    42. Data lines longer than 65536 caused pcretest to crash.
534    43. Clarified the data type for length and startoffset arguments for pcre_exec
535        and pcre_dfa_exec in the function-specific man pages, where they were
536        explicitly stated to be in bytes, never having been updated. I also added
537        some clarification to the pcreapi man page.
539    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
540        a segmentation fault.
543    Version 8.32 30-November-2012
544    -----------------------------
546    1.  Improved JIT compiler optimizations for first character search and single
547        character iterators.
549    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
550        Patch by Daniel Richard G.
552    3.  Single character iterator optimizations in the JIT compiler.
554    4.  Improved JIT compiler optimizations for character ranges.
556    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
557        Reported by Giuseppe D'Angelo.
559    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
560        of a line, was being set incorrectly in cases where .* appeared inside
561        atomic brackets at the start of a pattern, or where there was a subsequent
562        *PRUNE or *SKIP.
564    7.  Improved instruction cache flush for POWER/PowerPC.
565        Patch by Daniel Richard G.
567    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
568        grep:
570        (a) There is now no limit to the number of patterns to be matched.
572        (b) An error is given if a pattern is too long.
574        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
575            are now supported.
577        (d) --exclude-from and --include-from (multiple use) have been added.
579        (e) Exclusions and inclusions now apply to all files and directories, not
580            just to those obtained from scanning a directory recursively.
582        (f) Multiple uses of -f and --file-list are now supported.
584        (g) In a Windows environment, the default for -d has been changed from
585            "read" (the GNU grep default) to "skip", because otherwise the presence
586            of a directory in the file list provokes an error.
588        (h) The documentation has been revised and clarified in places.
590    9.  Improve the matching speed of capturing brackets.
592    10. Changed the meaning of \X so that it now matches a Unicode extended
593        grapheme cluster.
595    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
596        out POSIX threads when JIT support is configured.
598    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
600    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
601        structure could go wrong in environments where size_t is not the same size
602        as int.
604    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
605        set.
607    15. The EBCDIC support had decayed; later updates to the code had included
608        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
609        general tidy up of EBCDIC-related issues, and the documentation was also
610        not quite right. There is now a test that can be run on ASCII systems to
611        check some of the EBCDIC-related things (but is it not a full test).
613    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
614        in a small tidy to the code.
616    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
618    18. If the --only-matching (-o) option in pcregrep is specified multiple
619        times, each one causes appropriate output. For example, -o1 -o2 outputs the
620        substrings matched by the 1st and 2nd capturing parentheses. A separating
621        string can be specified by --om-separator (default empty).
623    19. Improving the first n character searches.
625    20. Turn case lists for horizontal and vertical white space into macros so that
626        they are defined only once.
628    21. This set of changes together give more compatible Unicode case-folding
629        behaviour for characters that have more than one other case when UCP
630        support is available.
632        (a) The Unicode property table now has offsets into a new table of sets of
633            three or more characters that are case-equivalent. The MultiStage2.py
634            script that generates these tables (the pcre_ucd.c file) now scans
635            CaseFolding.txt instead of UnicodeData.txt for character case
636            information.
638        (b) The code for adding characters or ranges of characters to a character
639            class has been abstracted into a generalized function that also handles
640            case-independence. In UTF-mode with UCP support, this uses the new data
641            to handle characters with more than one other case.
643        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
644            whose other case is greater than 256 are now correctly matched
645            caselessly. Previously, the high codepoint matched the low one, but not
646            vice versa.
648        (d) The processing of \h, \H, \v, and \ in character classes now makes use
649            of the new class addition function, using character lists defined as
650            macros alongside the case definitions of 20 above.
652        (e) Caseless back references now work with characters that have more than
653            one other case.
655        (f) General caseless matching of characters with more than one other case
656            is supported.
658    22. Unicode character properties were updated from Unicode 6.2.0
660    23. Improved CMake support under Windows. Patch by Daniel Richard G.
662    24. Add support for 32-bit character strings, and UTF-32
664    25. Major JIT compiler update (code refactoring and bugfixing).
665        Experimental Sparc 32 support is added.
667    26. Applied a modified version of Daniel Richard G's patch to create
668        pcre.h.generic and config.h.generic by "make" instead of in the
669        PrepareRelease script.
671    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
672        pcre_compile.c when checking for a zero character.
674    28. Introducing a native interface for JIT. Through this interface, the compiled
675        machine code can be directly executed. The purpose of this interface is to
676        provide fast pattern matching, so several sanity checks are not performed.
677        However, feature tests are still performed. The new interface provides
678        1.4x speedup compared to the old one.
680    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
681        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
682        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
684    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
685        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
686        the "old" RFC 2279). Instead, it ended up passing a negative length to
687        pcre_exec().
689    31. Add support for GCC's visibility feature to hide internal functions.
691    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
692        "unknown -C option" after outputting 0 or 1.
694    33. There is now support for generating a code coverage report for the test
695        suite in environments where gcc is the compiler and lcov is installed. This
696        is mainly for the benefit of the developers.
698    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
699        unaddressable using valgrind annotations, allowing valgrind to detect
700        invalid memory accesses. This is mainly for the benefit of the developers.
702    25. (*UTF) can now be used to start a pattern in any of the three libraries.
704    26. Give configure error if --enable-cpp but no C++ compiler found.
707    Version 8.31 06-July-2012
708    -------------------------
710    1.  Fixing a wrong JIT test case and some compiler warnings.
712    2.  Removed a bashism from the RunTest script.
714    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
715        to unsigned type, result still unsigned" that was given by an MS compiler
716        on encountering the code "-sizeof(xxx)".
718    4.  Partial matching support is added to the JIT compiler.
720    5.  Fixed several bugs concerned with partial matching of items that consist
721        of more than one character:
723        (a) /^(..)\1/ did not partially match "aba" because checking references was
724            done on an "all or nothing" basis. This also applied to repeated
725            references.
727        (b) \R did not give a hard partial match if \r was found at the end of the
728            subject.
730        (c) \X did not give a hard partial match after matching one or more
731            characters at the end of the subject.
733        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
734            a partial match for the string "\r".
736        (e) When newline was set to CRLF, the metacharacter "." did not recognize
737            a partial match for a CR character at the end of the subject string.
739    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
740        running pcretest, the text "(JIT)" added to the output whenever JIT is
741        actually used to run the match.
743    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
744        or /S+[+] with a digit between 1 and 7.
746    8.  OP_NOT now supports any UTF character not just single-byte ones.
748    9.  (*MARK) control verb is now supported by the JIT compiler.
750    10. The command "./RunTest list" lists the available tests without actually
751        running any of them. (Because I keep forgetting what they all are.)
755    12. Applied a (slightly modified) user-supplied patch that improves performance
756        when the heap is used for recursion (compiled with --disable-stack-for-
757        recursion). Instead of malloc and free for each heap frame each time a
758        logical recursion happens, frames are retained on a chain and re-used where
759        possible. This sometimes gives as much as 30% improvement.
761    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
762        call.
764    14. As documented, (*COMMIT) is now confined to within a positive assertion.
766    15. It is now possible to link pcretest with libedit as an alternative to
767        libreadline.
769    16. (*COMMIT) control verb is now supported by the JIT compiler.
771    17. The Unicode data tables have been updated to Unicode 6.1.0.
773    18. Added --file-list option to pcregrep.
775    19. Added binary file support to pcregrep, including the -a, --binary-files,
776        -I, and --text options.
778    20. The madvise function is renamed for posix_madvise for QNX compatibility
779        reasons. Fixed by Giuseppe D'Angelo.
781    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
783    22. Changed the option for creating symbolic links for 16-bit man pages from
784        -s to -sf so that re-installing does not cause issues.
786    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
788    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
789        matches in certain environments (the workspace was not being correctly
790        retained). Also added to pcre_dfa_exec() a simple plausibility check on
791        some of the workspace data at the beginning of a restart.
793    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
794        was not doing so when it should - probably a typo introduced by SVN 528
795        (change 8.10/14).
797    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
798        \w+ when the character tables indicated that \x{c4} was a word character.
799        There were several related cases, all because the tests for doing a table
800        lookup were testing for characters less than 127 instead of 255.
802    27. If a pattern contains capturing parentheses that are not used in a match,
803        their slots in the ovector are set to -1. For those that are higher than
804        any matched groups, this happens at the end of processing. In the case when
805        there were back references that the ovector was too small to contain
806        (causing temporary malloc'd memory to be used during matching), and the
807        highest capturing number was not used, memory off the end of the ovector
808        was incorrectly being set to -1. (It was using the size of the temporary
809        memory instead of the true size.)
811    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
812        ovector size, it uses memory at the end of the block that it has got.
814    29. Check for an overlong MARK name and give an error at compile time. The
815        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
817    30. JIT compiler update.
819    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
820        Rill for the patch.
822    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
824    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
826    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
827        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
828        was enabled.
830    35. Improve JIT code generation for greedy plus quantifier.
832    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
833        1 to "aa" instead of to an empty string. The bug affected repeated groups
834        that could potentially match an empty string.
836    37. Optimizing single character iterators in JIT.
838    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
839        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
840        codepoints that are too big for the mode are faulted, and in a UTF mode,
841        disallowed codepoints are also faulted.
843    39. If PCRE was compiled with UTF support, in three places in the DFA
844        matcher there was code that should only have been obeyed in UTF mode, but
845        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
846        processing when bytes with values greater than 127 were present. In 16-bit
847        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
848        both cases the values are those that cannot be the first data item in a UTF
849        character. The three items that might have provoked this were recursions,
850        possessively repeated groups, and atomic groups.
852    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
853        and pcregrep, because some OS require shared objects to be explicitly
854        passed to ld, causing the link step to fail if they are not.
856    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
857        mode, patterns that started with \h* or \R* might be incorrectly matched.
860    Version 8.30 04-February-2012
861    -----------------------------
863    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
864        name is defined in ctype.h.
866    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
867        only in quite long subpatterns.
869    3.  Removed the function pcre_info(), which has been obsolete and deprecated
870        since it was replaced by pcre_fullinfo() in February 2000.
872    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
873        match a (*MARK), and the match failed at the start of the subject, a
874        reference to memory before the start of the subject could occur. This bug
875        was introduced by fix 17 of release 8.21.
877    5.  A reference to an unset group with zero minimum repetition was giving
878        totally wrong answers (in non-JavaScript-compatibility mode). For example,
879        /(another)?(\1?)test/ matched against "hello world test". This bug was
880        introduced in release 8.13.
882    6.  Add support for 16-bit character strings (a large amount of work involving
883        many changes and refactorings).
885    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
886        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
887        from a file.
889    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
890        rounding is not applied in this particular case).
892    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
893        if they appear, or are escaped, in patterns.
895    10. Get rid of a number of -Wunused-but-set-variable warnings.
897    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
898        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
899        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
900        also returns the mark "x". This bug applied to capturing parentheses,
901        non-capturing parentheses, and atomic parentheses. It also applied to some
902        assertions.
904    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
905        information out of configure.ac instead of relying on pcre.h.generic, which
906        is not stored in the repository.
908    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
909        -lreadline.
911    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
913    15. Applied Graycode's patch to put the top-level frame on the stack rather
914        than the heap when not using the stack for recursion. This gives a
915        performance improvement in many cases when recursion is not deep.
917    16. Experimental code added to "pcretest -C" to output the stack frame size.
920    Version 8.21 12-Dec-2011
921    ------------------------
923    1.  Updating the JIT compiler.
925    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
926        are added as well.
928    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
929        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
930        calling _pcre_jit_exec. Some extra comments are added.
932    4.  (*MARK) settings inside atomic groups that do not contain any capturing
933        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
934        was introduced by change 18 for 8.20.
936    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
937        ECMA-262 standard.
939    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
940        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
941        This bug was probably introduced by change 9 of 8.13.
943    7.  While fixing 6 above, I noticed that a number of other items were being
944        incorrectly rejected as "not fixed length". This arose partly because newer
945        opcodes had not been added to the fixed-length checking code. I have (a)
946        corrected the bug and added tests for these items, and (b) arranged for an
947        error to occur if an unknown opcode is encountered while checking for fixed
948        length instead of just assuming "not fixed length". The items that were
949        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
950        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
951        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
953    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
954        being incorrectly compiled and would have given unpredicatble results.
956    9.  A possessively repeated subpattern with minimum repeat count greater than
957        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
958        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
959        the first (A) could occur when it should not.
961    10. Add a cast and remove a redundant test from the code.
963    11. JIT should use pcre_malloc/pcre_free for allocation.
965    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
966        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
967        is anything other than /usr, -L is still shown).
969    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
971    14. Perl does not support \N without a following name in a [] class; PCRE now
972        also gives an error.
974    15. If a forward reference was repeated with an upper limit of around 2000,
975        it caused the error "internal error: overran compiling workspace". The
976        maximum number of forward references (including repeats) was limited by the
977        internal workspace, and dependent on the LINK_SIZE. The code has been
978        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
979        the default depends on LINK_SIZE. There is a new upper limit (for safety)
980        of around 200,000 forward references. While doing this, I also speeded up
981        the filling in of repeated forward references.
983    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
984        incorrectly expecting the subject to contain another "a" after the start.
986    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
987        in the match, the SKIP should be ignored. This was not happening; instead
988        the SKIP was being treated as NOMATCH. For patterns such as
989        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
990        tested.
992    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
993        now much more compatible with Perl, in particular in cases where the result
994        is a non-match for a non-anchored pattern. For example, if
995        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
996        "m", where previously it did not return a name. A side effect of this
997        change is that for partial matches, the last encountered mark name is
998        returned, as for non matches. A number of tests that were previously not
999        Perl-compatible have been moved into the Perl-compatible test files. The
1000        refactoring has had the pleasing side effect of removing one argument from
1001        the match() function, thus reducing its stack requirements.
1003    19. If the /S+ option was used in pcretest to study a pattern using JIT,
1004        subsequent uses of /S (without +) incorrectly behaved like /S+.
1006    21. Retrieve executable code size support for the JIT compiler and fixing
1007        some warnings.
1009    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
1010        not work when the shorter character appeared right at the end of the
1011        subject string.
1013    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
1014        systems.
1016    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
1017        output it when the /M option is used in pcretest.
1019    25. The CheckMan script was not being included in the distribution. Also, added
1020        an explicit "perl" to run Perl scripts from the PrepareRelease script
1021        because this is reportedly needed in Windows.
1023    26. If study data was being save in a file and studying had not found a set of
1024        "starts with" bytes for the pattern, the data written to the file (though
1025        never used) was taken from uninitialized memory and so caused valgrind to
1026        complain.
1028    27. Updated RunTest.bat as provided by Sheri Pierce.
1030    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
1032    29. Computation of memory usage for the table of capturing group names was
1033        giving an unnecessarily large value.
1036    Version 8.20 21-Oct-2011
1037    ------------------------
1039    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
1040        a POSIX class. After further experiments with Perl, which convinced me that
1041        Perl has bugs and confusions, a closing square bracket is no longer allowed
1042        in a POSIX name. This bug also affected patterns with classes that started
1043        with full stops.
1045    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
1046        captured substring, but while checking the failing first alternative,
1047        substring 1 is temporarily captured. If the output vector supplied to
1048        pcre_exec() was not big enough for this capture, the yield of the function
1049        was still zero ("insufficient space for captured substrings"). This cannot
1050        be totally fixed without adding another stack variable, which seems a lot
1051        of expense for a edge case. However, I have improved the situation in cases
1052        such as /(a)(b)x|abc/ matched against "abc", where the return code
1053        indicates that fewer than the maximum number of slots in the ovector have
1054        been set.
1056    3.  Related to (2) above: when there are more back references in a pattern than
1057        slots in the output vector, pcre_exec() uses temporary memory during
1058        matching, and copies in the captures as far as possible afterwards. It was
1059        using the entire output vector, but this conflicts with the specification
1060        that only 2/3 is used for passing back captured substrings. Now it uses
1061        only the first 2/3, for compatibility. This is, of course, another edge
1062        case.
1064    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
1065        main code base, and can be used by building with --enable-jit. When this is
1066        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
1067        runtime --no-jit option is given.
1069    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
1070        ovector, the return from the function was zero, implying that there were
1071        other matches that did not fit. The correct "exactly full" value is now
1072        returned.
1074    6.  If a subpattern that was called recursively or as a subroutine contained
1075        (*PRUNE) or any other control that caused it to give a non-standard return,
1076        invalid errors such as "Error -26 (nested recursion at the same subject
1077        position)" or even infinite loops could occur.
1079    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
1080        computing the minimum length on reaching *ACCEPT, and so ended up with the
1081        wrong value of 1 rather than 0. Further investigation indicates that
1082        computing a minimum subject length in the presence of *ACCEPT is difficult
1083        (think back references, subroutine calls), and so I have changed the code
1084        so that no minimum is registered for a pattern that contains *ACCEPT.
1086    8.  If (*THEN) was present in the first (true) branch of a conditional group,
1087        it was not handled as intended. [But see 16 below.]
1089    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
1090        Sheri Pierce.
1092    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
1093        the first byte in a match must be "a".
1095    11. Change 17 for 8.13 increased the recursion depth for patterns like
1096        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
1097        pattern contains any instances of (*THEN). If it does not, the old
1098        optimizations are restored. It would be nice to do this on a per-group
1099        basis, but at the moment that is not feasible.
1101    12. In some environments, the output of pcretest -C is CRLF terminated. This
1102        broke RunTest's code that checks for the link size. A single white space
1103        character after the value is now allowed for.
1105    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
1106        For "fr", it uses the Windows-specific input and output files.
1108    14. If (*THEN) appeared in a group that was called recursively or as a
1109        subroutine, it did not work as intended. [But see next item.]
1111    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
1112        pattern fragments (but not containing any | characters). If A and B are
1113        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
1114        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
1115        D. In other words, Perl considers parentheses that do not contain any |
1116        characters to be part of a surrounding alternative, whereas PCRE was
1117        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
1118        differently. PCRE now behaves in the same way as Perl, except in the case
1119        of subroutine/recursion calls such as (?1) which have in any case always
1120        been different (but PCRE had them first :-).
1122    16. Related to 15 above: Perl does not treat the | in a conditional group as
1123        creating alternatives. Such a group is treated in the same way as an
1124        ordinary group without any | characters when processing (*THEN). PCRE has
1125        been changed to match Perl's behaviour.
1127    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
1128        RunGrepTest script failed.
1130    18. Change 22 for version 13 caused atomic groups to use more stack. This is
1131        inevitable for groups that contain captures, but it can lead to a lot of
1132        stack use in large patterns. The old behaviour has been restored for atomic
1133        groups that do not contain any capturing parentheses.
1135    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
1136        suppress the check for a minimum subject length at run time. (If it was
1137        given to pcre_exec() or pcre_dfa_exec() it did work.)
1139    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
1140        fail to work when decoding hex characters in data strings in EBCDIC
1141        environments.
1143    21. It appears that in at least one Mac OS environment, the isxdigit() function
1144        is implemented as a macro that evaluates to its argument more than once,
1145        contravening the C 90 Standard (I haven't checked a later standard). There
1146        was an instance in pcretest which caused it to go wrong when processing
1147        \x{...} escapes in subject strings. The has been rewritten to avoid using
1148        things like p++ in the argument of isxdigit().
1151    Version 8.13 16-Aug-2011
1152    ------------------------
1154    1.  The Unicode data tables have been updated to Unicode 6.0.0.
1156    2.  Two minor typos in pcre_internal.h have been fixed.
1158    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
1159        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
1160        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
1162    4.  There were a number of related bugs in the code for matching backrefences
1163        caselessly in UTF-8 mode when codes for the characters concerned were
1164        different numbers of bytes. For example, U+023A and U+2C65 are an upper
1165        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
1166        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
1167        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
1168        2-byte code at the end of the subject (it thought there wasn't enough data
1169        left).
1171    5.  Comprehensive information about what went wrong is now returned by
1172        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
1173        as the output vector has at least 2 elements. The offset of the start of
1174        the failing character and a reason code are placed in the vector.
1176    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
1177        now returned is for the first byte of the failing character, instead of the
1178        last byte inspected. This is an incompatible change, but I hope it is small
1179        enough not to be a problem. It makes the returned offset consistent with
1180        pcre_exec() and pcre_dfa_exec().
1182    7.  pcretest now gives a text phrase as well as the error number when
1183        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
1184        failure, the offset and reason code are output.
1186    8.  When \R was used with a maximizing quantifier it failed to skip backwards
1187        over a \r\n pair if the subsequent match failed. Instead, it just skipped
1188        back over a single character (\n). This seems wrong (because it treated the
1189        two characters as a single entity when going forwards), conflicts with the
1190        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
1191        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
1192        has been changed.
1194    9.  Some internal refactoring has changed the processing so that the handling
1195        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
1196        time (the PCRE_DOTALL option was changed this way some time ago: version
1197        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
1198        which was always a bit of a fudge. It also means that there is one less
1199        argument for the match() function, which reduces its stack requirements
1200        slightly. This change also fixes an incompatibility with Perl: the pattern
1201        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
1203    10. More internal refactoring has drastically reduced the number of recursive
1204        calls to match() for possessively repeated groups such as (abc)++ when
1205        using pcre_exec().
1207    11. While implementing 10, a number of bugs in the handling of groups were
1208        discovered and fixed:
1210        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
1211        (a|)*(?1) gave a compile-time internal error.
1212        ((a|)+)+  did not notice that the outer group could match an empty string.
1213        (^a|^)+   was not marked as anchored.
1214        (.*a|.*)+ was not marked as matching at start or after a newline.
1216    12. Yet more internal refactoring has removed another argument from the match()
1217        function. Special calls to this function are now indicated by setting a
1218        value in a variable in the "match data" data block.
1220    13. Be more explicit in pcre_study() instead of relying on "default" for
1221        opcodes that mean there is no starting character; this means that when new
1222        ones are added and accidentally left out of pcre_study(), testing should
1223        pick them up.
1225    14. The -s option of pcretest has been documented for ages as being an old
1226        synonym of -m (show memory usage). I have changed it to mean "force study
1227        for every regex", that is, assume /S for every regex. This is similar to -i
1228        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
1229        using it. It makes it easier to run collections of tests with and without
1230        study enabled, and thereby test pcre_study() more easily. All the standard
1231        tests are now run with and without -s (but some patterns can be marked as
1232        "never study" - see 20 below).
1234    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
1235        restoration of the capturing data to the outer values was not happening
1236        correctly.
1238    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
1239        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
1240        pattern had matched an empty string, and so incorrectly returned a no
1241        match.
1243    17. There was optimizing code for the last branch of non-capturing parentheses,
1244        and also for the obeyed branch of a conditional subexpression, which used
1245        tail recursion to cut down on stack usage. Unfortunately, now that there is
1246        the possibility of (*THEN) occurring in these branches, tail recursion is
1247        no longer possible because the return has to be checked for (*THEN). These
1248        two optimizations have therefore been removed. [But see 8.20/11 above.]
1250    18. If a pattern containing \R was studied, it was assumed that \R always
1251        matched two bytes, thus causing the minimum subject length to be
1252        incorrectly computed because \R can also match just one byte.
1254    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
1255        was incorrectly computed.
1257    20. If /S is present twice on a test pattern in pcretest input, it now
1258        *disables* studying, thereby overriding the use of -s on the command line
1259        (see 14 above). This is necessary for one or two tests to keep the output
1260        identical in both cases.
1262    21. When (*ACCEPT) was used in an assertion that matched an empty string and
1263        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
1265    22. When an atomic group that contained a capturing parenthesis was
1266        successfully matched, but the branch in which it appeared failed, the
1267        capturing was not being forgotten if a higher numbered group was later
1268        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
1269        group 1 to "a", when in fact it should be unset. This applied to multi-
1270        branched capturing and non-capturing groups, repeated or not, and also to
1271        positive assertions (capturing in negative assertions does not happen
1272        in PCRE) and also to nested atomic groups.
1274    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
1275        subject after a captured substring, to make it easier to tell which of a
1276        number of identical substrings has been captured.
1278    24. The way atomic groups are processed by pcre_exec() has been changed so that
1279        if they are repeated, backtracking one repetition now resets captured
1280        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
1281        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
1282        "aaa". Previously, it would have been "a". As part of this code
1283        refactoring, the way recursive calls are handled has also been changed.
1285    25. If an assertion condition captured any substrings, they were not passed
1286        back unless some other capturing happened later. For example, if
1287        (?(?=(a))a) was matched against "a", no capturing was returned.
1289    26. When studying a pattern that contained subroutine calls or assertions,
1290        the code for finding the minimum length of a possible match was handling
1291        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
1292        group 1 called group 2 while simultaneously a separate group 2 called group
1293        1). A stack overflow occurred in this case. I have fixed this by limiting
1294        the recursion depth to 10.
1296    27. Updated RunTest.bat in the distribution to the version supplied by Tom
1297        Fortmann. This supports explicit test numbers on the command line, and has
1298        argument validation and error reporting.
1300    28. An instance of \X with an unlimited repeat could fail if at any point the
1301        first character it looked at was a mark character.
1303    29. Some minor code refactoring concerning Unicode properties and scripts
1304        should reduce the stack requirement of match() slightly.
1306    30. Added the '=' option to pcretest to check the setting of unused capturing
1307        slots at the end of the pattern, which are documented as being -1, but are
1308        not included in the return count.
1310    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
1311        compiled something random. Now it gives a compile-time error (as does
1312        Perl).
1314    32. A *MARK encountered during the processing of a positive assertion is now
1315        recorded and passed back (compatible with Perl).
1317    33. If --only-matching or --colour was set on a pcregrep call whose pattern
1318        had alternative anchored branches, the search for a second match in a line
1319        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
1320        matched the line "0102" twice. The same bug affected patterns that started
1321        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
1322        twice.
1324    34. Previously, PCRE did not allow quantification of assertions. However, Perl
1325        does, and because of capturing effects, quantifying parenthesized
1326        assertions may at times be useful. Quantifiers are now allowed for
1327        parenthesized assertions.
1329    35. A minor code tidy in pcre_compile() when checking options for \R usage.
1331    36. \g was being checked for fancy things in a character class, when it should
1332        just be a literal "g".
1334    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
1335        appearance of a nested POSIX class supersedes an apparent external class.
1336        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
1337        unescaped square brackets may also appear as part of class names. For
1338        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
1339        more like Perl. (But see 8.20/1 above.)
1341    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
1342        was because it thought it was \N{name}, which is not supported).
1344    39. Add minix to OS list not supporting the -S option in pcretest.
1346    40. PCRE tries to detect cases of infinite recursion at compile time, but it
1347        cannot analyze patterns in sufficient detail to catch mutual recursions
1348        such as ((?1))((?2)). There is now a runtime test that gives an error if a
1349        subgroup is called recursively as a subpattern for a second time at the
1350        same position in the subject string. In previous releases this might have
1351        been caught by the recursion limit, or it might have run out of stack.
1353    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
1354        happen only once. PCRE was, however incorrectly giving a compile time error
1355        "recursive call could loop indefinitely" because it cannot analyze the
1356        pattern in sufficient detail. The compile time test no longer happens when
1357        PCRE is compiling a conditional subpattern, but actual runaway loops are
1358        now caught at runtime (see 40 above).
1360    42. It seems that Perl allows any characters other than a closing parenthesis
1361        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
1362        has been changed to be the same.
1364    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
1365        as not to get warnings when autogen.sh is called. Also changed
1366        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
1368    44. To help people who use pcregrep to scan files containing exceedingly long
1369        lines, the following changes have been made:
1371        (a) The default value of the buffer size parameter has been increased from
1372            8K to 20K. (The actual buffer used is three times this size.)
1374        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
1375            PCRE is built.
1377        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
1378            to be set at run time.
1380        (d) Numerical values in pcregrep options can be followed by K or M, for
1381            example --buffer-size=50K.
1383        (e) If a line being scanned overflows pcregrep's buffer, an error is now
1384            given and the return code is set to 2.
1386    45. Add a pointer to the latest mark to the callout data block.
1388    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
1389        partial match of an empty string instead of no match. This was specific to
1390        the use of ".".
1392    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
1393        complete match instead of a partial match. This bug was dependent on both
1394        the PCRE_UTF8 and PCRE_DOTALL options being set.
1396    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1397        starting byte set, because \b was not being ignored.
1400    Version 8.12 15-Jan-2011
1401    ------------------------
1403    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1404        checks for such things as part of the documentation building process.
1406    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1407        --match-limit and --recursion-limit options (added for 8.11). In
1408        particular, this made one of the standard tests fail. (The integer value
1409        went into the wrong half of a long int.)
1411    3.  If the --colour option was given to pcregrep with -v (invert match), it
1412        did strange things, either producing crazy output, or crashing. It should,
1413        of course, ignore a request for colour when reporting lines that do not
1414        match.
1416    4.  Another pcregrep bug caused similar problems if --colour was specified with
1417        -M (multiline) and the pattern match finished with a line ending.
1419    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1420        matched in multiline mode, the following line was shown as part of the
1421        match. This seems wrong, so I have changed it.
1423    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1424        the check for further matches in the same line (so they could be coloured)
1425        to overrun the end of the current line. If another match was found, it was
1426        incorrectly shown (and then shown again when found in the next line).
1428    7.  If pcregrep was compiled under Windows, there was a reference to the
1429        function pcregrep_exit() before it was defined. I am assuming this was
1430        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1431        reported by a user. I've moved the definition above the reference.
1434    Version 8.11 10-Dec-2010
1435    ------------------------
1437    1.  (*THEN) was not working properly if there were untried alternatives prior
1438        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1439        backtracked to try for "b" instead of moving to the next alternative branch
1440        at the same level (in this case, to look for "c"). The Perl documentation
1441        is clear that when (*THEN) is backtracked onto, it goes to the "next
1442        alternative in the innermost enclosing group".
1444    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1445        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1446        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1447        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1448        (*THEN).
1450    3.  If \s appeared in a character class, it removed the VT character from
1451        the class, even if it had been included by some previous item, for example
1452        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1453        of \s, but is part of the POSIX "space" class.)
1455    4.  A partial match never returns an empty string (because you can always
1456        match an empty string at the end of the subject); however the checking for
1457        an empty string was starting at the "start of match" point. This has been
1458        changed to the "earliest inspected character" point, because the returned
1459        data for a partial match starts at this character. This means that, for
1460        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1461        (previously it gave "no match").
1463    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1464        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1465        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1466        has an implication that the given string is incomplete (because a partial
1467        match is preferred over a full match). For this reason, these items now
1468        give a partial match in this situation. [Aside: previously, the one case
1469        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1470        match rather than a full match, which was wrong by the old rules, but is
1471        now correct.]
1473    6.  There was a bug in the handling of #-introduced comments, recognized when
1474        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1475        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1476        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1477        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1478        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1479        places in pcre_compile().
1481    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1482        comments when looking ahead for named forward references to subpatterns,
1483        the only newline sequence it recognized was NL. It now handles newlines
1484        according to the set newline convention.
1486    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1487        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1488        cater for a lack of strerror(). These oversights have been fixed.
1490    9.  Added --match-limit and --recursion-limit to pcregrep.
1492    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1494    11. When the -o option was used, pcregrep was setting a return code of 1, even
1495        when matches were found, and --line-buffered was not being honoured.
1497    12. Added an optional parentheses number to the -o and --only-matching options
1498        of pcregrep.
1500    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1501        can match an empty string. The code to do it in pcretest and pcredemo
1502        needed fixing:
1504        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1505            only one byte after an empty string match just before CRLF (this case
1506            just got forgotten; "any" and "anycrlf" were OK).
1508        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1509            mode when an empty string match preceded an ASCII character followed by
1510            a non-ASCII character. (The code for advancing by one character rather
1511            than one byte was nonsense.)
1513        (c) The pcredemo.c sample program did not have any code at all to handle
1514            the cases when CRLF is a valid newline sequence.
1516    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1517        as a starting offset was within the subject string. There is now a new
1518        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1519        negative or greater than the length of the string. In order to test this,
1520        pcretest is extended to allow the setting of negative starting offsets.
1522    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1523        starting offset points to the beginning of a UTF-8 character was
1524        unnecessarily clumsy. I tidied it up.
1526    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1527        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1529    17. Nobody had reported that the --include_dir option, which was added in
1530        release 7.7 should have been called --include-dir (hyphen, not underscore)
1531        for compatibility with GNU grep. I have changed it to --include-dir, but
1532        left --include_dir as an undocumented synonym, and the same for
1533        --exclude-dir, though that is not available in GNU grep, at least as of
1534        release 2.5.4.
1536    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1537        characters from a string of bytes) have been redefined so as not to use
1538        loops, in order to improve performance in some environments. At the same
1539        time, I abstracted some of the common code into auxiliary macros to save
1540        repetition (this should not affect the compiled code).
1542    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1543        compile-time error is now given if \c is not followed by an ASCII
1544        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1545        different, and any byte value is allowed.)
1547    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1548        START_OPTIMIZE option, which is now allowed at compile time - but just
1549        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1550        to pcregrep and other applications that have no direct access to PCRE
1551        options. The new /Y option in pcretest sets this option when calling
1552        pcre_compile().
1554    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1555        back references. Groups containing recursive back references were forced to
1556        be atomic by that change, but in the case of named groups, the amount of
1557        memory required was incorrectly computed, leading to "Failed: internal
1558        error: code overflow". This has been fixed.
1560    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1561        pcretest.c, to avoid build problems in some Borland environments.
1564    Version 8.10 25-Jun-2010
1565    ------------------------
1567    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1568        THEN.
1570    2.  (*ACCEPT) was not working when inside an atomic group.
1572    3.  Inside a character class, \B is treated as a literal by default, but
1573        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1574        causes the error). The code is unchanged, but I tidied the documentation.
1576    4.  Inside a character class, PCRE always treated \R and \X as literals,
1577        whereas Perl faults them if its -w option is set. I have changed PCRE so
1578        that it faults them when PCRE_EXTRA is set.
1580    5.  Added support for \N, which always matches any character other than
1581        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1583    6.  When compiling pcregrep with newer versions of gcc which may have
1584        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1585        declared with attribute warn_unused_result" were given. Just casting the
1586        result to (void) does not stop the warnings; a more elaborate fudge is
1587        needed. I've used a macro to implement this.
1589    7.  Minor change to pcretest.c to avoid a compiler warning.
1591    8.  Added four artifical Unicode properties to help with an option to make
1592        \s etc use properties (see next item). The new properties are: Xan
1593        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1595    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1596        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1597        this option. Modified pcretest to add /W to test this facility. Added
1598        REG_UCP to make it available via the POSIX interface.
1600    10. Added --line-buffered to pcregrep.
1602    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1603        studied, and the match started with a letter with a code point greater than
1604        127 whose first byte was different to the first byte of the other case of
1605        the letter, the other case of this starting letter was not recognized
1606        (#976).
1608    12. If a pattern that was studied started with a repeated Unicode property
1609        test, for example, \p{Nd}+, there was the theoretical possibility of
1610        setting up an incorrect bitmap of starting bytes, but fortunately it could
1611        not have actually happened in practice until change 8 above was made (it
1612        added property types that matched character-matching opcodes).
1614    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1615        possible starting bytes for non-anchored patterns.
1617    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1618        \R, and also a number of cases that involve Unicode properties, both
1619        explicit and implicit when PCRE_UCP is set.
1621    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1622        input, it could crash or give wrong results if characters with values
1623        greater than 0xc0 were present in the subject string. (Detail: it assumed
1624        UTF-8 input when processing these items.)
1626    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1627        size_t is 64-bit (#991).
1629    17. Added a check for running out of memory when PCRE is compiled with
1630        --disable-stack-for-recursion (#990).
1632    18. If the last data line in a file for pcretest does not have a newline on
1633        the end, a newline was missing in the output.
1635    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1636        less than 128) in its various bitmaps. However, there is a facility for
1637        generating tables according to the current locale when PCRE is compiled. It
1638        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1639        characters, are recognized by isspace() and therefore were getting set in
1640        these tables, and indeed these tables seem to approximate to ISO 8859. This
1641        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1642        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1643        which of course cannot start UTF-8 characters. I have changed the code so
1644        that only real ASCII characters (less than 128) and the correct starting
1645        bytes for UTF-8 encodings are set for characters greater than 127 when in
1646        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1647        altogether.)
1649    20. Added the /T option to pcretest so as to be able to run tests with non-
1650        standard character tables, thus making it possible to include the tests
1651        used for 19 above in the standard set of tests.
1653    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1654        reference to a subpattern the other side of a comment that contains an
1655        opening parenthesis caused either an internal compiling error, or a
1656        reference to the wrong subpattern.
1659    Version 8.02 19-Mar-2010
1660    ------------------------
1662    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1664    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1665        configured.
1667    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1668        original author of that file, following a query about its status.
1670    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1671        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1673    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1674        quantifier applied to a forward-referencing subroutine call, could compile
1675        incorrect code or give the error "internal error: previously-checked
1676        referenced subpattern not found".
1678    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1679        variables to point to external functions. For these systems, therefore,
1680        pcre_malloc etc. are now initialized to local functions that call the
1681        relevant global functions.
1683    7.  There were two entries missing in the vectors called coptable and poptable
1684        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1685        I've fixed the data, and added a kludgy way of testing at compile time that
1686        the lengths are correct (equal to the number of opcodes).
1688    8.  Following on from 7, I added a similar kludge to check the length of the
1689        eint vector in pcreposix.c.
1691    9.  Error texts for pcre_compile() are held as one long string to avoid too
1692        much relocation at load time. To find a text, the string is searched,
1693        counting zeros. There was no check for running off the end of the string,
1694        which could happen if a new error number was added without updating the
1695        string.
1697    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1699    11. \K was not working if it appeared in an atomic group or in a group that
1700        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1701        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1702        the assertion is positive, but not if it is negative.
1704    12. Change 11 fortuitously reduced the size of the stack frame used in the
1705        "match()" function of pcre_exec.c by one pointer. Forthcoming
1706        implementation of support for (*MARK) will need an extra pointer on the
1707        stack; I have reserved it now, so that the stack frame size does not
1708        decrease.
1710    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1711        item in branch that calls a recursion is a subroutine call - as in the
1712        second branch in the above example - was incorrectly given the compile-
1713        time error "recursive call could loop indefinitely" because pcre_compile()
1714        was not correctly checking the subroutine for matching a non-empty string.
1716    14. The checks for overrunning compiling workspace could trigger after an
1717        overrun had occurred. This is a "should never occur" error, but it can be
1718        triggered by pathological patterns such as hundreds of nested parentheses.
1719        The checks now trigger 100 bytes before the end of the workspace.
1721    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1724    Version 8.01 19-Jan-2010
1725    ------------------------
1727    1.  If a pattern contained a conditional subpattern with only one branch (in
1728        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1729        computed the wrong minimum data length (which is of course zero for such
1730        subpatterns). This could cause incorrect "no match" results.
1732    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1733        the pattern is reset in the first branch, pcre_compile() failed with
1734        "internal error: code overflow at offset...". This happened only when
1735        the reset was to the original external option setting. (An optimization
1736        abstracts leading options settings into an external setting, which was the
1737        cause of this.)
1739    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1740        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1741        assertion pattern did not match (meaning that the assertion was true), it
1742        was incorrectly treated as false if the SKIP had been reached during the
1743        matching. This also applied to assertions used as conditions.
1745    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1746        assertion subpattern, including such a pattern used as a condition,
1747        unpredictable results occurred, instead of the error return
1750    5.  The C++ GlobalReplace function was not working like Perl for the special
1751        situation when an empty string is matched. It now does the fancy magic
1752        stuff that is necessary.
1754    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1755        removed. (These were left over from very, very early versions of PCRE.)
1757    7.  Some cosmetic changes to the code to make life easier when compiling it
1758        as part of something else:
1760        (a) Change DEBUG to PCRE_DEBUG.
1762        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1763            called "current" as "current_branch", to prevent a collision with the
1764            Linux macro when compiled as a kernel module.
1766        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1767            prevent a collision with the Linux macro when compiled as a kernel
1768            module.
1770    8.  In pcre_compile() there are some checks for integer overflows that used to
1771        cast potentially large values to (double). This has been changed to that
1772        when building, a check for int64_t is made, and if it is found, it is used
1773        instead, thus avoiding the use of floating point arithmetic. (There is no
1774        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1775        double.
1777    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1778        2005 (difference between two addresses compared to an unsigned value).
1780    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1781        custom one, because of the following reported problem in Windows:
1783          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1784              under Win32.
1785          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1786              therefore missing the function definition.
1787          - The compiler thus generates a "C" signature for the test function.
1788          - The linker fails to find the "C" function.
1789          - PCRE fails to configure if asked to do so against libbz2.
1791    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1792        messages were output:
1794          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1795          rerunning libtoolize, to keep the correct libtool macros in-tree.
1796          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1798        I have done both of these things.
1800    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1801        most of the time, it *can* run out if it is given a pattern that contains a
1802        runaway infinite recursion. I updated the discussion in the pcrestack man
1803        page.
1805    13. Now that we have gone to the x.xx style of version numbers, the minor
1806        version may start with zero. Using 08 or 09 is a bad idea because users
1807        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1808        interpreted as invalid octal numbers. I've updated the previous comment in
1809        configure.ac, and also added a check that gives an error if 08 or 09 are
1810        used.
1812    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1813        causing partial matching to fail when the end of the subject matched \W
1814        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1816    15. There were some discrepancies between the declarations in pcre_internal.h
1817        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1818        their definitions. The declarations used "const uschar *" and the
1819        definitions used USPTR. Even though USPTR is normally defined as "const
1820        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1821        reported that: "This difference in casting confuses some C++ compilers, for
1822        example, SunCC recognizes above declarations as different functions and
1823        generates broken code for hbpcre." I have changed the declarations to use
1824        USPTR.
1826    16. GNU libtool is named differently on some systems. The autogen.sh script now
1827        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1828        (FreeBSD).
1830    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1831        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1832        comment: "Figure out how to create a longlong from a string: strtoll and
1833        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1834        instance, but it only takes 2 args instead of 3!"
1836    18. A subtle bug concerned with back references has been fixed by a change of
1837        specification, with a corresponding code fix. A pattern such as
1838        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1839        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1840        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1841        same bug. Such groups have to be quantified to be useful, or contained
1842        inside another quantified group. (If there's no repetition, the reference
1843        can never match.) The problem arises because, having left the group and
1844        moved on to the rest of the pattern, a later failure that backtracks into
1845        the group uses the captured value from the final iteration of the group
1846        rather than the correct earlier one. I have fixed this in PCRE by forcing
1847        any group that contains a reference to itself to be an atomic group; that
1848        is, there cannot be any backtracking into it once it has completed. This is
1849        similar to recursive and subroutine calls.
1852    Version 8.00 19-Oct-09
1853    ----------------------
1855    1.  The table for translating pcre_compile() error codes into POSIX error codes
1856        was out-of-date, and there was no check on the pcre_compile() error code
1857        being within the table. This could lead to an OK return being given in
1858        error.
1860    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1861        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1862        in a Windows environment.
1864    3.  The pcregrep --count option prints the count for each file even when it is
1865        zero, as does GNU grep. However, pcregrep was also printing all files when
1866        --files-with-matches was added. Now, when both options are given, it prints
1867        counts only for those files that have at least one match. (GNU grep just
1868        prints the file name in this circumstance, but including the count seems
1869        more useful - otherwise, why use --count?) Also ensured that the
1870        combination -clh just lists non-zero counts, with no names.
1872    4.  The long form of the pcregrep -F option was incorrectly implemented as
1873        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1874        but it seems right to fix it, and I didn't think it was worth preserving
1875        the old behaviour.
1877    5.  The command line items --regex=pattern and --regexp=pattern were not
1878        recognized by pcregrep, which required --regex pattern or --regexp pattern
1879        (with a space rather than an '='). The man page documented the '=' forms,
1880        which are compatible with GNU grep; these now work.
1882    6.  No libpcreposix.pc file was created for pkg-config; there was just
1883        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1885    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1886        when UCP support is not needed, by modifying the Python script that
1887        generates it from Unicode data files. This should not matter if the module
1888        is correctly used as a library, but I received one complaint about 50K of
1889        unwanted data. My guess is that the person linked everything into his
1890        program rather than using a library. Anyway, it does no harm.
1892    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1893        was a minimum greater than 1 for a wide character in a possessive
1894        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1895        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1896        character. Chaos in the form of incorrect output or a compiling loop could
1897        result.
1899    9.  The restrictions on what a pattern can contain when partial matching is
1900        requested for pcre_exec() have been removed. All patterns can now be
1901        partially matched by this function. In addition, if there are at least two
1902        slots in the offset vector, the offset of the earliest inspected character
1903        for the match and the offset of the end of the subject are set in them when
1904        PCRE_ERROR_PARTIAL is returned.
1906    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1907        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1908        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1909        and may be more useful for multi-segment matching.
1911    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1912        used to be given if ever the end of the subject was reached; now it is
1913        given only if matching could not proceed because another character was
1914        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1915        string "Z", which now yields "no match" instead of "partial match". In the
1916        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1917        final character ended with (*FAIL).
1919    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1920        if the pattern had a "must contain" character that was already found in the
1921        earlier partial match, unless partial matching was again requested. For
1922        example, with the pattern /dog.(body)?/, the "must contain" character is
1923        "g". If the first part-match was for the string "dog", restarting with
1924        "sbody" failed. This bug has been fixed.
1926    13. The string returned by pcre_dfa_exec() after a partial match has been
1927        changed so that it starts at the first inspected character rather than the
1928        first character of the match. This makes a difference only if the pattern
1929        starts with a lookbehind assertion or \b or \B (\K is not supported by
1930        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1931        matching functions compatible, and I think it's the right thing to do.
1933    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1934        so that the demonstration program is easily available in environments where
1935        PCRE has not been installed from source.
1937    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1938        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1939        library.
1941    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1942        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1943        is not the first non-POSIX option to be added. Clearly some people find
1944        these options useful.
1946    17. If a caller to the POSIX matching function regexec() passes a non-zero
1947        value for nmatch with a NULL value for pmatch, the value of
1948        nmatch is forced to zero.
1950    18. RunGrepTest did not have a test for the availability of the -u option of
1951        the diff command, as RunTest does. It now checks in the same way as
1952        RunTest, and also checks for the -b option.
1954    19. If an odd number of negated classes containing just a single character
1955        interposed, within parentheses, between a forward reference to a named
1956        subpattern and the definition of the subpattern, compilation crashed with
1957        an internal error, complaining that it could not find the referenced
1958        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1959        [The bug was that it was starting one character too far in when skipping
1960        over the character class, thus treating the ] as data rather than
1961        terminating the class. This meant it could skip too much.]
1963    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1964        /g option in pcretest when the pattern contains \K, which makes it possible
1965        to have an empty string match not at the start, even when the pattern is
1966        anchored. Updated pcretest and pcredemo to use this option.
1968    21. If the maximum number of capturing subpatterns in a recursion was greater
1969        than the maximum at the outer level, the higher number was returned, but
1970        with unset values at the outer level. The correct (outer level) value is
1971        now given.
1973    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1974        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1975        make it do so. The string so far is captured, making this feature
1976        compatible with Perl.
1978    23. The tests have been re-organized, adding tests 11 and 12, to make it
1979        possible to check the Perl 5.10 features against Perl 5.10.
1981    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1982        pattern matches a fixed length string. PCRE did not allow this; now it
1983        does. Neither allows recursion.
1985    25. I finally figured out how to implement a request to provide the minimum
1986        length of subject string that was needed in order to match a given pattern.
1987        (It was back references and recursion that I had previously got hung up
1988        on.) This code has now been added to pcre_study(); it finds a lower bound
1989        to the length of subject needed. It is not necessarily the greatest lower
1990        bound, but using it to avoid searching strings that are too short does give
1991        some useful speed-ups. The value is available to calling programs via
1992        pcre_fullinfo().
1994    26. While implementing 25, I discovered to my embarrassment that pcretest had
1995        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1996        study optimizations had never been tested with that matching function.
1997        Oops. What is worse, even when it was passed study data, there was a bug in
1998        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1999        were also very few tests of studied patterns with pcre_dfa_exec().
2001    27. If (?| is used to create subpatterns with duplicate numbers, they are now
2002        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
2003        on the other side of the coin, they are no longer allowed to have different
2004        names, because these cannot be distinguished in PCRE, and this has caused
2005        confusion. (This is a difference from Perl.)
2007    28. When duplicate subpattern names are present (necessarily with different
2008        numbers, as required by 27 above), and a test is made by name in a
2009        conditional pattern, either for a subpattern having been matched, or for
2010        recursion in such a pattern, all the associated numbered subpatterns are
2011        tested, and the overall condition is true if the condition is true for any
2012        one of them. This is the way Perl works, and is also more like the way
2013        testing by number works.
2016    Version 7.9 11-Apr-09
2017    ---------------------
2019    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
2020        (pcretest), all targets were linked against these libraries. This included
2021        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
2022        libraries. This caused unwanted dependencies to be created. This problem
2023        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
2024        pcretest is linked with readline.
2026    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
2027        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
2028        moved outside it again, because FALSE and TRUE are already defined in AIX,
2029        but BOOL is not.
2031    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
2032        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
2034    4.  The pcregrep documentation said spaces were inserted as well as colons (or
2035        hyphens) following file names and line numbers when outputting matching
2036        lines. This is not true; no spaces are inserted. I have also clarified the
2037        wording for the --colour (or --color) option.
2039    5.  In pcregrep, when --colour was used with -o, the list of matching strings
2040        was not coloured; this is different to GNU grep, so I have changed it to be
2041        the same.
2043    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
2044        each matching line was coloured. Now it goes on to look for further matches
2045        of any of the test patterns, which is the same behaviour as GNU grep.
2047    7.  A pattern that could match an empty string could cause pcregrep to loop; it
2048        doesn't make sense to accept an empty string match in pcregrep, so I have
2049        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
2050        seems to be how GNU grep behaves. [But see later change 40 for release
2051        8.33.]
2053    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
2054        start or after a newline", because the conditional assertion was not being
2055        correctly handled. The rule now is that both the assertion and what follows
2056        in the first alternative must satisfy the test.
2058    9.  If auto-callout was enabled in a pattern with a conditional group whose
2059        condition was an assertion, PCRE could crash during matching, both with
2060        pcre_exec() and pcre_dfa_exec().
2062    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
2063        used for matching.
2065    11. Unicode property support in character classes was not working for
2066        characters (bytes) greater than 127 when not in UTF-8 mode.
2068    12. Added the -M command line option to pcretest.
2070    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
2072    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
2074    16. Added comments and documentation about mis-use of no_arg in the C++
2075        wrapper.
2077    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
2078        from Martin Jerabek that uses macro names for all relevant character and
2079        string constants.
2081    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
2082        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
2083        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
2084        these, but not everybody uses configure.
2086    19. A conditional group that had only one branch was not being correctly
2087        recognized as an item that could match an empty string. This meant that an
2088        enclosing group might also not be so recognized, causing infinite looping
2089        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
2090        with the subject "ab", where knowledge that the repeated group can match
2091        nothing is needed in order to break the loop.
2093    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
2094        exec(), but without supplying a callout function, matching went wrong.
2096    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
2097        leak if the size of the offset vector was greater than 30. When the vector
2098        is smaller, the saved offsets during recursion go onto a local stack
2099        vector, but for larger vectors malloc() is used. It was failing to free
2100        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
2101        error, in fact).
2103    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
2104        heapframe that is used only when UTF-8 support is enabled. This caused no
2105        problem, but was untidy.
2107    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
2108        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
2109        included within another project.
2111    24. Steven Van Ingelgem's patches to add more options to the CMake support,
2112        slightly modified by me:
2114          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
2115              not building pcregrep.
2117          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
2118              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
2120    25. Forward references, both numeric and by name, in patterns that made use of
2121        duplicate group numbers, could behave incorrectly or give incorrect errors,
2122        because when scanning forward to find the reference group, PCRE was not
2123        taking into account the duplicate group numbers. A pattern such as
2124        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
2126    26. Changed a few more instances of "const unsigned char *" to USPTR, making
2127        the feature of a custom pointer more persuasive (as requested by a user).
2129    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
2130        pcretest.c, inside #ifndefs, because it seems they are sometimes already
2131        pre-defined.
2133    28. Added support for (*UTF8) at the start of a pattern.
2135    29. Arrange for flags added by the "release type" setting in CMake to be shown
2136        in the configuration summary.
2139    Version 7.8 05-Sep-08
2140    ---------------------
2142    1.  Replaced UCP searching code with optimized version as implemented for Ad
2143        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
2144        stage table and inline lookup instead of a function, giving speed ups of 2
2145        to 5 times on some simple patterns that I tested. Permission was given to
2146        distribute the MultiStage2.py script that generates the tables (it's not in
2147        the tarball, but is in the Subversion repository).
2149    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
2150        scripts.
2152    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
2153        a group with a zero qualifier. The result of the study could be incorrect,
2154        or the function might crash, depending on the pattern.
2156    4.  Caseless matching was not working for non-ASCII characters in back
2157        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
2158        It now works when Unicode Property Support is available.
2160    5.  In pcretest, an escape such as \x{de} in the data was always generating
2161        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
2162        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
2163        truncation.
2165    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
2167    7.  Added two (int) casts to pcregrep when printing the difference of two
2168        pointers, in case they are 64-bit values.
2170    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
2171        test 2 if it fails.
2173    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
2174        and a #define of that name to empty if it is not externally set. This is to
2175        allow users of MSVC to set it if necessary.
2177    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
2178        the convenience functions in the pcre_get.c source file.
2180    11. An option change at the start of a pattern that had top-level alternatives
2181        could cause overwriting and/or a crash. This command provoked a crash in
2182        some environments:
2184          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
2186        This potential security problem was recorded as CVE-2008-2371.
2188    12. For a pattern where the match had to start at the beginning or immediately
2189        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
2190        pcre_dfa_exec() could read past the end of the passed subject if there was
2191        no match. To help with detecting such bugs (e.g. with valgrind), I modified
2192        pcretest so that it places the subject at the end of its malloc-ed buffer.
2194    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
2195        exec() might read past the end of the data buffer in UTF-8 mode.
2197    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
2198        the data contained the byte 0x85 as part of a UTF-8 character within its
2199        first line. This applied both to normal and DFA matching.
2201    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
2202        /^[^d]*?$/8 failed to match "abc".
2204    16. Added a missing copyright notice to pcrecpp_internal.h.
2206    17. Make it more clear in the documentation that values returned from
2207        pcre_exec() in ovector are byte offsets, not character counts.
2209    18. Tidied a few places to stop certain compilers from issuing warnings.
2211    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
2212        supplied by Stefan Weber. I made a further small update for 7.8 because
2213        there is a change of source arrangements: the pcre_searchfuncs.c module is
2214        replaced by pcre_ucd.c.
2217    Version 7.7 07-May-08
2218    ---------------------
2220    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
2221        a string to a long long, pretend we don't even have a long long." This is
2222        done by checking for the strtoq, strtoll, and _strtoi64 functions.
2224    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
2225        pre-7.6 versions, which defined a global no_arg variable instead of putting
2226        it in the RE class. (See also #8 below.)
2228    3.  Remove a line of dead code, identified by coverity and reported by Nuno
2229        Lopes.
2231    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
2233        (1) The include/exclude patterns were being applied to the whole pathnames
2234            of files, instead of just to the final components.
2236        (2) If there was more than one level of directory, the subdirectories were
2237            skipped unless they satisfied the include/exclude conditions. This is
2238            inconsistent with GNU grep (and could even be seen as contrary to the
2239            pcregrep specification - which I improved to make it absolutely clear).
2240            The action now is always to scan all levels of directory, and just
2241            apply the include/exclude patterns to regular files.
2243    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
2244        --exclude_dir in the tests to avoid scanning .svn directories.
2246    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
2247        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
2248        doesn't support NULs in patterns.
2250    7.  Added some missing "const"s to declarations of static tables in
2251        pcre_compile.c and pcre_dfa_exec.c.
2253    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
2254        caused by fix #2  above. (Subsequently also a second patch to fix the
2255        first patch. And a third patch - this was a messy problem.)
2257    9.  Applied Craig's patch to remove the use of push_back().
2259    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
2260        matching function regexec().
2262    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
2263        which, however, unlike Perl's \g{...}, are subroutine calls, not back
2264        references. PCRE supports relative numbers with this syntax (I don't think
2265        Oniguruma does).
2267    12. Previously, a group with a zero repeat such as (...){0} was completely
2268        omitted from the compiled regex. However, this means that if the group
2269        was called as a subroutine from elsewhere in the pattern, things went wrong
2270        (an internal error was given). Such groups are now left in the compiled
2271        pattern, with a new opcode that causes them to be skipped at execution
2272        time.
2274    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
2275        to the way PCRE behaves:
2277        (a) A lone ] character is dis-allowed (Perl treats it as data).
2279        (b) A back reference to an unmatched subpattern matches an empty string
2280            (Perl fails the current match path).
2282        (c) A data ] in a character class must be notated as \] because if the
2283            first data character in a class is ], it defines an empty class. (In
2284            Perl it is not possible to have an empty class.) The empty class []
2285            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
2286            The negative empty class [^] matches any one character, independently
2287            of the DOTALL setting.
2289    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
2290        non-existent subpattern following a character class starting with ']' and
2291        containing () gave an internal compiling error instead of "reference to
2292        non-existent subpattern". Fortunately, when the pattern did exist, the
2293        compiled code was correct. (When scanning forwards to check for the
2294        existence of the subpattern, it was treating the data ']' as terminating
2295        the class, so got the count wrong. When actually compiling, the reference
2296        was subsequently set up correctly.)
2298    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
2299        it was being rejected as not supported by pcre_dfa_exec(), even though
2300        other assertions are supported. I have made pcre_dfa_exec() support
2301        (*FAIL).
2303    16. The implementation of 13c above involved the invention of a new opcode,
2304        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
2305        cannot be changed at match time, I realized I could make a small
2306        improvement to matching performance by compiling OP_ALLANY instead of
2307        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
2308        on the OP_ANY path.
2310    17. Compiling pcretest on Windows with readline support failed without the
2311        following two fixes: (1) Make the unistd.h include conditional on
2312        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
2314    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
2315        ncurses library to be included for pcretest when ReadLine support is
2316        requested, but also to allow for it to be overridden. This patch came from
2317        Daniel Bergström.
2319    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
2320        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
2321        any errors with the current Unicode tables. Thanks to Peter Kankowski for
2322        spotting this.
2325    Version 7.6 28-Jan-08
2326    ---------------------
2328    1.  A character class containing a very large number of characters with
2329        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
2330        overflow.
2332    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
2333        HAVE_LONG_LONG is not defined.
2335    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
2336        bring them up to date and include new features. This patch includes:
2338        - Fixed PH's badly added libz and libbz2 support.
2339        - Fixed a problem with static linking.
2340        - Added pcredemo. [But later removed - see 7 below.]
2341        - Fixed dftables problem and added an option.
2342        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
2343            HAVE_LONG_LONG.
2344        - Added readline support for pcretest.
2345        - Added an listing of the option settings after cmake has run.
2347    4.  A user submitted a patch to Makefile that makes it easy to create
2348        "pcre.dll" under mingw when using Configure/Make. I added stuff to
2349        Makefile.am that cause it to include this special target, without
2350        affecting anything else. Note that the same mingw target plus all
2351        the other distribution libraries and programs are now supported
2352        when configuring with CMake (see 6 below) instead of with
2353        Configure/Make.
2355    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
2356        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
2357        exported in the Windows port". It has not yet been confirmed that the patch
2358        solves the problem, but it does no harm.
2360    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
2361        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
2362        with CMake, and also correct the comment about stack recursion.
2364    7.  Remove the automatic building of pcredemo from the ./configure system and
2365        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
2366        of a program that users should build themselves after PCRE is installed, so
2367        building it automatically is not really right. What is more, it gave
2368        trouble in some build environments.
2370    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
2373    Version 7.5 10-Jan-08
2374    ---------------------
2376    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
2377        values in parens when parsing an RE using the C++ wrapper."
2379    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
2380        Characters greater than 255 were excluded from the class instead of being
2381        included.
2383    3.  The same bug as (2) above applied to negated POSIX classes such as
2384        [:^space:].
2386    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
2387        defined or documented. It seems to have been a typo for PCRE_STATIC, so
2388        I have changed it.
2390    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
2391        first named subpattern) and a construct such as (?&a) would reference the
2392        first named subpattern whose name started with "a" (in other words, the
2393        length check was missing). Both these problems are fixed. "Subpattern name
2394        expected" is now given for (?&) (a zero-length name), and this patch also
2395        makes it give the same error for \k'' (previously it complained that that
2396        was a reference to a non-existent subpattern).
2398    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
2399        this is right because (?- can be followed by option settings as well as by
2400        digits. I have, however, made the messages clearer.
2402    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
2403        than the number used in the conditional) now cause a compile-time error.
2404        This is actually not compatible with Perl, which accepts such patterns, but
2405        treats the conditional as always being FALSE (as PCRE used to), but it
2406        seems to me that giving a diagnostic is better.
2408    8.  Change "alphameric" to the more common word "alphanumeric" in comments
2409        and messages.
2411    9.  Fix two occurrences of "backslash" in comments that should have been
2412        "backspace".
2414    10. Remove two redundant lines of code that can never be obeyed (their function
2415        was moved elsewhere).
2417    11. The program that makes PCRE's Unicode character property table had a bug
2418        which caused it to generate incorrect table entries for sequences of
2419        characters that have the same character type, but are in different scripts.
2420        It amalgamated them into a single range, with the script of the first of
2421        them. In other words, some characters were in the wrong script. There were
2422        thirteen such cases, affecting characters in the following ranges:
2424          U+002b0 - U+002c1
2425          U+0060c - U+0060d
2426          U+0061e - U+00612
2427          U+0064b - U+0065e
2428          U+0074d - U+0076d
2429          U+01800 - U+01805
2430          U+01d00 - U+01d77
2431          U+01d9b - U+01dbf
2432          U+0200b - U+0200f
2433          U+030fc - U+030fe
2434          U+03260 - U+0327f
2435          U+0fb46 - U+0fbb1
2436          U+10450 - U+1049d
2438    12. The -o option (show only the matching part of a line) for pcregrep was not
2439        compatible with GNU grep in that, if there was more than one match in a
2440        line, it showed only the first of them. It now behaves in the same way as
2441        GNU grep.
2443    13. If the -o and -v options were combined for pcregrep, it printed a blank
2444        line for every non-matching line. GNU grep prints nothing, and pcregrep now
2445        does the same. The return code can be used to tell if there were any
2446        non-matching lines.
2448    14. Added --file-offsets and --line-offsets to pcregrep.
2450    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
2451        infinitely looping recursion. The bug was that positive lookaheads were not
2452        being skipped when checking for a possible empty match (negative lookaheads
2453        and both kinds of lookbehind were skipped).
2455    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
2456        inclusion of <windows.h> to before rather than after the definition of
2457        INVALID_FILE_ATTRIBUTES (patch from David Byron).
2459    17. Specifying a possessive quantifier with a specific limit for a Unicode
2460        character property caused pcre_compile() to compile bad code, which led at
2461        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
2462        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
2463        caused the error; without that there was no problem.
2465    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
2467    19. Added --enable-pcretest-libreadline.
2469    20. In pcrecpp.cc, the variable 'count' was incremented twice in
2470        RE::GlobalReplace(). As a result, the number of replacements returned was
2471        double what it should be. I removed one of the increments, but Craig sent a
2472        later patch that removed the other one (the right fix) and added unit tests
2473        that check the return values (which was not done before).
2475    21. Several CMake things:
2477        (1) Arranged that, when cmake is used on Unix, the libraries end up with
2478            the names libpcre and libpcreposix, not just pcre and pcreposix.
2480        (2) The above change means that pcretest and pcregrep are now correctly
2481            linked with the newly-built libraries, not previously installed ones.
2485    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
2486        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
2487        UTF-8 newline character). The key issue is that the pattern starts .*;
2488        this means that the match must be either at the beginning, or after a
2489        newline. The bug was in the code for advancing after a failed match and
2490        checking that the new position followed a newline. It was not taking
2491        account of UTF-8 characters correctly.
2493    23. PCRE was behaving differently from Perl in the way it recognized POSIX
2494        character classes. PCRE was not treating the sequence [:...:] as a
2495        character class unless the ... were all letters. Perl, however, seems to
2496        allow any characters between [: and :], though of course it rejects as
2497        unknown any "names" that contain non-letters, because all the known class
2498        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
2499        for example, whereas PCRE did not - it did not recognize a POSIX character
2500        class. This seemed a bit dangerous, so the code has been changed to be
2501        closer to Perl. The behaviour is not identical to Perl, because PCRE will
2502        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
2503        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
2504        Perl does, and where it didn't before.
2506    24. Rewrite so as to remove the single use of %n from pcregrep because in some
2507        Windows environments %n is disabled by default.
2510    Version 7.4 21-Sep-07
2511    ---------------------
2513    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
2514        means that a class such as [\s] counted as "explicit reference to CR or
2515        LF". That isn't really right - the whole point of the change was to try to
2516        help when there was an actual mention of one of the two characters. So now
2517        the change happens only if \r or \n (or a literal CR or LF) character is
2518        encountered.
2520    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
2521        of both had grown to the point where there were only 3 bits left.
2522        Fortunately, there was spare space in the data structure, and so I have
2523        moved the internal flags into a new 16-bit field to free up more option
2524        bits.
2526    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
2527        but did not set the internal JCHANGED flag - either of these is enough to
2528        control the way the "get" function works - but the PCRE_INFO_JCHANGED
2529        facility is supposed to tell if (?J) was ever used, so now (?J) at the
2530        start sets both bits.
2532    4.  Added options (at build time, compile time, exec time) to change \R from
2533        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
2535    5.  doc/pcresyntax.html was missing from the distribution.
2537    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
2538        compatibility, even though it is no longer used.
2540    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
2541        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
2542        windows.h file is present (where different names are used). [This was
2543        reversed later after testing - see 16 below.]
2545    8.  Changed all #include <config.h> to #include "config.h". There were also
2546        some further <pcre.h> cases that I changed to "pcre.h".
2548    9.  When pcregrep was used with the --colour option, it missed the line ending
2549        sequence off the lines that it output.
2551    10. It was pointed out to me that arrays of string pointers cause lots of
2552        relocations when a shared library is dynamically loaded. A technique of
2553        using a single long string with a table of offsets can drastically reduce
2554        these. I have refactored PCRE in four places to do this. The result is
2555        dramatic:
2557          Originally:                          290
2558          After changing UCP table:            187
2559          After changing error message table:   43
2560          After changing table of "verbs"       36
2561          After changing table of Posix names   22
2563        Thanks to the folks working on Gregex for glib for this insight.
2565    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
2566        unicode-properties was also set.
2568    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
2570    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
2571        checked only for CRLF.
2573    14. Added casts to pcretest.c to avoid compiler warnings.
2575    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
2577    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
2578        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2579        entirely. This removes changes made in 7 above.
2581    17. The CMake files have been updated, and there is now more information about
2582        building with CMake in the NON-UNIX-USE document.
2585    Version 7.3 28-Aug-07
2586  ---------------------  ---------------------
2588   1. In the rejigging of the build system that eventually resulted in 7.1, the   1. In the rejigging of the build system that eventually resulted in 7.1, the
# Line 98  Version 7.3 09-Aug-07 Line 2679  Version 7.3 09-Aug-07
2679      the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the      the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
2680      full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still      full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
2681      does: it's just the validity check that is more restrictive.      does: it's just the validity check that is more restrictive.
2683  16. Inserted checks for integer overflows during escape sequence (backslash)  16. Inserted checks for integer overflows during escape sequence (backslash)
2684      processing, and also fixed erroneous offset values for syntax errors during      processing, and also fixed erroneous offset values for syntax errors during
2685      backslash processing.      backslash processing.
2687  17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)  17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
2688      for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".      for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
2690    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
2691        caused an overrun.
2693    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
2694        something other than just ASCII characters) inside a group that had an
2695        unlimited repeat caused a loop at compile time (while checking to see
2696        whether the group could match an empty string).
2698    20. Debugging a pattern containing \p or \P could cause a crash. For example,
2699        [\P{Any}] did so. (Error in the code for printing property names.)
2701    21. An orphan \E inside a character class could cause a crash.
2703    22. A repeated capturing bracket such as (A)? could cause a wild memory
2704        reference during compilation.
2706    23. There are several functions in pcre_compile() that scan along a compiled
2707        expression for various reasons (e.g. to see if it's fixed length for look
2708        behind). There were bugs in these functions when a repeated \p or \P was
2709        present in the pattern. These operators have additional parameters compared
2710        with \d, etc, and these were not being taken into account when moving along
2711        the compiled data. Specifically:
2713        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2714            length.
2716        (b) An item such as \pL+ within a repeated group could cause crashes or
2717            loops.
2719        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2720            "reference to non-existent subpattern" error.
2722        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2724    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2725        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2727    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2729    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2730        character were causing crashes (broken optimization).
2732    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2733        \p or \P) caused a compile-time loop.
2735    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2736        break. For example, the unstudied pattern [\r\n]A does not match the string
2737        "\r\nA" because change 7.0/46 below moves the current point on by two
2738        characters after failing to match at the start. However, the pattern \nA
2739        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2740        the same is true. There doesn't seem any very clean way out of this, but
2741        what I have chosen to do makes the common cases work: PCRE now takes note
2742        of whether there can be an explicit match for \r or \n anywhere in the
2743        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2744        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2745        pattern has explicit CR or LF references.
2747    29. Added (*CR) etc for changing newline setting at start of pattern.
2750  Version 7.2 19-Jun-07  Version 7.2 19-Jun-07

Removed from v.214  
changed lines
  Added in v.1480

  ViewVC Help
Powered by ViewVC 1.1.5